Noisy Oracle + Constraints

In [ ]:

Copied!





import os
import sys
from google.colab import drive
drive.mount('/content/drive')
sys.path.append('/content/drive/MyDrive/Projects/2023-HJ-Prox/src/')

from hj_prox import *
from functions import *
import numpy as np
import torch
import torch.nn as nn
import matplotlib.pyplot as plt
import time

torch.set_default_dtype(torch.float64)
torch.manual_seed(0)
device = 'cuda:0'
import os
import sys
from google.colab import drive
drive.mount('/content/drive')
sys.path.append('/content/drive/MyDrive/Projects/2023-HJ-Prox/src/')

from hj_prox import *
from functions import *
import numpy as np
import torch
import torch.nn as nn
import matplotlib.pyplot as plt
import time

torch.set_default_dtype(torch.float64)
torch.manual_seed(0)
device = 'cuda:0'

Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount("/content/drive", force_remount=True).

In [ ]:

Copied!

def l1_norm_noisy(x):
  # assumes shape of x is (n_samples x dim)
  return (1 + 5e-3*torch.randn(x.shape[0], device=x.device))*torch.norm(x, dim=1, p=1)
def l1_norm_noisy(x):
  # assumes shape of x is (n_samples x dim)
  return (1 + 5e-3*torch.randn(x.shape[0], device=x.device))*torch.norm(x, dim=1, p=1)

LADMM Algorithm to obtain true solution of¶

\begin{equation} \min_{x\in\mathbb{R}^{1000}} \; \|Wx\|_1 \quad \text{such that} \quad Ax=b \end{equation}

given by

\begin{align} p^{k+1} & = \eta_\lambda \left(p^k + \lambda(\nu_1^k + \alpha(Wx^k - p^k)) \right) \\ \nu_1^{k+1} &= \nu_1^k + \alpha(Wx^k - p^{k+1}) \\ \nu_2^{k+1} &= \nu_2^k + \alpha(Ax^k -b) \\ r^{k} & = W^\top \left( 2\nu_1^{k+1} - \nu_1^k \right) + A^\top \left( 2\nu_2^{k+1} - \nu_2^k \right)\\ x^{k+1} & = x^k - \beta r^k, \end{align}

where $\alpha$, $\beta$, and $\lambda$ are chosen parameters

In [ ]:

Copied!





torch.manual_seed(1)
dim = int(1e3)
A = torch.randn(int(dim/2), dim, device=device)
W = 1e-1*torch.randn(dim, dim, device=device)
b = torch.randn(int(dim/2), 1, device=device)

alpha = 1e-3
beta = 1e-1
lambd = 1e-2 

def ladmm(x0, W, A, b, max_iters, tol=1e-6, alpha=1e-2, beta=1e-1, lambd=1e-2, verbose=True):

  assert len(x0.shape)==2 and x0.shape[1]==1
  dim1 = A.shape[1]
  dim2 = A.shape[0]
  # note if A is underdetermined, dim1 < dim2 

  device = x0.device
  xk = x0
  pk = torch.zeros(dim1, 1, device=device)
  nu1k = torch.zeros(dim1, 1, device=device)
  nu2k = torch.zeros(dim2, 1, device=device)

  fk_hist = torch.zeros(max_iters)
  fixed_point_residual_hist = torch.zeros(max_iters)

  for i in range(max_iters):
    Wxk = W.matmul(xk)
    temp = pk + lambd*(nu1k + alpha*(Wxk - pk))
    pk = l1_norm_prox(temp, t=lambd) 
    nu1k_plus1 = nu1k + alpha*(Wxk - pk)
    nu2k_plus1 = nu2k + alpha*(A.matmul(xk) - b)
    rk = W.t().matmul(2*nu1k_plus1 - nu1k) + A.t().matmul(2*nu2k_plus1 - nu2k)
    xk = xk - beta*rk

    fk = l1_norm(W.matmul(xk).permute(1,0)).cpu()
    fk_hist[i] = fk
    fixed_point_residual = torch.norm(beta*rk).cpu()
    fixed_point_residual_hist[i] = fixed_point_residual

    if verbose:
      print('iter: ', i, ', fk: ', fk, ', fixed pt res: ', fixed_point_residual)

    if fixed_point_residual <= tol:
      x_sol = xk
      fk_hist = fk_hist[0:i]
      fixed_point_residual_hist = fixed_point_residual_hist[0:i]
      print('LADMM converged to fixed pt. res. of', fixed_point_residual, ', after ', i, ' iterations ')
      break

  return xk, fk_hist, fixed_point_residual_hist

def hj_mm(x0, W, A, b, max_iters, delta=1e-1, int_samples=int(1e2), step_size=1e-1, tol=1e-8, lambd=0.5, verbose=True, x_true=[], print_freq=1, alpha=1.0):

  assert len(x0.shape)==2 and x0.shape[1]==1
  dim1 = A.shape[1]
  dim2 = A.shape[0]

  assert b.shape[1]==1 and b.shape[0]==dim2
  # note if A is underdetermined, dim1 < dim2 

  device = x0.device
  xk = x0
  uk = torch.zeros(dim2, 1, device=device)

  fk_hist = torch.zeros(max_iters)
  fixed_point_residual_hist = torch.zeros(max_iters)
  rel_err_hist = torch.zeros(max_iters)
  linesearch_iters_hist = torch.zeros(max_iters)


  def f_noisy(x):
    return l1_norm_noisy(W.matmul(x.permute(1,0)).permute(1,0))

  res = A.matmul(xk)-b

  assert res.shape[0]==dim2 and res.shape[1]==1


  for i in range(max_iters):

    start_time = time.time()

    xold = xk.clone()

    res = A.matmul(xk)-b
    xk = xk - step_size*A.t().matmul(uk + lambd*res)
    xk, linesearch_iters, temp_envelope = compute_prox(xk, step_size, f_noisy, delta=delta, int_samples=int_samples, alpha=alpha, device=device)

    res = A.matmul(xk)-b
    uk = uk + lambd*(res)

    assert res.shape[0]==A.shape[0] and res.shape[1]==1
    assert uk.shape==res.shape

    Wxk = W.matmul(xk)

    fk = l1_norm(Wxk.permute(1,0)).cpu()
    fk_hist[i] = fk
    fixed_point_residual = torch.norm(xk - xold).cpu()
    fixed_point_residual_hist[i] = fixed_point_residual.cpu()
    rel_err = (torch.norm(xk - x_true)/torch.norm(x_true)).cpu()
    rel_err_hist[i] = rel_err.cpu()
    linesearch_iters_hist[i] = linesearch_iters

    end_time = time.time()
    iter_time = end_time - start_time

    if verbose:
      if (i+1)%print_freq==0:
        print('iter: ', i+1, ', fixed pt res: ', fixed_point_residual, ', ls: ', linesearch_iters,  ', rel_err: ', rel_err, ', time: ', iter_time)

  return xk, fk_hist, fixed_point_residual_hist, rel_err_hist, linesearch_iters_hist

def grad_descent(x0, A, b, max_iters, tol=1e-6, step_size=1e-1, verbose=True, x_true=[]):

  xk = x0
  grad_norm_hist = torch.zeros(max_iters)
  rel_err_hist = torch.zeros(max_iters)

  for i in range(max_iters):
    start_time = time.time()
    grad_term = A.t().matmul(A.matmul(xk)-b)
    xk =  xk - step_size * grad_term

    grad_norm_hist[i] = torch.norm(grad_term).cpu()
    rel_err = (torch.norm(xk - x_true)/torch.norm(x_true)).cpu()
    rel_err_hist[i] = rel_err
    
    end_time = time.time()
    iter_time = end_time - start_time

    if verbose:
      print('iter = ', i+1, ', grad_norm = ', grad_norm_hist[i], ', rel_err = ', rel_err)

  return xk, grad_norm_hist, rel_err_hist
torch.manual_seed(1)
dim = int(1e3)
A = torch.randn(int(dim/2), dim, device=device)
W = 1e-1*torch.randn(dim, dim, device=device)
b = torch.randn(int(dim/2), 1, device=device)

alpha = 1e-3
beta = 1e-1
lambd = 1e-2 

def ladmm(x0, W, A, b, max_iters, tol=1e-6, alpha=1e-2, beta=1e-1, lambd=1e-2, verbose=True):

  assert len(x0.shape)==2 and x0.shape[1]==1
  dim1 = A.shape[1]
  dim2 = A.shape[0]
  # note if A is underdetermined, dim1 < dim2 

  device = x0.device
  xk = x0
  pk = torch.zeros(dim1, 1, device=device)
  nu1k = torch.zeros(dim1, 1, device=device)
  nu2k = torch.zeros(dim2, 1, device=device)

  fk_hist = torch.zeros(max_iters)
  fixed_point_residual_hist = torch.zeros(max_iters)

  for i in range(max_iters):
    Wxk = W.matmul(xk)
    temp = pk + lambd*(nu1k + alpha*(Wxk - pk))
    pk = l1_norm_prox(temp, t=lambd) 
    nu1k_plus1 = nu1k + alpha*(Wxk - pk)
    nu2k_plus1 = nu2k + alpha*(A.matmul(xk) - b)
    rk = W.t().matmul(2*nu1k_plus1 - nu1k) + A.t().matmul(2*nu2k_plus1 - nu2k)
    xk = xk - beta*rk

    fk = l1_norm(W.matmul(xk).permute(1,0)).cpu()
    fk_hist[i] = fk
    fixed_point_residual = torch.norm(beta*rk).cpu()
    fixed_point_residual_hist[i] = fixed_point_residual

    if verbose:
      print('iter: ', i, ', fk: ', fk, ', fixed pt res: ', fixed_point_residual)

    if fixed_point_residual <= tol:
      x_sol = xk
      fk_hist = fk_hist[0:i]
      fixed_point_residual_hist = fixed_point_residual_hist[0:i]
      print('LADMM converged to fixed pt. res. of', fixed_point_residual, ', after ', i, ' iterations ')
      break

  return xk, fk_hist, fixed_point_residual_hist

def hj_mm(x0, W, A, b, max_iters, delta=1e-1, int_samples=int(1e2), step_size=1e-1, tol=1e-8, lambd=0.5, verbose=True, x_true=[], print_freq=1, alpha=1.0):

  assert len(x0.shape)==2 and x0.shape[1]==1
  dim1 = A.shape[1]
  dim2 = A.shape[0]

  assert b.shape[1]==1 and b.shape[0]==dim2
  # note if A is underdetermined, dim1 < dim2 

  device = x0.device
  xk = x0
  uk = torch.zeros(dim2, 1, device=device)

  fk_hist = torch.zeros(max_iters)
  fixed_point_residual_hist = torch.zeros(max_iters)
  rel_err_hist = torch.zeros(max_iters)
  linesearch_iters_hist = torch.zeros(max_iters)


  def f_noisy(x):
    return l1_norm_noisy(W.matmul(x.permute(1,0)).permute(1,0))

  res = A.matmul(xk)-b

  assert res.shape[0]==dim2 and res.shape[1]==1


  for i in range(max_iters):

    start_time = time.time()

    xold = xk.clone()

    res = A.matmul(xk)-b
    xk = xk - step_size*A.t().matmul(uk + lambd*res)
    xk, linesearch_iters, temp_envelope = compute_prox(xk, step_size, f_noisy, delta=delta, int_samples=int_samples, alpha=alpha, device=device)

    res = A.matmul(xk)-b
    uk = uk + lambd*(res)

    assert res.shape[0]==A.shape[0] and res.shape[1]==1
    assert uk.shape==res.shape

    Wxk = W.matmul(xk)

    fk = l1_norm(Wxk.permute(1,0)).cpu()
    fk_hist[i] = fk
    fixed_point_residual = torch.norm(xk - xold).cpu()
    fixed_point_residual_hist[i] = fixed_point_residual.cpu()
    rel_err = (torch.norm(xk - x_true)/torch.norm(x_true)).cpu()
    rel_err_hist[i] = rel_err.cpu()
    linesearch_iters_hist[i] = linesearch_iters

    end_time = time.time()
    iter_time = end_time - start_time

    if verbose:
      if (i+1)%print_freq==0:
        print('iter: ', i+1, ', fixed pt res: ', fixed_point_residual, ', ls: ', linesearch_iters,  ', rel_err: ', rel_err, ', time: ', iter_time)

  return xk, fk_hist, fixed_point_residual_hist, rel_err_hist, linesearch_iters_hist

def grad_descent(x0, A, b, max_iters, tol=1e-6, step_size=1e-1, verbose=True, x_true=[]):

  xk = x0
  grad_norm_hist = torch.zeros(max_iters)
  rel_err_hist = torch.zeros(max_iters)

  for i in range(max_iters):
    start_time = time.time()
    grad_term = A.t().matmul(A.matmul(xk)-b)
    xk =  xk - step_size * grad_term

    grad_norm_hist[i] = torch.norm(grad_term).cpu()
    rel_err = (torch.norm(xk - x_true)/torch.norm(x_true)).cpu()
    rel_err_hist[i] = rel_err
    
    end_time = time.time()
    iter_time = end_time - start_time

    if verbose:
      print('iter = ', i+1, ', grad_norm = ', grad_norm_hist[i], ', rel_err = ', rel_err)

  return xk, grad_norm_hist, rel_err_hist

Find Numerical True Solution with Linearized ADMM.¶

Numerically true solution is obtained by averaging three runs

In [ ]:

Copied!





n_trials_true = 3
x_true_array = torch.randn(n_trials_true, dim, device=device)
for i in range(n_trials_true):
  x0 = torch.randn(dim, 1, device=device)

  start_time_ladmm = time.time()
  x_true, fk_admm_hist, fixed_point_residual_admm_hist = ladmm(x0, W, A, b, int(5e4), alpha=alpha, lambd=lambd, beta=beta, verbose=False, tol=1e-6)
  end_time_ladmm = time.time()
  time_ladmm = end_time_ladmm - start_time_ladmm
  print('trial ', i, ', time = ', time_ladmm)
  x_true_array[i, :] = x_true.view(-1)
  assert(max(x_true)<np.inf)
n_trials_true = 3
x_true_array = torch.randn(n_trials_true, dim, device=device)
for i in range(n_trials_true):
  x0 = torch.randn(dim, 1, device=device)

  start_time_ladmm = time.time()
  x_true, fk_admm_hist, fixed_point_residual_admm_hist = ladmm(x0, W, A, b, int(5e4), alpha=alpha, lambd=lambd, beta=beta, verbose=False, tol=1e-6)
  end_time_ladmm = time.time()
  time_ladmm = end_time_ladmm - start_time_ladmm
  print('trial ', i, ', time = ', time_ladmm)
  x_true_array[i, :] = x_true.view(-1)
  assert(max(x_true)

LADMM converged to fixed pt. res. of tensor(9.9999e-07) , after  33947  iterations 
trial  0 , time =  17.05860137939453
LADMM converged to fixed pt. res. of tensor(1.0000e-06) , after  34690  iterations 
trial  1 , time =  17.195754289627075
LADMM converged to fixed pt. res. of tensor(9.9986e-07) , after  31409  iterations 
trial  2 , time =  15.552043437957764

Compute Gradient Descent Solution¶

This is to ensure that the proximal operator is not simply the identity

In [ ]:

Copied!





# get relative errors for ladmm_no_prox.
max_iters = int(1e4)
U, s, Vt = torch.svd(A.t().matmul(A))
step_size = 1/s.max().cpu() - 1e-8
lambd=0.5

xopt_GD, grad_norm_hist_GD, rel_err_hist_GD = grad_descent(x0, A, b, max_iters, step_size=step_size, verbose=False, x_true=x_true)
# get relative errors for ladmm_no_prox.
max_iters = int(1e4)
U, s, Vt = torch.svd(A.t().matmul(A))
step_size = 1/s.max().cpu() - 1e-8
lambd=0.5

xopt_GD, grad_norm_hist_GD, rel_err_hist_GD = grad_descent(x0, A, b, max_iters, step_size=step_size, verbose=False, x_true=x_true)

Blackbox Oracle HJ Prox Experiment 1: Fix number of samples and vary $\delta$¶

In [ ]:

Copied!





# Average runs over 30 trials: standalone
n_trials = 1
delta_array = [1e-2, 1e-1, 1.0, 10, 100]
alpha_array = [2**(-12), 2**(-9), (2**(-5) + 2**(-6))/2, 2**(-2), 2**(-1)]

xopt_HJ_lmm_delta_array = torch.zeros(len(delta_array), x_true.shape[0], x_true.shape[1])
fk_HJ_lmm_hist_delta_array = torch.zeros(len(delta_array), max_iters)
norm_diff_HJ_lmm_hist_delta_array = torch.zeros(len(delta_array), max_iters)
rel_err_lmm_hist_delta_array = torch.zeros(len(delta_array), max_iters)
ls_iters_lmm_hist_delta_array = torch.zeros(len(delta_array), max_iters)

samples_for_varying_delta = int(1e4)

for delta_index in range(len(delta_array)):
  print('delta =', delta_array[delta_index])
  for i in range(n_trials):
    start_time_hj_lmm = time.time()
    current_delta = delta_array[delta_index]
    xopt_HJ_lmm_temp, fk_HJ_lmm_hist_temp, norm_diff_HJ_lmm_hist_temp, rel_err_hist_temp, ls_iters_hist_temp = hj_mm(x0, W, A, b, max_iters, 
                                                                                                  int_samples=samples_for_varying_delta, 
                                                                                                  delta=current_delta,
                                                                                                  lambd = lambd, 
                                                                                                  step_size=step_size,
                                                                                                  verbose=True, 
                                                                                                  print_freq=50,
                                                                                                  x_true=x_true,
                                                                                                  alpha=alpha_array[delta_index])
    end_time_hj_lmm = time.time()
    time_hj_lmm = end_time_hj_lmm - start_time_hj_lmm
    print('trial ', i+1, ' finished after', time_hj_lmm)
      

    print('\n\n\n\n CHANGE, delta = ', current_delta)

    if i==0:
      xopt_HJ_lmm = xopt_HJ_lmm_temp
      fk_HJ_lmm_hist = fk_HJ_lmm_hist_temp
      norm_diff_HJ_lmm_hist = norm_diff_HJ_lmm_hist_temp
      rel_err_hist = rel_err_hist_temp
      ls_iters_hist = ls_iters_hist_temp
    else:
      xopt_HJ_lmm = xopt_HJ_lmm + xopt_HJ_lmm_temp
      fk_HJ_lmm_hist = fk_HJ_lmm_hist + fk_HJ_lmm_hist_temp
      norm_diff_HJ_lmm_hist = norm_diff_HJ_lmm_hist + norm_diff_HJ_lmm_hist_temp
      rel_err_hist = rel_err_hist + rel_err_hist_temp
      ls_iters_hist = ls_iters_hist + ls_iters_hist_temp


  xopt_HJ_lmm_delta_array[delta_index, :, :] = xopt_HJ_lmm/n_trials
  fk_HJ_lmm_hist_delta_array[delta_index, :] = fk_HJ_lmm_hist/n_trials
  norm_diff_HJ_lmm_hist_delta_array[delta_index, :] = norm_diff_HJ_lmm_hist/n_trials
  rel_err_lmm_hist_delta_array[delta_index, :] = rel_err_hist/n_trials 
  ls_iters_lmm_hist_delta_array[delta_index,:] = ls_iters_hist/n_trials
  print('norm(rel_err_hist_temp - rel_err_lmm_hist_delta_array[sample_index,:]) = ', torch.norm(rel_err_hist_temp - rel_err_lmm_hist_delta_array[delta_index,:]))
# Average runs over 30 trials: standalone
n_trials = 1
delta_array = [1e-2, 1e-1, 1.0, 10, 100]
alpha_array = [2**(-12), 2**(-9), (2**(-5) + 2**(-6))/2, 2**(-2), 2**(-1)]

xopt_HJ_lmm_delta_array = torch.zeros(len(delta_array), x_true.shape[0], x_true.shape[1])
fk_HJ_lmm_hist_delta_array = torch.zeros(len(delta_array), max_iters)
norm_diff_HJ_lmm_hist_delta_array = torch.zeros(len(delta_array), max_iters)
rel_err_lmm_hist_delta_array = torch.zeros(len(delta_array), max_iters)
ls_iters_lmm_hist_delta_array = torch.zeros(len(delta_array), max_iters)

samples_for_varying_delta = int(1e4)

for delta_index in range(len(delta_array)):
  print('delta =', delta_array[delta_index])
  for i in range(n_trials):
    start_time_hj_lmm = time.time()
    current_delta = delta_array[delta_index]
    xopt_HJ_lmm_temp, fk_HJ_lmm_hist_temp, norm_diff_HJ_lmm_hist_temp, rel_err_hist_temp, ls_iters_hist_temp = hj_mm(x0, W, A, b, max_iters, 
                                                                                                  int_samples=samples_for_varying_delta, 
                                                                                                  delta=current_delta,
                                                                                                  lambd = lambd, 
                                                                                                  step_size=step_size,
                                                                                                  verbose=True, 
                                                                                                  print_freq=50,
                                                                                                  x_true=x_true,
                                                                                                  alpha=alpha_array[delta_index])
    end_time_hj_lmm = time.time()
    time_hj_lmm = end_time_hj_lmm - start_time_hj_lmm
    print('trial ', i+1, ' finished after', time_hj_lmm)
      

    print('\n\n\n\n CHANGE, delta = ', current_delta)

    if i==0:
      xopt_HJ_lmm = xopt_HJ_lmm_temp
      fk_HJ_lmm_hist = fk_HJ_lmm_hist_temp
      norm_diff_HJ_lmm_hist = norm_diff_HJ_lmm_hist_temp
      rel_err_hist = rel_err_hist_temp
      ls_iters_hist = ls_iters_hist_temp
    else:
      xopt_HJ_lmm = xopt_HJ_lmm + xopt_HJ_lmm_temp
      fk_HJ_lmm_hist = fk_HJ_lmm_hist + fk_HJ_lmm_hist_temp
      norm_diff_HJ_lmm_hist = norm_diff_HJ_lmm_hist + norm_diff_HJ_lmm_hist_temp
      rel_err_hist = rel_err_hist + rel_err_hist_temp
      ls_iters_hist = ls_iters_hist + ls_iters_hist_temp


  xopt_HJ_lmm_delta_array[delta_index, :, :] = xopt_HJ_lmm/n_trials
  fk_HJ_lmm_hist_delta_array[delta_index, :] = fk_HJ_lmm_hist/n_trials
  norm_diff_HJ_lmm_hist_delta_array[delta_index, :] = norm_diff_HJ_lmm_hist/n_trials
  rel_err_lmm_hist_delta_array[delta_index, :] = rel_err_hist/n_trials 
  ls_iters_lmm_hist_delta_array[delta_index,:] = ls_iters_hist/n_trials
  print('norm(rel_err_hist_temp - rel_err_lmm_hist_delta_array[sample_index,:]) = ', torch.norm(rel_err_hist_temp - rel_err_lmm_hist_delta_array[delta_index,:]))

delta = 0.01
iter:  50 , fixed pt res:  tensor(0.6014) , ls:  1 , rel_err:  tensor(15.2876) , time:  0.002866983413696289
iter:  100 , fixed pt res:  tensor(0.2610) , ls:  1 , rel_err:  tensor(14.1650) , time:  0.0026578903198242188
iter:  150 , fixed pt res:  tensor(0.1266) , ls:  1 , rel_err:  tensor(13.2045) , time:  0.002607583999633789
iter:  200 , fixed pt res:  tensor(0.0741) , ls:  1 , rel_err:  tensor(12.3301) , time:  0.0027289390563964844
iter:  250 , fixed pt res:  tensor(0.0612) , ls:  1 , rel_err:  tensor(11.5111) , time:  0.0026788711547851562
iter:  300 , fixed pt res:  tensor(0.0564) , ls:  1 , rel_err:  tensor(10.7342) , time:  0.0026938915252685547
iter:  350 , fixed pt res:  tensor(0.0528) , ls:  1 , rel_err:  tensor(10.0056) , time:  0.0026712417602539062
iter:  400 , fixed pt res:  tensor(0.0527) , ls:  1 , rel_err:  tensor(9.3370) , time:  0.0026133060455322266
iter:  450 , fixed pt res:  tensor(0.0507) , ls:  1 , rel_err:  tensor(8.7010) , time:  0.0026159286499023438
iter:  500 , fixed pt res:  tensor(0.0490) , ls:  1 , rel_err:  tensor(8.1159) , time:  0.002603769302368164
iter:  550 , fixed pt res:  tensor(0.0506) , ls:  1 , rel_err:  tensor(7.5500) , time:  0.0025870800018310547
iter:  600 , fixed pt res:  tensor(0.0502) , ls:  1 , rel_err:  tensor(7.0455) , time:  0.002651691436767578
iter:  650 , fixed pt res:  tensor(0.0490) , ls:  1 , rel_err:  tensor(6.5675) , time:  0.0026242733001708984
iter:  700 , fixed pt res:  tensor(0.0472) , ls:  1 , rel_err:  tensor(6.1281) , time:  0.0026123523712158203
iter:  750 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(5.7053) , time:  0.0027201175689697266
iter:  800 , fixed pt res:  tensor(0.0473) , ls:  1 , rel_err:  tensor(5.3180) , time:  0.002632617950439453
iter:  850 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(4.9593) , time:  0.002621889114379883
iter:  900 , fixed pt res:  tensor(0.0463) , ls:  1 , rel_err:  tensor(4.6238) , time:  0.002607107162475586
iter:  950 , fixed pt res:  tensor(0.0459) , ls:  1 , rel_err:  tensor(4.3144) , time:  0.002595186233520508
iter:  1000 , fixed pt res:  tensor(0.0452) , ls:  1 , rel_err:  tensor(4.0185) , time:  0.002645254135131836
iter:  1050 , fixed pt res:  tensor(0.0463) , ls:  1 , rel_err:  tensor(3.7540) , time:  0.002599954605102539
iter:  1100 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(3.5047) , time:  0.0026836395263671875
iter:  1150 , fixed pt res:  tensor(0.0456) , ls:  1 , rel_err:  tensor(3.2836) , time:  0.0026116371154785156
iter:  1200 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(3.0738) , time:  0.002595663070678711
iter:  1250 , fixed pt res:  tensor(0.0455) , ls:  1 , rel_err:  tensor(2.8702) , time:  0.002620220184326172
iter:  1300 , fixed pt res:  tensor(0.0455) , ls:  1 , rel_err:  tensor(2.6800) , time:  0.0026030540466308594
iter:  1350 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(2.5230) , time:  0.002622842788696289
iter:  1400 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(2.3789) , time:  0.0026531219482421875
iter:  1450 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(2.2404) , time:  0.002665281295776367
iter:  1500 , fixed pt res:  tensor(0.0439) , ls:  1 , rel_err:  tensor(2.1036) , time:  0.002603769302368164
iter:  1550 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(1.9756) , time:  0.0026116371154785156
iter:  1600 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(1.8512) , time:  0.0026519298553466797
iter:  1650 , fixed pt res:  tensor(0.0447) , ls:  1 , rel_err:  tensor(1.7429) , time:  0.0026175975799560547
iter:  1700 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(1.6449) , time:  0.0026214122772216797
iter:  1750 , fixed pt res:  tensor(0.0447) , ls:  1 , rel_err:  tensor(1.5489) , time:  0.002623319625854492
iter:  1800 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(1.4627) , time:  0.002610445022583008
iter:  1850 , fixed pt res:  tensor(0.0426) , ls:  1 , rel_err:  tensor(1.3820) , time:  0.0026204586029052734
iter:  1900 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(1.3105) , time:  0.002740621566772461
iter:  1950 , fixed pt res:  tensor(0.0461) , ls:  1 , rel_err:  tensor(1.2419) , time:  0.0026121139526367188
iter:  2000 , fixed pt res:  tensor(0.0445) , ls:  1 , rel_err:  tensor(1.1821) , time:  0.0026407241821289062
iter:  2050 , fixed pt res:  tensor(0.0466) , ls:  1 , rel_err:  tensor(1.1262) , time:  0.0026285648345947266
iter:  2100 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(1.0700) , time:  0.0027396678924560547
iter:  2150 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(1.0336) , time:  0.0026540756225585938
iter:  2200 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(0.9851) , time:  0.0028276443481445312
iter:  2250 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.9466) , time:  0.0026481151580810547
iter:  2300 , fixed pt res:  tensor(0.0469) , ls:  1 , rel_err:  tensor(0.8937) , time:  0.002695798873901367
iter:  2350 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.8420) , time:  0.00261688232421875
iter:  2400 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(0.8041) , time:  0.0026023387908935547
iter:  2450 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(0.7707) , time:  0.0026216506958007812
iter:  2500 , fixed pt res:  tensor(0.0431) , ls:  1 , rel_err:  tensor(0.7425) , time:  0.0026090145111083984
iter:  2550 , fixed pt res:  tensor(0.0427) , ls:  1 , rel_err:  tensor(0.7211) , time:  0.0026030540466308594
iter:  2600 , fixed pt res:  tensor(0.0445) , ls:  1 , rel_err:  tensor(0.6999) , time:  0.0026161670684814453
iter:  2650 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.6637) , time:  0.0026285648345947266
iter:  2700 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(0.6484) , time:  0.002592325210571289
iter:  2750 , fixed pt res:  tensor(0.0462) , ls:  1 , rel_err:  tensor(0.6204) , time:  0.0026221275329589844
iter:  2800 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.5907) , time:  0.0026400089263916016
iter:  2850 , fixed pt res:  tensor(0.0461) , ls:  1 , rel_err:  tensor(0.5667) , time:  0.002603769302368164
iter:  2900 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(0.5479) , time:  0.002847433090209961
iter:  2950 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(0.5221) , time:  0.002637147903442383
iter:  3000 , fixed pt res:  tensor(0.0442) , ls:  1 , rel_err:  tensor(0.4944) , time:  0.002875804901123047
iter:  3050 , fixed pt res:  tensor(0.0435) , ls:  1 , rel_err:  tensor(0.4865) , time:  0.0026242733001708984
iter:  3100 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.4677) , time:  0.002634286880493164
iter:  3150 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.4603) , time:  0.0026273727416992188
iter:  3200 , fixed pt res:  tensor(0.0449) , ls:  1 , rel_err:  tensor(0.4451) , time:  0.0027008056640625
iter:  3250 , fixed pt res:  tensor(0.0436) , ls:  1 , rel_err:  tensor(0.4363) , time:  0.002660512924194336
iter:  3300 , fixed pt res:  tensor(0.0434) , ls:  1 , rel_err:  tensor(0.4198) , time:  0.002737283706665039
iter:  3350 , fixed pt res:  tensor(0.0456) , ls:  1 , rel_err:  tensor(0.4052) , time:  0.002620697021484375
iter:  3400 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(0.3929) , time:  0.002596616744995117
iter:  3450 , fixed pt res:  tensor(0.0468) , ls:  1 , rel_err:  tensor(0.3751) , time:  0.002627134323120117
iter:  3500 , fixed pt res:  tensor(0.0465) , ls:  1 , rel_err:  tensor(0.3594) , time:  0.0026710033416748047
iter:  3550 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.3492) , time:  0.0026192665100097656
iter:  3600 , fixed pt res:  tensor(0.0462) , ls:  1 , rel_err:  tensor(0.3329) , time:  0.002612590789794922
iter:  3650 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(0.3213) , time:  0.0026526451110839844
iter:  3700 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.3138) , time:  0.002615213394165039
iter:  3750 , fixed pt res:  tensor(0.0445) , ls:  1 , rel_err:  tensor(0.3068) , time:  0.0026254653930664062
iter:  3800 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(0.2985) , time:  0.0026793479919433594
iter:  3850 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.2954) , time:  0.0026276111602783203
iter:  3900 , fixed pt res:  tensor(0.0452) , ls:  1 , rel_err:  tensor(0.2995) , time:  0.0027604103088378906
iter:  3950 , fixed pt res:  tensor(0.0455) , ls:  1 , rel_err:  tensor(0.2918) , time:  0.0026252269744873047
iter:  4000 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.2829) , time:  0.002598285675048828
iter:  4050 , fixed pt res:  tensor(0.0442) , ls:  1 , rel_err:  tensor(0.2883) , time:  0.0026128292083740234
iter:  4100 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(0.2823) , time:  0.0026764869689941406
iter:  4150 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(0.2749) , time:  0.0026464462280273438
iter:  4200 , fixed pt res:  tensor(0.0456) , ls:  1 , rel_err:  tensor(0.2738) , time:  0.002658843994140625
iter:  4250 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(0.2839) , time:  0.002664327621459961
iter:  4300 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(0.2846) , time:  0.002600431442260742
iter:  4350 , fixed pt res:  tensor(0.0447) , ls:  1 , rel_err:  tensor(0.2822) , time:  0.002614736557006836
iter:  4400 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(0.2766) , time:  0.0026111602783203125
iter:  4450 , fixed pt res:  tensor(0.0462) , ls:  1 , rel_err:  tensor(0.2851) , time:  0.002942800521850586
iter:  4500 , fixed pt res:  tensor(0.0452) , ls:  1 , rel_err:  tensor(0.2790) , time:  0.002645254135131836
iter:  4550 , fixed pt res:  tensor(0.0458) , ls:  1 , rel_err:  tensor(0.2737) , time:  0.0026237964630126953
iter:  4600 , fixed pt res:  tensor(0.0425) , ls:  1 , rel_err:  tensor(0.2737) , time:  0.0026972293853759766
iter:  4650 , fixed pt res:  tensor(0.0439) , ls:  1 , rel_err:  tensor(0.2738) , time:  0.002602815628051758
iter:  4700 , fixed pt res:  tensor(0.0459) , ls:  1 , rel_err:  tensor(0.2681) , time:  0.0027227401733398438
iter:  4750 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(0.2600) , time:  0.0026187896728515625
iter:  4800 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(0.2689) , time:  0.002664327621459961
iter:  4850 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.2647) , time:  0.002623319625854492
iter:  4900 , fixed pt res:  tensor(0.0442) , ls:  1 , rel_err:  tensor(0.2610) , time:  0.0026009082794189453
iter:  4950 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(0.2547) , time:  0.0026657581329345703
iter:  5000 , fixed pt res:  tensor(0.0422) , ls:  1 , rel_err:  tensor(0.2452) , time:  0.0026068687438964844
iter:  5050 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(0.2534) , time:  0.002641439437866211
iter:  5100 , fixed pt res:  tensor(0.0469) , ls:  1 , rel_err:  tensor(0.2546) , time:  0.0025894641876220703
iter:  5150 , fixed pt res:  tensor(0.0439) , ls:  1 , rel_err:  tensor(0.2565) , time:  0.0026578903198242188
iter:  5200 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.2602) , time:  0.002780437469482422
iter:  5250 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(0.2538) , time:  0.002653360366821289
iter:  5300 , fixed pt res:  tensor(0.0468) , ls:  1 , rel_err:  tensor(0.2541) , time:  0.0026001930236816406
iter:  5350 , fixed pt res:  tensor(0.0465) , ls:  1 , rel_err:  tensor(0.2446) , time:  0.0026025772094726562
iter:  5400 , fixed pt res:  tensor(0.0468) , ls:  1 , rel_err:  tensor(0.2299) , time:  0.002604246139526367
iter:  5450 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(0.2321) , time:  0.002683401107788086
iter:  5500 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.2332) , time:  0.0026204586029052734
iter:  5550 , fixed pt res:  tensor(0.0436) , ls:  1 , rel_err:  tensor(0.2307) , time:  0.0025882720947265625
iter:  5600 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.2264) , time:  0.0026459693908691406
iter:  5650 , fixed pt res:  tensor(0.0456) , ls:  1 , rel_err:  tensor(0.2368) , time:  0.002635955810546875
iter:  5700 , fixed pt res:  tensor(0.0426) , ls:  1 , rel_err:  tensor(0.2348) , time:  0.002643585205078125
iter:  5750 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(0.2338) , time:  0.002627849578857422
iter:  5800 , fixed pt res:  tensor(0.0466) , ls:  1 , rel_err:  tensor(0.2340) , time:  0.002646923065185547
iter:  5850 , fixed pt res:  tensor(0.0460) , ls:  1 , rel_err:  tensor(0.2335) , time:  0.002660036087036133
iter:  5900 , fixed pt res:  tensor(0.0460) , ls:  1 , rel_err:  tensor(0.2385) , time:  0.002616405487060547
iter:  5950 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(0.2460) , time:  0.002605915069580078
iter:  6000 , fixed pt res:  tensor(0.0452) , ls:  1 , rel_err:  tensor(0.2416) , time:  0.0026574134826660156
iter:  6050 , fixed pt res:  tensor(0.0456) , ls:  1 , rel_err:  tensor(0.2449) , time:  0.0026526451110839844
iter:  6100 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(0.2512) , time:  0.0025959014892578125
iter:  6150 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(0.2495) , time:  0.0025892257690429688
iter:  6200 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2530) , time:  0.002648591995239258
iter:  6250 , fixed pt res:  tensor(0.0447) , ls:  1 , rel_err:  tensor(0.2474) , time:  0.0026361942291259766
iter:  6300 , fixed pt res:  tensor(0.0468) , ls:  1 , rel_err:  tensor(0.2457) , time:  0.0026311874389648438
iter:  6350 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.2481) , time:  0.002582073211669922
iter:  6400 , fixed pt res:  tensor(0.0442) , ls:  1 , rel_err:  tensor(0.2603) , time:  0.0027589797973632812
iter:  6450 , fixed pt res:  tensor(0.0458) , ls:  1 , rel_err:  tensor(0.2571) , time:  0.002671957015991211
iter:  6500 , fixed pt res:  tensor(0.0481) , ls:  1 , rel_err:  tensor(0.2515) , time:  0.0026404857635498047
iter:  6550 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.2546) , time:  0.002638101577758789
iter:  6600 , fixed pt res:  tensor(0.0431) , ls:  1 , rel_err:  tensor(0.2515) , time:  0.002616405487060547
iter:  6650 , fixed pt res:  tensor(0.0429) , ls:  1 , rel_err:  tensor(0.2477) , time:  0.0026443004608154297
iter:  6700 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(0.2431) , time:  0.0026798248291015625
iter:  6750 , fixed pt res:  tensor(0.0460) , ls:  1 , rel_err:  tensor(0.2456) , time:  0.0026657581329345703
iter:  6800 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.2512) , time:  0.00266265869140625
iter:  6850 , fixed pt res:  tensor(0.0455) , ls:  1 , rel_err:  tensor(0.2436) , time:  0.0026111602783203125
iter:  6900 , fixed pt res:  tensor(0.0449) , ls:  1 , rel_err:  tensor(0.2404) , time:  0.0026488304138183594
iter:  6950 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(0.2493) , time:  0.0025832653045654297
iter:  7000 , fixed pt res:  tensor(0.0459) , ls:  1 , rel_err:  tensor(0.2581) , time:  0.002736330032348633
iter:  7050 , fixed pt res:  tensor(0.0439) , ls:  1 , rel_err:  tensor(0.2536) , time:  0.0026400089263916016
iter:  7100 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(0.2552) , time:  0.002611398696899414
iter:  7150 , fixed pt res:  tensor(0.0442) , ls:  1 , rel_err:  tensor(0.2662) , time:  0.0026242733001708984
iter:  7200 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(0.2723) , time:  0.002617359161376953
iter:  7250 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(0.2547) , time:  0.0026268959045410156
iter:  7300 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2492) , time:  0.0026891231536865234
iter:  7350 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(0.2514) , time:  0.0025877952575683594
iter:  7400 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.2516) , time:  0.0026044845581054688
iter:  7450 , fixed pt res:  tensor(0.0435) , ls:  1 , rel_err:  tensor(0.2598) , time:  0.0025823116302490234
iter:  7500 , fixed pt res:  tensor(0.0458) , ls:  1 , rel_err:  tensor(0.2602) , time:  0.0026979446411132812
iter:  7550 , fixed pt res:  tensor(0.0447) , ls:  1 , rel_err:  tensor(0.2565) , time:  0.002629995346069336
iter:  7600 , fixed pt res:  tensor(0.0462) , ls:  1 , rel_err:  tensor(0.2516) , time:  0.002672433853149414
iter:  7650 , fixed pt res:  tensor(0.0455) , ls:  1 , rel_err:  tensor(0.2441) , time:  0.0026144981384277344
iter:  7700 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2406) , time:  0.0026328563690185547
iter:  7750 , fixed pt res:  tensor(0.0436) , ls:  1 , rel_err:  tensor(0.2406) , time:  0.0026073455810546875
iter:  7800 , fixed pt res:  tensor(0.0442) , ls:  1 , rel_err:  tensor(0.2393) , time:  0.0026388168334960938
iter:  7850 , fixed pt res:  tensor(0.0439) , ls:  1 , rel_err:  tensor(0.2345) , time:  0.0026521682739257812
iter:  7900 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(0.2379) , time:  0.0026383399963378906
iter:  7950 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(0.2370) , time:  0.002651214599609375
iter:  8000 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.2446) , time:  0.002850055694580078
iter:  8050 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(0.2443) , time:  0.0026161670684814453
iter:  8100 , fixed pt res:  tensor(0.0461) , ls:  1 , rel_err:  tensor(0.2481) , time:  0.0026061534881591797
iter:  8150 , fixed pt res:  tensor(0.0432) , ls:  1 , rel_err:  tensor(0.2444) , time:  0.0026128292083740234
iter:  8200 , fixed pt res:  tensor(0.0464) , ls:  1 , rel_err:  tensor(0.2433) , time:  0.0027930736541748047
iter:  8250 , fixed pt res:  tensor(0.0435) , ls:  1 , rel_err:  tensor(0.2440) , time:  0.0026159286499023438
iter:  8300 , fixed pt res:  tensor(0.0442) , ls:  1 , rel_err:  tensor(0.2494) , time:  0.00262451171875
iter:  8350 , fixed pt res:  tensor(0.0468) , ls:  1 , rel_err:  tensor(0.2557) , time:  0.002601146697998047
iter:  8400 , fixed pt res:  tensor(0.0434) , ls:  1 , rel_err:  tensor(0.2539) , time:  0.002947568893432617
iter:  8450 , fixed pt res:  tensor(0.0452) , ls:  1 , rel_err:  tensor(0.2474) , time:  0.0026085376739501953
iter:  8500 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(0.2328) , time:  0.0026068687438964844
iter:  8550 , fixed pt res:  tensor(0.0461) , ls:  1 , rel_err:  tensor(0.2369) , time:  0.0025882720947265625
iter:  8600 , fixed pt res:  tensor(0.0445) , ls:  1 , rel_err:  tensor(0.2276) , time:  0.002688169479370117
iter:  8650 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(0.2380) , time:  0.0025942325592041016
iter:  8700 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(0.2387) , time:  0.0026848316192626953
iter:  8750 , fixed pt res:  tensor(0.0420) , ls:  1 , rel_err:  tensor(0.2441) , time:  0.002606630325317383
iter:  8800 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2478) , time:  0.0026895999908447266
iter:  8850 , fixed pt res:  tensor(0.0459) , ls:  1 , rel_err:  tensor(0.2411) , time:  0.0025746822357177734
iter:  8900 , fixed pt res:  tensor(0.0452) , ls:  1 , rel_err:  tensor(0.2358) , time:  0.0026254653930664062
iter:  8950 , fixed pt res:  tensor(0.0461) , ls:  1 , rel_err:  tensor(0.2361) , time:  0.0026378631591796875
iter:  9000 , fixed pt res:  tensor(0.0426) , ls:  1 , rel_err:  tensor(0.2347) , time:  0.0026183128356933594
iter:  9050 , fixed pt res:  tensor(0.0455) , ls:  1 , rel_err:  tensor(0.2319) , time:  0.002596139907836914
iter:  9100 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.2398) , time:  0.002637624740600586
iter:  9150 , fixed pt res:  tensor(0.0456) , ls:  1 , rel_err:  tensor(0.2427) , time:  0.0026619434356689453
iter:  9200 , fixed pt res:  tensor(0.0452) , ls:  1 , rel_err:  tensor(0.2579) , time:  0.002629995346069336
iter:  9250 , fixed pt res:  tensor(0.0458) , ls:  1 , rel_err:  tensor(0.2582) , time:  0.0026335716247558594
iter:  9300 , fixed pt res:  tensor(0.0459) , ls:  1 , rel_err:  tensor(0.2596) , time:  0.002610445022583008
iter:  9350 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.2467) , time:  0.0026264190673828125
iter:  9400 , fixed pt res:  tensor(0.0432) , ls:  1 , rel_err:  tensor(0.2438) , time:  0.0026292800903320312
iter:  9450 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(0.2456) , time:  0.0026323795318603516
iter:  9500 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.2347) , time:  0.002623319625854492
iter:  9550 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(0.2371) , time:  0.002624988555908203
iter:  9600 , fixed pt res:  tensor(0.0452) , ls:  1 , rel_err:  tensor(0.2468) , time:  0.0026865005493164062
iter:  9650 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.2422) , time:  0.002601146697998047
iter:  9700 , fixed pt res:  tensor(0.0442) , ls:  1 , rel_err:  tensor(0.2472) , time:  0.00258636474609375
iter:  9750 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2434) , time:  0.0026235580444335938
iter:  9800 , fixed pt res:  tensor(0.0469) , ls:  1 , rel_err:  tensor(0.2460) , time:  0.0028467178344726562
iter:  9850 , fixed pt res:  tensor(0.0466) , ls:  1 , rel_err:  tensor(0.2459) , time:  0.0026454925537109375
iter:  9900 , fixed pt res:  tensor(0.0452) , ls:  1 , rel_err:  tensor(0.2351) , time:  0.0026373863220214844
iter:  9950 , fixed pt res:  tensor(0.0432) , ls:  1 , rel_err:  tensor(0.2353) , time:  0.002624034881591797
iter:  10000 , fixed pt res:  tensor(0.0460) , ls:  1 , rel_err:  tensor(0.2458) , time:  0.002570629119873047
trial  1  finished after 27.3780620098114




 CHANGE, delta =  0.01
norm(rel_err_hist_temp - rel_err_lmm_hist_delta_array[sample_index,:]) =  tensor(0.)
delta = 0.1
iter:  50 , fixed pt res:  tensor(0.6024) , ls:  1 , rel_err:  tensor(15.2734) , time:  0.0026395320892333984
iter:  100 , fixed pt res:  tensor(0.2571) , ls:  1 , rel_err:  tensor(14.1573) , time:  0.002633810043334961
iter:  150 , fixed pt res:  tensor(0.1314) , ls:  1 , rel_err:  tensor(13.2116) , time:  0.002636432647705078
iter:  200 , fixed pt res:  tensor(0.0796) , ls:  1 , rel_err:  tensor(12.3254) , time:  0.002630472183227539
iter:  250 , fixed pt res:  tensor(0.0655) , ls:  1 , rel_err:  tensor(11.4973) , time:  0.0026044845581054688
iter:  300 , fixed pt res:  tensor(0.0626) , ls:  1 , rel_err:  tensor(10.7382) , time:  0.002668619155883789
iter:  350 , fixed pt res:  tensor(0.0591) , ls:  1 , rel_err:  tensor(10.0184) , time:  0.0026521682739257812
iter:  400 , fixed pt res:  tensor(0.0576) , ls:  1 , rel_err:  tensor(9.3451) , time:  0.0036263465881347656
iter:  450 , fixed pt res:  tensor(0.0553) , ls:  1 , rel_err:  tensor(8.7179) , time:  0.002643108367919922
iter:  500 , fixed pt res:  tensor(0.0547) , ls:  1 , rel_err:  tensor(8.1263) , time:  0.0026276111602783203
iter:  550 , fixed pt res:  tensor(0.0541) , ls:  1 , rel_err:  tensor(7.5703) , time:  0.002591371536254883
iter:  600 , fixed pt res:  tensor(0.0519) , ls:  1 , rel_err:  tensor(7.0570) , time:  0.00260162353515625
iter:  650 , fixed pt res:  tensor(0.0512) , ls:  1 , rel_err:  tensor(6.5821) , time:  0.0026454925537109375
iter:  700 , fixed pt res:  tensor(0.0519) , ls:  1 , rel_err:  tensor(6.1244) , time:  0.0026962757110595703
iter:  750 , fixed pt res:  tensor(0.0525) , ls:  1 , rel_err:  tensor(5.7183) , time:  0.0025899410247802734
iter:  800 , fixed pt res:  tensor(0.0516) , ls:  1 , rel_err:  tensor(5.3294) , time:  0.0026574134826660156
iter:  850 , fixed pt res:  tensor(0.0498) , ls:  1 , rel_err:  tensor(4.9821) , time:  0.0026454925537109375
iter:  900 , fixed pt res:  tensor(0.0508) , ls:  1 , rel_err:  tensor(4.6579) , time:  0.002607583999633789
iter:  950 , fixed pt res:  tensor(0.0522) , ls:  1 , rel_err:  tensor(4.3529) , time:  0.002629995346069336
iter:  1000 , fixed pt res:  tensor(0.0509) , ls:  1 , rel_err:  tensor(4.0844) , time:  0.0026361942291259766
iter:  1050 , fixed pt res:  tensor(0.0516) , ls:  1 , rel_err:  tensor(3.8181) , time:  0.0026645660400390625
iter:  1100 , fixed pt res:  tensor(0.0519) , ls:  1 , rel_err:  tensor(3.5839) , time:  0.002616405487060547
iter:  1150 , fixed pt res:  tensor(0.0508) , ls:  1 , rel_err:  tensor(3.3641) , time:  0.002714872360229492
iter:  1200 , fixed pt res:  tensor(0.0504) , ls:  1 , rel_err:  tensor(3.1551) , time:  0.002900838851928711
iter:  1250 , fixed pt res:  tensor(0.0512) , ls:  1 , rel_err:  tensor(2.9600) , time:  0.002619028091430664
iter:  1300 , fixed pt res:  tensor(0.0514) , ls:  1 , rel_err:  tensor(2.7889) , time:  0.002590656280517578
iter:  1350 , fixed pt res:  tensor(0.0500) , ls:  1 , rel_err:  tensor(2.6252) , time:  0.002590656280517578
iter:  1400 , fixed pt res:  tensor(0.0496) , ls:  1 , rel_err:  tensor(2.4691) , time:  0.002591371536254883
iter:  1450 , fixed pt res:  tensor(0.0520) , ls:  1 , rel_err:  tensor(2.3233) , time:  0.0026082992553710938
iter:  1500 , fixed pt res:  tensor(0.0529) , ls:  1 , rel_err:  tensor(2.1884) , time:  0.0026760101318359375
iter:  1550 , fixed pt res:  tensor(0.0515) , ls:  1 , rel_err:  tensor(2.0734) , time:  0.002660512924194336
iter:  1600 , fixed pt res:  tensor(0.0516) , ls:  1 , rel_err:  tensor(1.9564) , time:  0.0026001930236816406
iter:  1650 , fixed pt res:  tensor(0.0502) , ls:  1 , rel_err:  tensor(1.8567) , time:  0.002652883529663086
iter:  1700 , fixed pt res:  tensor(0.0491) , ls:  1 , rel_err:  tensor(1.7603) , time:  0.0026578903198242188
iter:  1750 , fixed pt res:  tensor(0.0499) , ls:  1 , rel_err:  tensor(1.6678) , time:  0.002605438232421875
iter:  1800 , fixed pt res:  tensor(0.0499) , ls:  1 , rel_err:  tensor(1.5920) , time:  0.002622365951538086
iter:  1850 , fixed pt res:  tensor(0.0497) , ls:  1 , rel_err:  tensor(1.5209) , time:  0.002629518508911133
iter:  1900 , fixed pt res:  tensor(0.0484) , ls:  1 , rel_err:  tensor(1.4413) , time:  0.0026397705078125
iter:  1950 , fixed pt res:  tensor(0.0494) , ls:  1 , rel_err:  tensor(1.3834) , time:  0.0026040077209472656
iter:  2000 , fixed pt res:  tensor(0.0497) , ls:  1 , rel_err:  tensor(1.3253) , time:  0.002625703811645508
iter:  2050 , fixed pt res:  tensor(0.0506) , ls:  1 , rel_err:  tensor(1.2545) , time:  0.002617359161376953
iter:  2100 , fixed pt res:  tensor(0.0496) , ls:  1 , rel_err:  tensor(1.2010) , time:  0.0026395320892333984
iter:  2150 , fixed pt res:  tensor(0.0528) , ls:  1 , rel_err:  tensor(1.1449) , time:  0.0026330947875976562
iter:  2200 , fixed pt res:  tensor(0.0486) , ls:  1 , rel_err:  tensor(1.0987) , time:  0.0026540756225585938
iter:  2250 , fixed pt res:  tensor(0.0503) , ls:  1 , rel_err:  tensor(1.0528) , time:  0.002614259719848633
iter:  2300 , fixed pt res:  tensor(0.0516) , ls:  1 , rel_err:  tensor(1.0073) , time:  0.0025959014892578125
iter:  2350 , fixed pt res:  tensor(0.0493) , ls:  1 , rel_err:  tensor(0.9671) , time:  0.0025959014892578125
iter:  2400 , fixed pt res:  tensor(0.0498) , ls:  1 , rel_err:  tensor(0.9244) , time:  0.002643585205078125
iter:  2450 , fixed pt res:  tensor(0.0521) , ls:  1 , rel_err:  tensor(0.8814) , time:  0.0026264190673828125
iter:  2500 , fixed pt res:  tensor(0.0491) , ls:  1 , rel_err:  tensor(0.8438) , time:  0.0028371810913085938
iter:  2550 , fixed pt res:  tensor(0.0502) , ls:  1 , rel_err:  tensor(0.8141) , time:  0.002639293670654297
iter:  2600 , fixed pt res:  tensor(0.0494) , ls:  1 , rel_err:  tensor(0.7701) , time:  0.0028192996978759766
iter:  2650 , fixed pt res:  tensor(0.0492) , ls:  1 , rel_err:  tensor(0.7341) , time:  0.0026717185974121094
iter:  2700 , fixed pt res:  tensor(0.0497) , ls:  1 , rel_err:  tensor(0.7033) , time:  0.002650737762451172
iter:  2750 , fixed pt res:  tensor(0.0494) , ls:  1 , rel_err:  tensor(0.6702) , time:  0.0025954246520996094
iter:  2800 , fixed pt res:  tensor(0.0520) , ls:  1 , rel_err:  tensor(0.6541) , time:  0.002783060073852539
iter:  2850 , fixed pt res:  tensor(0.0505) , ls:  1 , rel_err:  tensor(0.6294) , time:  0.0026559829711914062
iter:  2900 , fixed pt res:  tensor(0.0490) , ls:  1 , rel_err:  tensor(0.6014) , time:  0.0029981136322021484
iter:  2950 , fixed pt res:  tensor(0.0465) , ls:  1 , rel_err:  tensor(0.5876) , time:  0.0027670860290527344
iter:  3000 , fixed pt res:  tensor(0.0522) , ls:  1 , rel_err:  tensor(0.5688) , time:  0.0028235912322998047
iter:  3050 , fixed pt res:  tensor(0.0508) , ls:  1 , rel_err:  tensor(0.5508) , time:  0.0026345252990722656
iter:  3100 , fixed pt res:  tensor(0.0499) , ls:  1 , rel_err:  tensor(0.5394) , time:  0.0027310848236083984
iter:  3150 , fixed pt res:  tensor(0.0514) , ls:  1 , rel_err:  tensor(0.5302) , time:  0.002622365951538086
iter:  3200 , fixed pt res:  tensor(0.0507) , ls:  1 , rel_err:  tensor(0.5260) , time:  0.002646923065185547
iter:  3250 , fixed pt res:  tensor(0.0486) , ls:  1 , rel_err:  tensor(0.5135) , time:  0.0025806427001953125
iter:  3300 , fixed pt res:  tensor(0.0500) , ls:  1 , rel_err:  tensor(0.5036) , time:  0.002665281295776367
iter:  3350 , fixed pt res:  tensor(0.0510) , ls:  1 , rel_err:  tensor(0.4874) , time:  0.002624034881591797
iter:  3400 , fixed pt res:  tensor(0.0485) , ls:  1 , rel_err:  tensor(0.4782) , time:  0.0026574134826660156
iter:  3450 , fixed pt res:  tensor(0.0526) , ls:  1 , rel_err:  tensor(0.4766) , time:  0.0026869773864746094
iter:  3500 , fixed pt res:  tensor(0.0483) , ls:  1 , rel_err:  tensor(0.4638) , time:  0.0027141571044921875
iter:  3550 , fixed pt res:  tensor(0.0479) , ls:  1 , rel_err:  tensor(0.4446) , time:  0.002619028091430664
iter:  3600 , fixed pt res:  tensor(0.0510) , ls:  1 , rel_err:  tensor(0.4431) , time:  0.002580881118774414
iter:  3650 , fixed pt res:  tensor(0.0515) , ls:  1 , rel_err:  tensor(0.4307) , time:  0.002605915069580078
iter:  3700 , fixed pt res:  tensor(0.0504) , ls:  1 , rel_err:  tensor(0.4222) , time:  0.00266265869140625
iter:  3750 , fixed pt res:  tensor(0.0497) , ls:  1 , rel_err:  tensor(0.4080) , time:  0.0026161670684814453
iter:  3800 , fixed pt res:  tensor(0.0492) , ls:  1 , rel_err:  tensor(0.4069) , time:  0.0025949478149414062
iter:  3850 , fixed pt res:  tensor(0.0502) , ls:  1 , rel_err:  tensor(0.3987) , time:  0.002596139907836914
iter:  3900 , fixed pt res:  tensor(0.0499) , ls:  1 , rel_err:  tensor(0.3832) , time:  0.002691507339477539
iter:  3950 , fixed pt res:  tensor(0.0507) , ls:  1 , rel_err:  tensor(0.3856) , time:  0.002644777297973633
iter:  4000 , fixed pt res:  tensor(0.0516) , ls:  1 , rel_err:  tensor(0.3708) , time:  0.0026993751525878906
iter:  4050 , fixed pt res:  tensor(0.0493) , ls:  1 , rel_err:  tensor(0.3730) , time:  0.002619028091430664
iter:  4100 , fixed pt res:  tensor(0.0494) , ls:  1 , rel_err:  tensor(0.3689) , time:  0.0026683807373046875
iter:  4150 , fixed pt res:  tensor(0.0479) , ls:  1 , rel_err:  tensor(0.3694) , time:  0.00261688232421875
iter:  4200 , fixed pt res:  tensor(0.0516) , ls:  1 , rel_err:  tensor(0.3652) , time:  0.002650737762451172
iter:  4250 , fixed pt res:  tensor(0.0516) , ls:  1 , rel_err:  tensor(0.3573) , time:  0.0026361942291259766
iter:  4300 , fixed pt res:  tensor(0.0491) , ls:  1 , rel_err:  tensor(0.3519) , time:  0.0031049251556396484
iter:  4350 , fixed pt res:  tensor(0.0490) , ls:  1 , rel_err:  tensor(0.3418) , time:  0.0026412010192871094
iter:  4400 , fixed pt res:  tensor(0.0509) , ls:  1 , rel_err:  tensor(0.3438) , time:  0.002599477767944336
iter:  4450 , fixed pt res:  tensor(0.0533) , ls:  1 , rel_err:  tensor(0.3472) , time:  0.002591848373413086
iter:  4500 , fixed pt res:  tensor(0.0492) , ls:  1 , rel_err:  tensor(0.3378) , time:  0.002658367156982422
iter:  4550 , fixed pt res:  tensor(0.0467) , ls:  1 , rel_err:  tensor(0.3343) , time:  0.002669095993041992
iter:  4600 , fixed pt res:  tensor(0.0477) , ls:  1 , rel_err:  tensor(0.3372) , time:  0.0025866031646728516
iter:  4650 , fixed pt res:  tensor(0.0489) , ls:  1 , rel_err:  tensor(0.3435) , time:  0.002625703811645508
iter:  4700 , fixed pt res:  tensor(0.0513) , ls:  1 , rel_err:  tensor(0.3364) , time:  0.0026092529296875
iter:  4750 , fixed pt res:  tensor(0.0501) , ls:  1 , rel_err:  tensor(0.3308) , time:  0.0026960372924804688
iter:  4800 , fixed pt res:  tensor(0.0481) , ls:  1 , rel_err:  tensor(0.3289) , time:  0.0026102066040039062
iter:  4850 , fixed pt res:  tensor(0.0487) , ls:  1 , rel_err:  tensor(0.3347) , time:  0.0026645660400390625
iter:  4900 , fixed pt res:  tensor(0.0505) , ls:  1 , rel_err:  tensor(0.3225) , time:  0.002632617950439453
iter:  4950 , fixed pt res:  tensor(0.0496) , ls:  1 , rel_err:  tensor(0.3249) , time:  0.0026404857635498047
iter:  5000 , fixed pt res:  tensor(0.0520) , ls:  1 , rel_err:  tensor(0.3180) , time:  0.0026352405548095703
iter:  5050 , fixed pt res:  tensor(0.0503) , ls:  1 , rel_err:  tensor(0.3167) , time:  0.0026044845581054688
iter:  5100 , fixed pt res:  tensor(0.0513) , ls:  1 , rel_err:  tensor(0.3107) , time:  0.0026082992553710938
iter:  5150 , fixed pt res:  tensor(0.0506) , ls:  1 , rel_err:  tensor(0.3191) , time:  0.002648591995239258
iter:  5200 , fixed pt res:  tensor(0.0493) , ls:  1 , rel_err:  tensor(0.3144) , time:  0.002597808837890625
iter:  5250 , fixed pt res:  tensor(0.0502) , ls:  1 , rel_err:  tensor(0.3067) , time:  0.0026259422302246094
iter:  5300 , fixed pt res:  tensor(0.0509) , ls:  1 , rel_err:  tensor(0.3126) , time:  0.0026192665100097656
iter:  5350 , fixed pt res:  tensor(0.0503) , ls:  1 , rel_err:  tensor(0.2975) , time:  0.002664804458618164
iter:  5400 , fixed pt res:  tensor(0.0516) , ls:  1 , rel_err:  tensor(0.3082) , time:  0.0026803016662597656
iter:  5450 , fixed pt res:  tensor(0.0515) , ls:  1 , rel_err:  tensor(0.3042) , time:  0.0026454925537109375
iter:  5500 , fixed pt res:  tensor(0.0502) , ls:  1 , rel_err:  tensor(0.3087) , time:  0.0026674270629882812
iter:  5550 , fixed pt res:  tensor(0.0492) , ls:  1 , rel_err:  tensor(0.3069) , time:  0.002651214599609375
iter:  5600 , fixed pt res:  tensor(0.0507) , ls:  1 , rel_err:  tensor(0.2987) , time:  0.0025959014892578125
iter:  5650 , fixed pt res:  tensor(0.0494) , ls:  1 , rel_err:  tensor(0.2988) , time:  0.002653360366821289
iter:  5700 , fixed pt res:  tensor(0.0521) , ls:  1 , rel_err:  tensor(0.2915) , time:  0.002653360366821289
iter:  5750 , fixed pt res:  tensor(0.0492) , ls:  1 , rel_err:  tensor(0.2909) , time:  0.0026268959045410156
iter:  5800 , fixed pt res:  tensor(0.0496) , ls:  1 , rel_err:  tensor(0.2915) , time:  0.0026569366455078125
iter:  5850 , fixed pt res:  tensor(0.0481) , ls:  1 , rel_err:  tensor(0.2881) , time:  0.0026044845581054688
iter:  5900 , fixed pt res:  tensor(0.0497) , ls:  1 , rel_err:  tensor(0.2900) , time:  0.0026781558990478516
iter:  5950 , fixed pt res:  tensor(0.0500) , ls:  1 , rel_err:  tensor(0.2745) , time:  0.0025932788848876953
iter:  6000 , fixed pt res:  tensor(0.0509) , ls:  1 , rel_err:  tensor(0.2660) , time:  0.0025959014892578125
iter:  6050 , fixed pt res:  tensor(0.0507) , ls:  1 , rel_err:  tensor(0.2699) , time:  0.0026063919067382812
iter:  6100 , fixed pt res:  tensor(0.0506) , ls:  1 , rel_err:  tensor(0.2744) , time:  0.0027208328247070312
iter:  6150 , fixed pt res:  tensor(0.0481) , ls:  1 , rel_err:  tensor(0.2756) , time:  0.002628326416015625
iter:  6200 , fixed pt res:  tensor(0.0493) , ls:  1 , rel_err:  tensor(0.2902) , time:  0.002673625946044922
iter:  6250 , fixed pt res:  tensor(0.0501) , ls:  1 , rel_err:  tensor(0.2795) , time:  0.005936861038208008
iter:  6300 , fixed pt res:  tensor(0.0507) , ls:  1 , rel_err:  tensor(0.2847) , time:  0.002641439437866211
iter:  6350 , fixed pt res:  tensor(0.0514) , ls:  1 , rel_err:  tensor(0.2862) , time:  0.002637624740600586
iter:  6400 , fixed pt res:  tensor(0.0495) , ls:  1 , rel_err:  tensor(0.2855) , time:  0.0026657581329345703
iter:  6450 , fixed pt res:  tensor(0.0509) , ls:  1 , rel_err:  tensor(0.2882) , time:  0.0026798248291015625
iter:  6500 , fixed pt res:  tensor(0.0507) , ls:  1 , rel_err:  tensor(0.2736) , time:  0.0026044845581054688
iter:  6550 , fixed pt res:  tensor(0.0508) , ls:  1 , rel_err:  tensor(0.2734) , time:  0.0026023387908935547
iter:  6600 , fixed pt res:  tensor(0.0484) , ls:  1 , rel_err:  tensor(0.2813) , time:  0.0026292800903320312
iter:  6650 , fixed pt res:  tensor(0.0475) , ls:  1 , rel_err:  tensor(0.2846) , time:  0.002617359161376953
iter:  6700 , fixed pt res:  tensor(0.0501) , ls:  1 , rel_err:  tensor(0.2871) , time:  0.0026006698608398438
iter:  6750 , fixed pt res:  tensor(0.0496) , ls:  1 , rel_err:  tensor(0.2827) , time:  0.0026998519897460938
iter:  6800 , fixed pt res:  tensor(0.0491) , ls:  1 , rel_err:  tensor(0.2914) , time:  0.002641916275024414
iter:  6850 , fixed pt res:  tensor(0.0511) , ls:  1 , rel_err:  tensor(0.2767) , time:  0.0026628971099853516
iter:  6900 , fixed pt res:  tensor(0.0490) , ls:  1 , rel_err:  tensor(0.2685) , time:  0.002627849578857422
iter:  6950 , fixed pt res:  tensor(0.0485) , ls:  1 , rel_err:  tensor(0.2683) , time:  0.0026264190673828125
iter:  7000 , fixed pt res:  tensor(0.0517) , ls:  1 , rel_err:  tensor(0.2680) , time:  0.002621889114379883
iter:  7050 , fixed pt res:  tensor(0.0487) , ls:  1 , rel_err:  tensor(0.2720) , time:  0.002657175064086914
iter:  7100 , fixed pt res:  tensor(0.0504) , ls:  1 , rel_err:  tensor(0.2722) , time:  0.0027887821197509766
iter:  7150 , fixed pt res:  tensor(0.0503) , ls:  1 , rel_err:  tensor(0.2694) , time:  0.0026350021362304688
iter:  7200 , fixed pt res:  tensor(0.0490) , ls:  1 , rel_err:  tensor(0.2759) , time:  0.0026476383209228516
iter:  7250 , fixed pt res:  tensor(0.0515) , ls:  1 , rel_err:  tensor(0.2741) , time:  0.0025992393493652344
iter:  7300 , fixed pt res:  tensor(0.0479) , ls:  1 , rel_err:  tensor(0.2775) , time:  0.0026497840881347656
iter:  7350 , fixed pt res:  tensor(0.0494) , ls:  1 , rel_err:  tensor(0.2756) , time:  0.0026824474334716797
iter:  7400 , fixed pt res:  tensor(0.0517) , ls:  1 , rel_err:  tensor(0.2752) , time:  0.002943754196166992
iter:  7450 , fixed pt res:  tensor(0.0494) , ls:  1 , rel_err:  tensor(0.2795) , time:  0.0026013851165771484
iter:  7500 , fixed pt res:  tensor(0.0494) , ls:  1 , rel_err:  tensor(0.2867) , time:  0.002633810043334961
iter:  7550 , fixed pt res:  tensor(0.0489) , ls:  1 , rel_err:  tensor(0.2829) , time:  0.0026514530181884766
iter:  7600 , fixed pt res:  tensor(0.0526) , ls:  1 , rel_err:  tensor(0.2755) , time:  0.002739429473876953
iter:  7650 , fixed pt res:  tensor(0.0489) , ls:  1 , rel_err:  tensor(0.2817) , time:  0.002626657485961914
iter:  7700 , fixed pt res:  tensor(0.0518) , ls:  1 , rel_err:  tensor(0.2754) , time:  0.002665281295776367
iter:  7750 , fixed pt res:  tensor(0.0502) , ls:  1 , rel_err:  tensor(0.2794) , time:  0.0026416778564453125
iter:  7800 , fixed pt res:  tensor(0.0515) , ls:  1 , rel_err:  tensor(0.2688) , time:  0.0026836395263671875
iter:  7850 , fixed pt res:  tensor(0.0509) , ls:  1 , rel_err:  tensor(0.2722) , time:  0.0026421546936035156
iter:  7900 , fixed pt res:  tensor(0.0481) , ls:  1 , rel_err:  tensor(0.2603) , time:  0.002622842788696289
iter:  7950 , fixed pt res:  tensor(0.0513) , ls:  1 , rel_err:  tensor(0.2730) , time:  0.0026264190673828125
iter:  8000 , fixed pt res:  tensor(0.0507) , ls:  1 , rel_err:  tensor(0.2707) , time:  0.0026280879974365234
iter:  8050 , fixed pt res:  tensor(0.0531) , ls:  1 , rel_err:  tensor(0.2837) , time:  0.002634286880493164
iter:  8100 , fixed pt res:  tensor(0.0479) , ls:  1 , rel_err:  tensor(0.2819) , time:  0.002626180648803711
iter:  8150 , fixed pt res:  tensor(0.0490) , ls:  1 , rel_err:  tensor(0.2658) , time:  0.0026307106018066406
iter:  8200 , fixed pt res:  tensor(0.0498) , ls:  1 , rel_err:  tensor(0.2765) , time:  0.0026407241821289062
iter:  8250 , fixed pt res:  tensor(0.0507) , ls:  1 , rel_err:  tensor(0.2701) , time:  0.002642393112182617
iter:  8300 , fixed pt res:  tensor(0.0512) , ls:  1 , rel_err:  tensor(0.2767) , time:  0.002636432647705078
iter:  8350 , fixed pt res:  tensor(0.0525) , ls:  1 , rel_err:  tensor(0.2821) , time:  0.0026226043701171875
iter:  8400 , fixed pt res:  tensor(0.0503) , ls:  1 , rel_err:  tensor(0.2873) , time:  0.002777576446533203
iter:  8450 , fixed pt res:  tensor(0.0510) , ls:  1 , rel_err:  tensor(0.2763) , time:  0.002614259719848633
iter:  8500 , fixed pt res:  tensor(0.0502) , ls:  1 , rel_err:  tensor(0.2895) , time:  0.0026481151580810547
iter:  8550 , fixed pt res:  tensor(0.0496) , ls:  1 , rel_err:  tensor(0.2829) , time:  0.0026159286499023438
iter:  8600 , fixed pt res:  tensor(0.0485) , ls:  1 , rel_err:  tensor(0.2801) , time:  0.00262451171875
iter:  8650 , fixed pt res:  tensor(0.0495) , ls:  1 , rel_err:  tensor(0.2850) , time:  0.0025970935821533203
iter:  8700 , fixed pt res:  tensor(0.0503) , ls:  1 , rel_err:  tensor(0.2773) , time:  0.0026483535766601562
iter:  8750 , fixed pt res:  tensor(0.0501) , ls:  1 , rel_err:  tensor(0.2770) , time:  0.0026476383209228516
iter:  8800 , fixed pt res:  tensor(0.0504) , ls:  1 , rel_err:  tensor(0.2859) , time:  0.002651214599609375
iter:  8850 , fixed pt res:  tensor(0.0509) , ls:  1 , rel_err:  tensor(0.2834) , time:  0.002624034881591797
iter:  8900 , fixed pt res:  tensor(0.0502) , ls:  1 , rel_err:  tensor(0.2798) , time:  0.002610921859741211
iter:  8950 , fixed pt res:  tensor(0.0477) , ls:  1 , rel_err:  tensor(0.2901) , time:  0.002869129180908203
iter:  9000 , fixed pt res:  tensor(0.0513) , ls:  1 , rel_err:  tensor(0.2827) , time:  0.0026993751525878906
iter:  9050 , fixed pt res:  tensor(0.0496) , ls:  1 , rel_err:  tensor(0.2763) , time:  0.0026035308837890625
iter:  9100 , fixed pt res:  tensor(0.0494) , ls:  1 , rel_err:  tensor(0.2757) , time:  0.0026476383209228516
iter:  9150 , fixed pt res:  tensor(0.0501) , ls:  1 , rel_err:  tensor(0.2769) , time:  0.002643108367919922
iter:  9200 , fixed pt res:  tensor(0.0497) , ls:  1 , rel_err:  tensor(0.2807) , time:  0.0026454925537109375
iter:  9250 , fixed pt res:  tensor(0.0493) , ls:  1 , rel_err:  tensor(0.2818) , time:  0.002629518508911133
iter:  9300 , fixed pt res:  tensor(0.0502) , ls:  1 , rel_err:  tensor(0.2832) , time:  0.0027909278869628906
iter:  9350 , fixed pt res:  tensor(0.0483) , ls:  1 , rel_err:  tensor(0.2743) , time:  0.0026569366455078125
iter:  9400 , fixed pt res:  tensor(0.0506) , ls:  1 , rel_err:  tensor(0.2720) , time:  0.002588510513305664
iter:  9450 , fixed pt res:  tensor(0.0493) , ls:  1 , rel_err:  tensor(0.2639) , time:  0.0026352405548095703
iter:  9500 , fixed pt res:  tensor(0.0474) , ls:  1 , rel_err:  tensor(0.2650) , time:  0.0026216506958007812
iter:  9550 , fixed pt res:  tensor(0.0505) , ls:  1 , rel_err:  tensor(0.2635) , time:  0.002645730972290039
iter:  9600 , fixed pt res:  tensor(0.0511) , ls:  1 , rel_err:  tensor(0.2621) , time:  0.002592802047729492
iter:  9650 , fixed pt res:  tensor(0.0504) , ls:  1 , rel_err:  tensor(0.2597) , time:  0.002633333206176758
iter:  9700 , fixed pt res:  tensor(0.0482) , ls:  1 , rel_err:  tensor(0.2604) , time:  0.002636432647705078
iter:  9750 , fixed pt res:  tensor(0.0485) , ls:  1 , rel_err:  tensor(0.2763) , time:  0.0026273727416992188
iter:  9800 , fixed pt res:  tensor(0.0473) , ls:  1 , rel_err:  tensor(0.2771) , time:  0.002584695816040039
iter:  9850 , fixed pt res:  tensor(0.0492) , ls:  1 , rel_err:  tensor(0.2740) , time:  0.002599477767944336
iter:  9900 , fixed pt res:  tensor(0.0486) , ls:  1 , rel_err:  tensor(0.2774) , time:  0.002656221389770508
iter:  9950 , fixed pt res:  tensor(0.0490) , ls:  1 , rel_err:  tensor(0.2829) , time:  0.002659320831298828
iter:  10000 , fixed pt res:  tensor(0.0501) , ls:  1 , rel_err:  tensor(0.2859) , time:  0.0026171207427978516
trial  1  finished after 27.39184856414795




 CHANGE, delta =  0.1
norm(rel_err_hist_temp - rel_err_lmm_hist_delta_array[sample_index,:]) =  tensor(0.)
delta = 1.0
iter:  50 , fixed pt res:  tensor(0.6005) , ls:  1 , rel_err:  tensor(15.2768) , time:  0.0026540756225585938
iter:  100 , fixed pt res:  tensor(0.2584) , ls:  1 , rel_err:  tensor(14.1514) , time:  0.002636432647705078
iter:  150 , fixed pt res:  tensor(0.1262) , ls:  1 , rel_err:  tensor(13.2173) , time:  0.0026426315307617188
iter:  200 , fixed pt res:  tensor(0.0774) , ls:  1 , rel_err:  tensor(12.3428) , time:  0.0026314258575439453
iter:  250 , fixed pt res:  tensor(0.0582) , ls:  1 , rel_err:  tensor(11.5279) , time:  0.0026137828826904297
iter:  300 , fixed pt res:  tensor(0.0570) , ls:  1 , rel_err:  tensor(10.7427) , time:  0.0026383399963378906
iter:  350 , fixed pt res:  tensor(0.0529) , ls:  1 , rel_err:  tensor(10.0210) , time:  0.002575397491455078
iter:  400 , fixed pt res:  tensor(0.0512) , ls:  1 , rel_err:  tensor(9.3471) , time:  0.0026078224182128906
iter:  450 , fixed pt res:  tensor(0.0520) , ls:  1 , rel_err:  tensor(8.7221) , time:  0.0026149749755859375
iter:  500 , fixed pt res:  tensor(0.0506) , ls:  1 , rel_err:  tensor(8.1162) , time:  0.0028815269470214844
iter:  550 , fixed pt res:  tensor(0.0497) , ls:  1 , rel_err:  tensor(7.5650) , time:  0.0026204586029052734
iter:  600 , fixed pt res:  tensor(0.0510) , ls:  1 , rel_err:  tensor(7.0442) , time:  0.002576112747192383
iter:  650 , fixed pt res:  tensor(0.0493) , ls:  1 , rel_err:  tensor(6.5619) , time:  0.0026092529296875
iter:  700 , fixed pt res:  tensor(0.0482) , ls:  1 , rel_err:  tensor(6.1032) , time:  0.002608776092529297
iter:  750 , fixed pt res:  tensor(0.0479) , ls:  1 , rel_err:  tensor(5.6962) , time:  0.002565622329711914
iter:  800 , fixed pt res:  tensor(0.0467) , ls:  1 , rel_err:  tensor(5.3075) , time:  0.0025768280029296875
iter:  850 , fixed pt res:  tensor(0.0467) , ls:  1 , rel_err:  tensor(4.9531) , time:  0.002599954605102539
iter:  900 , fixed pt res:  tensor(0.0480) , ls:  1 , rel_err:  tensor(4.6159) , time:  0.0026197433471679688
iter:  950 , fixed pt res:  tensor(0.0478) , ls:  1 , rel_err:  tensor(4.3099) , time:  0.002585887908935547
iter:  1000 , fixed pt res:  tensor(0.0481) , ls:  1 , rel_err:  tensor(4.0246) , time:  0.002577066421508789
iter:  1050 , fixed pt res:  tensor(0.0474) , ls:  1 , rel_err:  tensor(3.7718) , time:  0.0026106834411621094
iter:  1100 , fixed pt res:  tensor(0.0468) , ls:  1 , rel_err:  tensor(3.5245) , time:  0.0026373863220214844
iter:  1150 , fixed pt res:  tensor(0.0470) , ls:  1 , rel_err:  tensor(3.3049) , time:  0.002629518508911133
iter:  1200 , fixed pt res:  tensor(0.0473) , ls:  1 , rel_err:  tensor(3.0891) , time:  0.0026445388793945312
iter:  1250 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(2.8850) , time:  0.0026628971099853516
iter:  1300 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(2.7110) , time:  0.00262451171875
iter:  1350 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(2.5294) , time:  0.0026557445526123047
iter:  1400 , fixed pt res:  tensor(0.0435) , ls:  1 , rel_err:  tensor(2.3791) , time:  0.0026009082794189453
iter:  1450 , fixed pt res:  tensor(0.0459) , ls:  1 , rel_err:  tensor(2.2291) , time:  0.0026459693908691406
iter:  1500 , fixed pt res:  tensor(0.0456) , ls:  1 , rel_err:  tensor(2.0951) , time:  0.002675294876098633
iter:  1550 , fixed pt res:  tensor(0.0459) , ls:  1 , rel_err:  tensor(1.9579) , time:  0.0025932788848876953
iter:  1600 , fixed pt res:  tensor(0.0456) , ls:  1 , rel_err:  tensor(1.8430) , time:  0.002668619155883789
iter:  1650 , fixed pt res:  tensor(0.0447) , ls:  1 , rel_err:  tensor(1.7305) , time:  0.0026373863220214844
iter:  1700 , fixed pt res:  tensor(0.0459) , ls:  1 , rel_err:  tensor(1.6303) , time:  0.002611398696899414
iter:  1750 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(1.5428) , time:  0.002649068832397461
iter:  1800 , fixed pt res:  tensor(0.0458) , ls:  1 , rel_err:  tensor(1.4608) , time:  0.0026705265045166016
iter:  1850 , fixed pt res:  tensor(0.0458) , ls:  1 , rel_err:  tensor(1.3826) , time:  0.0026466846466064453
iter:  1900 , fixed pt res:  tensor(0.0455) , ls:  1 , rel_err:  tensor(1.3036) , time:  0.002669095993041992
iter:  1950 , fixed pt res:  tensor(0.0473) , ls:  1 , rel_err:  tensor(1.2278) , time:  0.0026056766510009766
iter:  2000 , fixed pt res:  tensor(0.0471) , ls:  1 , rel_err:  tensor(1.1679) , time:  0.0026493072509765625
iter:  2050 , fixed pt res:  tensor(0.0473) , ls:  1 , rel_err:  tensor(1.1019) , time:  0.0026268959045410156
iter:  2100 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(1.0464) , time:  0.00262451171875
iter:  2150 , fixed pt res:  tensor(0.0455) , ls:  1 , rel_err:  tensor(1.0033) , time:  0.002627849578857422
iter:  2200 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.9490) , time:  0.002641916275024414
iter:  2250 , fixed pt res:  tensor(0.0462) , ls:  1 , rel_err:  tensor(0.8957) , time:  0.0026352405548095703
iter:  2300 , fixed pt res:  tensor(0.0463) , ls:  1 , rel_err:  tensor(0.8654) , time:  0.002621173858642578
iter:  2350 , fixed pt res:  tensor(0.0479) , ls:  1 , rel_err:  tensor(0.8243) , time:  0.0027053356170654297
iter:  2400 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.7950) , time:  0.0026209354400634766
iter:  2450 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(0.7708) , time:  0.0026350021362304688
iter:  2500 , fixed pt res:  tensor(0.0462) , ls:  1 , rel_err:  tensor(0.7380) , time:  0.002679109573364258
iter:  2550 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.7190) , time:  0.002644777297973633
iter:  2600 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.6741) , time:  0.0026051998138427734
iter:  2650 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.6421) , time:  0.0026192665100097656
iter:  2700 , fixed pt res:  tensor(0.0463) , ls:  1 , rel_err:  tensor(0.6235) , time:  0.0026314258575439453
iter:  2750 , fixed pt res:  tensor(0.0469) , ls:  1 , rel_err:  tensor(0.5928) , time:  0.002638578414916992
iter:  2800 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(0.5695) , time:  0.002611398696899414
iter:  2850 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(0.5487) , time:  0.0026407241821289062
iter:  2900 , fixed pt res:  tensor(0.0456) , ls:  1 , rel_err:  tensor(0.5377) , time:  0.002595186233520508
iter:  2950 , fixed pt res:  tensor(0.0462) , ls:  1 , rel_err:  tensor(0.5236) , time:  0.002619504928588867
iter:  3000 , fixed pt res:  tensor(0.0458) , ls:  1 , rel_err:  tensor(0.4881) , time:  0.0026061534881591797
iter:  3050 , fixed pt res:  tensor(0.0470) , ls:  1 , rel_err:  tensor(0.4761) , time:  0.0026426315307617188
iter:  3100 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(0.4674) , time:  0.002640962600708008
iter:  3150 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.4544) , time:  0.0026710033416748047
iter:  3200 , fixed pt res:  tensor(0.0461) , ls:  1 , rel_err:  tensor(0.4424) , time:  0.0026140213012695312
iter:  3250 , fixed pt res:  tensor(0.0455) , ls:  1 , rel_err:  tensor(0.4442) , time:  0.0026242733001708984
iter:  3300 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(0.4376) , time:  0.0026433467864990234
iter:  3350 , fixed pt res:  tensor(0.0458) , ls:  1 , rel_err:  tensor(0.4240) , time:  0.0026373863220214844
iter:  3400 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(0.4060) , time:  0.002625703811645508
iter:  3450 , fixed pt res:  tensor(0.0458) , ls:  1 , rel_err:  tensor(0.3985) , time:  0.002706766128540039
iter:  3500 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.3738) , time:  0.0026137828826904297
iter:  3550 , fixed pt res:  tensor(0.0475) , ls:  1 , rel_err:  tensor(0.3622) , time:  0.0026497840881347656
iter:  3600 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.3532) , time:  0.0025882720947265625
iter:  3650 , fixed pt res:  tensor(0.0442) , ls:  1 , rel_err:  tensor(0.3436) , time:  0.0026092529296875
iter:  3700 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.3315) , time:  0.0029354095458984375
iter:  3750 , fixed pt res:  tensor(0.0472) , ls:  1 , rel_err:  tensor(0.3287) , time:  0.002648591995239258
iter:  3800 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(0.3257) , time:  0.0026578903198242188
iter:  3850 , fixed pt res:  tensor(0.0467) , ls:  1 , rel_err:  tensor(0.3218) , time:  0.0026292800903320312
iter:  3900 , fixed pt res:  tensor(0.0445) , ls:  1 , rel_err:  tensor(0.3271) , time:  0.0026068687438964844
iter:  3950 , fixed pt res:  tensor(0.0455) , ls:  1 , rel_err:  tensor(0.3251) , time:  0.0026171207427978516
iter:  4000 , fixed pt res:  tensor(0.0449) , ls:  1 , rel_err:  tensor(0.3210) , time:  0.002641916275024414
iter:  4050 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.3092) , time:  0.002651214599609375
iter:  4100 , fixed pt res:  tensor(0.0469) , ls:  1 , rel_err:  tensor(0.3099) , time:  0.0026879310607910156
iter:  4150 , fixed pt res:  tensor(0.0466) , ls:  1 , rel_err:  tensor(0.3023) , time:  0.0026187896728515625
iter:  4200 , fixed pt res:  tensor(0.0465) , ls:  1 , rel_err:  tensor(0.2965) , time:  0.0026483535766601562
iter:  4250 , fixed pt res:  tensor(0.0459) , ls:  1 , rel_err:  tensor(0.2969) , time:  0.002678394317626953
iter:  4300 , fixed pt res:  tensor(0.0447) , ls:  1 , rel_err:  tensor(0.2840) , time:  0.0026154518127441406
iter:  4350 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.2805) , time:  0.0026285648345947266
iter:  4400 , fixed pt res:  tensor(0.0436) , ls:  1 , rel_err:  tensor(0.2865) , time:  0.002627134323120117
iter:  4450 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2870) , time:  0.002590179443359375
iter:  4500 , fixed pt res:  tensor(0.0463) , ls:  1 , rel_err:  tensor(0.2870) , time:  0.0026192665100097656
iter:  4550 , fixed pt res:  tensor(0.0470) , ls:  1 , rel_err:  tensor(0.2886) , time:  0.002604246139526367
iter:  4600 , fixed pt res:  tensor(0.0474) , ls:  1 , rel_err:  tensor(0.2938) , time:  0.002635478973388672
iter:  4650 , fixed pt res:  tensor(0.0472) , ls:  1 , rel_err:  tensor(0.2835) , time:  0.0026128292083740234
iter:  4700 , fixed pt res:  tensor(0.0470) , ls:  1 , rel_err:  tensor(0.2962) , time:  0.0026760101318359375
iter:  4750 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2938) , time:  0.002630472183227539
iter:  4800 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2811) , time:  0.0026006698608398438
iter:  4850 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(0.2715) , time:  0.0026330947875976562
iter:  4900 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.2852) , time:  0.0026869773864746094
iter:  4950 , fixed pt res:  tensor(0.0439) , ls:  1 , rel_err:  tensor(0.2726) , time:  0.002587556838989258
iter:  5000 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(0.2675) , time:  0.002599000930786133
iter:  5050 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.2669) , time:  0.002650737762451172
iter:  5100 , fixed pt res:  tensor(0.0472) , ls:  1 , rel_err:  tensor(0.2728) , time:  0.0026428699493408203
iter:  5150 , fixed pt res:  tensor(0.0461) , ls:  1 , rel_err:  tensor(0.2719) , time:  0.0026433467864990234
iter:  5200 , fixed pt res:  tensor(0.0458) , ls:  1 , rel_err:  tensor(0.2842) , time:  0.002634286880493164
iter:  5250 , fixed pt res:  tensor(0.0471) , ls:  1 , rel_err:  tensor(0.2828) , time:  0.002868175506591797
iter:  5300 , fixed pt res:  tensor(0.0466) , ls:  1 , rel_err:  tensor(0.2798) , time:  0.0026540756225585938
iter:  5350 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.2775) , time:  0.002628803253173828
iter:  5400 , fixed pt res:  tensor(0.0462) , ls:  1 , rel_err:  tensor(0.2736) , time:  0.0025908946990966797
iter:  5450 , fixed pt res:  tensor(0.0452) , ls:  1 , rel_err:  tensor(0.2676) , time:  0.0025773048400878906
iter:  5500 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(0.2721) , time:  0.002618551254272461
iter:  5550 , fixed pt res:  tensor(0.0447) , ls:  1 , rel_err:  tensor(0.2685) , time:  0.0026159286499023438
iter:  5600 , fixed pt res:  tensor(0.0475) , ls:  1 , rel_err:  tensor(0.2680) , time:  0.0026025772094726562
iter:  5650 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.2586) , time:  0.0025658607482910156
iter:  5700 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(0.2543) , time:  0.0026159286499023438
iter:  5750 , fixed pt res:  tensor(0.0466) , ls:  1 , rel_err:  tensor(0.2570) , time:  0.002605438232421875
iter:  5800 , fixed pt res:  tensor(0.0462) , ls:  1 , rel_err:  tensor(0.2530) , time:  0.003300905227661133
iter:  5850 , fixed pt res:  tensor(0.0470) , ls:  1 , rel_err:  tensor(0.2553) , time:  0.002602100372314453
iter:  5900 , fixed pt res:  tensor(0.0462) , ls:  1 , rel_err:  tensor(0.2602) , time:  0.0026242733001708984
iter:  5950 , fixed pt res:  tensor(0.0458) , ls:  1 , rel_err:  tensor(0.2625) , time:  0.0026111602783203125
iter:  6000 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(0.2580) , time:  0.0029571056365966797
iter:  6050 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.2541) , time:  0.0026388168334960938
iter:  6100 , fixed pt res:  tensor(0.0476) , ls:  1 , rel_err:  tensor(0.2559) , time:  0.0026400089263916016
iter:  6150 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(0.2496) , time:  0.002628803253173828
iter:  6200 , fixed pt res:  tensor(0.0456) , ls:  1 , rel_err:  tensor(0.2486) , time:  0.002633333206176758
iter:  6250 , fixed pt res:  tensor(0.0471) , ls:  1 , rel_err:  tensor(0.2526) , time:  0.0025932788848876953
iter:  6300 , fixed pt res:  tensor(0.0464) , ls:  1 , rel_err:  tensor(0.2486) , time:  0.0026178359985351562
iter:  6350 , fixed pt res:  tensor(0.0471) , ls:  1 , rel_err:  tensor(0.2381) , time:  0.002589702606201172
iter:  6400 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.2374) , time:  0.0026047229766845703
iter:  6450 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2355) , time:  0.002631664276123047
iter:  6500 , fixed pt res:  tensor(0.0465) , ls:  1 , rel_err:  tensor(0.2353) , time:  0.0025975704193115234
iter:  6550 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(0.2260) , time:  0.002611875534057617
iter:  6600 , fixed pt res:  tensor(0.0461) , ls:  1 , rel_err:  tensor(0.2244) , time:  0.002643585205078125
iter:  6650 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(0.2314) , time:  0.002623319625854492
iter:  6700 , fixed pt res:  tensor(0.0471) , ls:  1 , rel_err:  tensor(0.2334) , time:  0.0026040077209472656
iter:  6750 , fixed pt res:  tensor(0.0445) , ls:  1 , rel_err:  tensor(0.2349) , time:  0.002633333206176758
iter:  6800 , fixed pt res:  tensor(0.0459) , ls:  1 , rel_err:  tensor(0.2346) , time:  0.0026187896728515625
iter:  6850 , fixed pt res:  tensor(0.0466) , ls:  1 , rel_err:  tensor(0.2335) , time:  0.002584218978881836
iter:  6900 , fixed pt res:  tensor(0.0478) , ls:  1 , rel_err:  tensor(0.2362) , time:  0.002745389938354492
iter:  6950 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2329) , time:  0.00258636474609375
iter:  7000 , fixed pt res:  tensor(0.0449) , ls:  1 , rel_err:  tensor(0.2421) , time:  0.002652883529663086
iter:  7050 , fixed pt res:  tensor(0.0470) , ls:  1 , rel_err:  tensor(0.2474) , time:  0.0026178359985351562
iter:  7100 , fixed pt res:  tensor(0.0467) , ls:  1 , rel_err:  tensor(0.2483) , time:  0.0026175975799560547
iter:  7150 , fixed pt res:  tensor(0.0458) , ls:  1 , rel_err:  tensor(0.2538) , time:  0.0026121139526367188
iter:  7200 , fixed pt res:  tensor(0.0465) , ls:  1 , rel_err:  tensor(0.2634) , time:  0.0026302337646484375
iter:  7250 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.2635) , time:  0.002633333206176758
iter:  7300 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.2602) , time:  0.0025920867919921875
iter:  7350 , fixed pt res:  tensor(0.0461) , ls:  1 , rel_err:  tensor(0.2569) , time:  0.00262451171875
iter:  7400 , fixed pt res:  tensor(0.0468) , ls:  1 , rel_err:  tensor(0.2582) , time:  0.0026018619537353516
iter:  7450 , fixed pt res:  tensor(0.0452) , ls:  1 , rel_err:  tensor(0.2591) , time:  0.002658367156982422
iter:  7500 , fixed pt res:  tensor(0.0460) , ls:  1 , rel_err:  tensor(0.2521) , time:  0.0026044845581054688
iter:  7550 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(0.2517) , time:  0.002680063247680664
iter:  7600 , fixed pt res:  tensor(0.0463) , ls:  1 , rel_err:  tensor(0.2522) , time:  0.002724170684814453
iter:  7650 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(0.2457) , time:  0.002649068832397461
iter:  7700 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2462) , time:  0.002586841583251953
iter:  7750 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.2563) , time:  0.0026137828826904297
iter:  7800 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2561) , time:  0.002600431442260742
iter:  7850 , fixed pt res:  tensor(0.0464) , ls:  1 , rel_err:  tensor(0.2601) , time:  0.002637147903442383
iter:  7900 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(0.2634) , time:  0.0025911331176757812
iter:  7950 , fixed pt res:  tensor(0.0480) , ls:  1 , rel_err:  tensor(0.2607) , time:  0.0025832653045654297
iter:  8000 , fixed pt res:  tensor(0.0449) , ls:  1 , rel_err:  tensor(0.2636) , time:  0.0030062198638916016
iter:  8050 , fixed pt res:  tensor(0.0472) , ls:  1 , rel_err:  tensor(0.2702) , time:  0.0026273727416992188
iter:  8100 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2676) , time:  0.0026102066040039062
iter:  8150 , fixed pt res:  tensor(0.0474) , ls:  1 , rel_err:  tensor(0.2655) , time:  0.002643108367919922
iter:  8200 , fixed pt res:  tensor(0.0470) , ls:  1 , rel_err:  tensor(0.2644) , time:  0.002836942672729492
iter:  8250 , fixed pt res:  tensor(0.0452) , ls:  1 , rel_err:  tensor(0.2599) , time:  0.0026481151580810547
iter:  8300 , fixed pt res:  tensor(0.0458) , ls:  1 , rel_err:  tensor(0.2528) , time:  0.0026917457580566406
iter:  8350 , fixed pt res:  tensor(0.0460) , ls:  1 , rel_err:  tensor(0.2515) , time:  0.002638101577758789
iter:  8400 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.2581) , time:  0.0026161670684814453
iter:  8450 , fixed pt res:  tensor(0.0460) , ls:  1 , rel_err:  tensor(0.2615) , time:  0.0026121139526367188
iter:  8500 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.2630) , time:  0.002630472183227539
iter:  8550 , fixed pt res:  tensor(0.0468) , ls:  1 , rel_err:  tensor(0.2544) , time:  0.0026378631591796875
iter:  8600 , fixed pt res:  tensor(0.0456) , ls:  1 , rel_err:  tensor(0.2536) , time:  0.0028374195098876953
iter:  8650 , fixed pt res:  tensor(0.0469) , ls:  1 , rel_err:  tensor(0.2555) , time:  0.0026123523712158203
iter:  8700 , fixed pt res:  tensor(0.0460) , ls:  1 , rel_err:  tensor(0.2586) , time:  0.0026521682739257812
iter:  8750 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.2590) , time:  0.0026025772094726562
iter:  8800 , fixed pt res:  tensor(0.0445) , ls:  1 , rel_err:  tensor(0.2606) , time:  0.0026471614837646484
iter:  8850 , fixed pt res:  tensor(0.0473) , ls:  1 , rel_err:  tensor(0.2578) , time:  0.0026187896728515625
iter:  8900 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.2623) , time:  0.002620697021484375
iter:  8950 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.2704) , time:  0.002619504928588867
iter:  9000 , fixed pt res:  tensor(0.0473) , ls:  1 , rel_err:  tensor(0.2746) , time:  0.00292205810546875
iter:  9050 , fixed pt res:  tensor(0.0445) , ls:  1 , rel_err:  tensor(0.2834) , time:  0.002611875534057617
iter:  9100 , fixed pt res:  tensor(0.0460) , ls:  1 , rel_err:  tensor(0.2765) , time:  0.002656698226928711
iter:  9150 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2818) , time:  0.0026679039001464844
iter:  9200 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(0.2774) , time:  0.002641916275024414
iter:  9250 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(0.2712) , time:  0.0025887489318847656
iter:  9300 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.2731) , time:  0.002611875534057617
iter:  9350 , fixed pt res:  tensor(0.0439) , ls:  1 , rel_err:  tensor(0.2686) , time:  0.002651691436767578
iter:  9400 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2698) , time:  0.0026099681854248047
iter:  9450 , fixed pt res:  tensor(0.0459) , ls:  1 , rel_err:  tensor(0.2642) , time:  0.002611398696899414
iter:  9500 , fixed pt res:  tensor(0.0494) , ls:  1 , rel_err:  tensor(0.2521) , time:  0.0028138160705566406
iter:  9550 , fixed pt res:  tensor(0.0478) , ls:  1 , rel_err:  tensor(0.2604) , time:  0.002634763717651367
iter:  9600 , fixed pt res:  tensor(0.0476) , ls:  1 , rel_err:  tensor(0.2566) , time:  0.0026116371154785156
iter:  9650 , fixed pt res:  tensor(0.0473) , ls:  1 , rel_err:  tensor(0.2515) , time:  0.002725362777709961
iter:  9700 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2516) , time:  0.0025947093963623047
iter:  9750 , fixed pt res:  tensor(0.0462) , ls:  1 , rel_err:  tensor(0.2511) , time:  0.0026650428771972656
iter:  9800 , fixed pt res:  tensor(0.0480) , ls:  1 , rel_err:  tensor(0.2353) , time:  0.002584218978881836
iter:  9850 , fixed pt res:  tensor(0.0459) , ls:  1 , rel_err:  tensor(0.2436) , time:  0.0028257369995117188
iter:  9900 , fixed pt res:  tensor(0.0466) , ls:  1 , rel_err:  tensor(0.2401) , time:  0.002621173858642578
iter:  9950 , fixed pt res:  tensor(0.0459) , ls:  1 , rel_err:  tensor(0.2476) , time:  0.002672910690307617
iter:  10000 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(0.2453) , time:  0.0026237964630126953
trial  1  finished after 27.315054655075073




 CHANGE, delta =  1.0
norm(rel_err_hist_temp - rel_err_lmm_hist_delta_array[sample_index,:]) =  tensor(0.)
delta = 10
iter:  50 , fixed pt res:  tensor(0.6019) , ls:  1 , rel_err:  tensor(15.2753) , time:  0.0026540756225585938
iter:  100 , fixed pt res:  tensor(0.2578) , ls:  1 , rel_err:  tensor(14.1615) , time:  0.0026514530181884766
iter:  150 , fixed pt res:  tensor(0.1269) , ls:  1 , rel_err:  tensor(13.2275) , time:  0.0026235580444335938
iter:  200 , fixed pt res:  tensor(0.0762) , ls:  1 , rel_err:  tensor(12.3565) , time:  0.002614736557006836
iter:  250 , fixed pt res:  tensor(0.0597) , ls:  1 , rel_err:  tensor(11.5210) , time:  0.002669095993041992
iter:  300 , fixed pt res:  tensor(0.0571) , ls:  1 , rel_err:  tensor(10.7411) , time:  0.0026438236236572266
iter:  350 , fixed pt res:  tensor(0.0520) , ls:  1 , rel_err:  tensor(10.0191) , time:  0.002608060836791992
iter:  400 , fixed pt res:  tensor(0.0518) , ls:  1 , rel_err:  tensor(9.3347) , time:  0.0026454925537109375
iter:  450 , fixed pt res:  tensor(0.0510) , ls:  1 , rel_err:  tensor(8.6942) , time:  0.002699136734008789
iter:  500 , fixed pt res:  tensor(0.0485) , ls:  1 , rel_err:  tensor(8.0995) , time:  0.002779245376586914
iter:  550 , fixed pt res:  tensor(0.0510) , ls:  1 , rel_err:  tensor(7.5484) , time:  0.0026400089263916016
iter:  600 , fixed pt res:  tensor(0.0489) , ls:  1 , rel_err:  tensor(7.0212) , time:  0.0026590824127197266
iter:  650 , fixed pt res:  tensor(0.0483) , ls:  1 , rel_err:  tensor(6.5332) , time:  0.0026786327362060547
iter:  700 , fixed pt res:  tensor(0.0466) , ls:  1 , rel_err:  tensor(6.0990) , time:  0.002670764923095703
iter:  750 , fixed pt res:  tensor(0.0494) , ls:  1 , rel_err:  tensor(5.6841) , time:  0.0026121139526367188
iter:  800 , fixed pt res:  tensor(0.0467) , ls:  1 , rel_err:  tensor(5.3001) , time:  0.002619504928588867
iter:  850 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(4.9403) , time:  0.002660512924194336
iter:  900 , fixed pt res:  tensor(0.0465) , ls:  1 , rel_err:  tensor(4.5992) , time:  0.002605438232421875
iter:  950 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(4.3029) , time:  0.0026481151580810547
iter:  1000 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(4.0176) , time:  0.002672910690307617
iter:  1050 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(3.7485) , time:  0.0026397705078125
iter:  1100 , fixed pt res:  tensor(0.0456) , ls:  1 , rel_err:  tensor(3.5050) , time:  0.0026443004608154297
iter:  1150 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(3.2698) , time:  0.002622842788696289
iter:  1200 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(3.0443) , time:  0.0027205944061279297
iter:  1250 , fixed pt res:  tensor(0.0427) , ls:  1 , rel_err:  tensor(2.8493) , time:  0.002618551254272461
iter:  1300 , fixed pt res:  tensor(0.0456) , ls:  1 , rel_err:  tensor(2.6783) , time:  0.0026373863220214844
iter:  1350 , fixed pt res:  tensor(0.0414) , ls:  1 , rel_err:  tensor(2.5112) , time:  0.0026564598083496094
iter:  1400 , fixed pt res:  tensor(0.0455) , ls:  1 , rel_err:  tensor(2.3520) , time:  0.0026865005493164062
iter:  1450 , fixed pt res:  tensor(0.0429) , ls:  1 , rel_err:  tensor(2.2008) , time:  0.0027163028717041016
iter:  1500 , fixed pt res:  tensor(0.0468) , ls:  1 , rel_err:  tensor(2.0657) , time:  0.0026688575744628906
iter:  1550 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(1.9482) , time:  0.002709627151489258
iter:  1600 , fixed pt res:  tensor(0.0456) , ls:  1 , rel_err:  tensor(1.8259) , time:  0.0026383399963378906
iter:  1650 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(1.7236) , time:  0.0026502609252929688
iter:  1700 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(1.6258) , time:  0.0026776790618896484
iter:  1750 , fixed pt res:  tensor(0.0434) , ls:  1 , rel_err:  tensor(1.5325) , time:  0.002692699432373047
iter:  1800 , fixed pt res:  tensor(0.0459) , ls:  1 , rel_err:  tensor(1.4510) , time:  0.0027060508728027344
iter:  1850 , fixed pt res:  tensor(0.0431) , ls:  1 , rel_err:  tensor(1.3683) , time:  0.0026793479919433594
iter:  1900 , fixed pt res:  tensor(0.0434) , ls:  1 , rel_err:  tensor(1.2950) , time:  0.002918243408203125
iter:  1950 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(1.2262) , time:  0.0026624202728271484
iter:  2000 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(1.1582) , time:  0.0028617382049560547
iter:  2050 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(1.1049) , time:  0.002691507339477539
iter:  2100 , fixed pt res:  tensor(0.0447) , ls:  1 , rel_err:  tensor(1.0510) , time:  0.0027077198028564453
iter:  2150 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.9996) , time:  0.0026416778564453125
iter:  2200 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(0.9575) , time:  0.002642393112182617
iter:  2250 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.9075) , time:  0.002653837203979492
iter:  2300 , fixed pt res:  tensor(0.0430) , ls:  1 , rel_err:  tensor(0.8576) , time:  0.002691984176635742
iter:  2350 , fixed pt res:  tensor(0.0435) , ls:  1 , rel_err:  tensor(0.8157) , time:  0.002669811248779297
iter:  2400 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.7775) , time:  0.002611398696899414
iter:  2450 , fixed pt res:  tensor(0.0433) , ls:  1 , rel_err:  tensor(0.7434) , time:  0.002624034881591797
iter:  2500 , fixed pt res:  tensor(0.0427) , ls:  1 , rel_err:  tensor(0.7074) , time:  0.0026209354400634766
iter:  2550 , fixed pt res:  tensor(0.0449) , ls:  1 , rel_err:  tensor(0.6806) , time:  0.0026488304138183594
iter:  2600 , fixed pt res:  tensor(0.0434) , ls:  1 , rel_err:  tensor(0.6435) , time:  0.002690553665161133
iter:  2650 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.6135) , time:  0.0027344226837158203
iter:  2700 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.5929) , time:  0.002749204635620117
iter:  2750 , fixed pt res:  tensor(0.0447) , ls:  1 , rel_err:  tensor(0.5653) , time:  0.0027985572814941406
iter:  2800 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.5382) , time:  0.002658843994140625
iter:  2850 , fixed pt res:  tensor(0.0468) , ls:  1 , rel_err:  tensor(0.5252) , time:  0.002630949020385742
iter:  2900 , fixed pt res:  tensor(0.0455) , ls:  1 , rel_err:  tensor(0.5129) , time:  0.002830982208251953
iter:  2950 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.4813) , time:  0.0026781558990478516
iter:  3000 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(0.4677) , time:  0.0026559829711914062
iter:  3050 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.4501) , time:  0.002631664276123047
iter:  3100 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(0.4472) , time:  0.0026276111602783203
iter:  3150 , fixed pt res:  tensor(0.0436) , ls:  1 , rel_err:  tensor(0.4336) , time:  0.002597808837890625
iter:  3200 , fixed pt res:  tensor(0.0456) , ls:  1 , rel_err:  tensor(0.4200) , time:  0.002661466598510742
iter:  3250 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.4142) , time:  0.002621889114379883
iter:  3300 , fixed pt res:  tensor(0.0421) , ls:  1 , rel_err:  tensor(0.3954) , time:  0.002713441848754883
iter:  3350 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(0.3863) , time:  0.0026400089263916016
iter:  3400 , fixed pt res:  tensor(0.0431) , ls:  1 , rel_err:  tensor(0.3733) , time:  0.002678394317626953
iter:  3450 , fixed pt res:  tensor(0.0434) , ls:  1 , rel_err:  tensor(0.3668) , time:  0.002679109573364258
iter:  3500 , fixed pt res:  tensor(0.0456) , ls:  1 , rel_err:  tensor(0.3535) , time:  0.0026700496673583984
iter:  3550 , fixed pt res:  tensor(0.0439) , ls:  1 , rel_err:  tensor(0.3507) , time:  0.0026290416717529297
iter:  3600 , fixed pt res:  tensor(0.0459) , ls:  1 , rel_err:  tensor(0.3416) , time:  0.0026311874389648438
iter:  3650 , fixed pt res:  tensor(0.0452) , ls:  1 , rel_err:  tensor(0.3300) , time:  0.0027027130126953125
iter:  3700 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(0.3323) , time:  0.0026731491088867188
iter:  3750 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.3244) , time:  0.002677440643310547
iter:  3800 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(0.3207) , time:  0.0026001930236816406
iter:  3850 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.3108) , time:  0.0026946067810058594
iter:  3900 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.3088) , time:  0.0026557445526123047
iter:  3950 , fixed pt res:  tensor(0.0466) , ls:  1 , rel_err:  tensor(0.3105) , time:  0.0026497840881347656
iter:  4000 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(0.3087) , time:  0.002724170684814453
iter:  4050 , fixed pt res:  tensor(0.0442) , ls:  1 , rel_err:  tensor(0.3149) , time:  0.002661466598510742
iter:  4100 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.3114) , time:  0.0026755332946777344
iter:  4150 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.3084) , time:  0.002646923065185547
iter:  4200 , fixed pt res:  tensor(0.0435) , ls:  1 , rel_err:  tensor(0.3067) , time:  0.0026345252990722656
iter:  4250 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(0.2983) , time:  0.002586841583251953
iter:  4300 , fixed pt res:  tensor(0.0459) , ls:  1 , rel_err:  tensor(0.3007) , time:  0.002732515335083008
iter:  4350 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.2996) , time:  0.00261688232421875
iter:  4400 , fixed pt res:  tensor(0.0449) , ls:  1 , rel_err:  tensor(0.2934) , time:  0.002596139907836914
iter:  4450 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(0.2900) , time:  0.002726316452026367
iter:  4500 , fixed pt res:  tensor(0.0431) , ls:  1 , rel_err:  tensor(0.2807) , time:  0.002689361572265625
iter:  4550 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.2853) , time:  0.0026869773864746094
iter:  4600 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.2876) , time:  0.0026433467864990234
iter:  4650 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.2823) , time:  0.0026700496673583984
iter:  4700 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.2841) , time:  0.0026471614837646484
iter:  4750 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.2762) , time:  0.0026712417602539062
iter:  4800 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.2731) , time:  0.002760648727416992
iter:  4850 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(0.2585) , time:  0.002687215805053711
iter:  4900 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.2593) , time:  0.00275421142578125
iter:  4950 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(0.2579) , time:  0.0027098655700683594
iter:  5000 , fixed pt res:  tensor(0.0432) , ls:  1 , rel_err:  tensor(0.2592) , time:  0.002589702606201172
iter:  5050 , fixed pt res:  tensor(0.0452) , ls:  1 , rel_err:  tensor(0.2644) , time:  0.0025992393493652344
iter:  5100 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(0.2573) , time:  0.002879619598388672
iter:  5150 , fixed pt res:  tensor(0.0431) , ls:  1 , rel_err:  tensor(0.2536) , time:  0.002627849578857422
iter:  5200 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(0.2591) , time:  0.002704143524169922
iter:  5250 , fixed pt res:  tensor(0.0462) , ls:  1 , rel_err:  tensor(0.2442) , time:  0.002785921096801758
iter:  5300 , fixed pt res:  tensor(0.0469) , ls:  1 , rel_err:  tensor(0.2435) , time:  0.0026731491088867188
iter:  5350 , fixed pt res:  tensor(0.0462) , ls:  1 , rel_err:  tensor(0.2527) , time:  0.0026497840881347656
iter:  5400 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(0.2567) , time:  0.002666473388671875
iter:  5450 , fixed pt res:  tensor(0.0435) , ls:  1 , rel_err:  tensor(0.2584) , time:  0.0027420520782470703
iter:  5500 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(0.2566) , time:  0.002741575241088867
iter:  5550 , fixed pt res:  tensor(0.0427) , ls:  1 , rel_err:  tensor(0.2675) , time:  0.0027132034301757812
iter:  5600 , fixed pt res:  tensor(0.0455) , ls:  1 , rel_err:  tensor(0.2643) , time:  0.002718210220336914
iter:  5650 , fixed pt res:  tensor(0.0439) , ls:  1 , rel_err:  tensor(0.2668) , time:  0.002752065658569336
iter:  5700 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(0.2581) , time:  0.0026581287384033203
iter:  5750 , fixed pt res:  tensor(0.0428) , ls:  1 , rel_err:  tensor(0.2603) , time:  0.002724170684814453
iter:  5800 , fixed pt res:  tensor(0.0442) , ls:  1 , rel_err:  tensor(0.2698) , time:  0.0027420520782470703
iter:  5850 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(0.2633) , time:  0.0026688575744628906
iter:  5900 , fixed pt res:  tensor(0.0430) , ls:  1 , rel_err:  tensor(0.2637) , time:  0.0026891231536865234
iter:  5950 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.2634) , time:  0.002692699432373047
iter:  6000 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.2569) , time:  0.002665996551513672
iter:  6050 , fixed pt res:  tensor(0.0449) , ls:  1 , rel_err:  tensor(0.2561) , time:  0.002684354782104492
iter:  6100 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(0.2504) , time:  0.002671957015991211
iter:  6150 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.2581) , time:  0.0026755332946777344
iter:  6200 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2574) , time:  0.0026607513427734375
iter:  6250 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(0.2477) , time:  0.002636432647705078
iter:  6300 , fixed pt res:  tensor(0.0433) , ls:  1 , rel_err:  tensor(0.2521) , time:  0.0027189254760742188
iter:  6350 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(0.2508) , time:  0.0027017593383789062
iter:  6400 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.2479) , time:  0.0026891231536865234
iter:  6450 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(0.2471) , time:  0.002687692642211914
iter:  6500 , fixed pt res:  tensor(0.0432) , ls:  1 , rel_err:  tensor(0.2560) , time:  0.002639293670654297
iter:  6550 , fixed pt res:  tensor(0.0436) , ls:  1 , rel_err:  tensor(0.2571) , time:  0.002622842788696289
iter:  6600 , fixed pt res:  tensor(0.0431) , ls:  1 , rel_err:  tensor(0.2565) , time:  0.0026454925537109375
iter:  6650 , fixed pt res:  tensor(0.0434) , ls:  1 , rel_err:  tensor(0.2631) , time:  0.002637147903442383
iter:  6700 , fixed pt res:  tensor(0.0436) , ls:  1 , rel_err:  tensor(0.2600) , time:  0.0026543140411376953
iter:  6750 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(0.2621) , time:  0.00264739990234375
iter:  6800 , fixed pt res:  tensor(0.0461) , ls:  1 , rel_err:  tensor(0.2623) , time:  0.0032041072845458984
iter:  6850 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.2492) , time:  0.0026340484619140625
iter:  6900 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.2476) , time:  0.002618551254272461
iter:  6950 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(0.2500) , time:  0.0026683807373046875
iter:  7000 , fixed pt res:  tensor(0.0442) , ls:  1 , rel_err:  tensor(0.2462) , time:  0.0026750564575195312
iter:  7050 , fixed pt res:  tensor(0.0439) , ls:  1 , rel_err:  tensor(0.2455) , time:  0.0026268959045410156
iter:  7100 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.2419) , time:  0.0026750564575195312
iter:  7150 , fixed pt res:  tensor(0.0442) , ls:  1 , rel_err:  tensor(0.2353) , time:  0.0025870800018310547
iter:  7200 , fixed pt res:  tensor(0.0426) , ls:  1 , rel_err:  tensor(0.2313) , time:  0.0027091503143310547
iter:  7250 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.2405) , time:  0.0025959014892578125
iter:  7300 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(0.2351) , time:  0.0026917457580566406
iter:  7350 , fixed pt res:  tensor(0.0419) , ls:  1 , rel_err:  tensor(0.2425) , time:  0.0026624202728271484
iter:  7400 , fixed pt res:  tensor(0.0427) , ls:  1 , rel_err:  tensor(0.2339) , time:  0.002680063247680664
iter:  7450 , fixed pt res:  tensor(0.0431) , ls:  1 , rel_err:  tensor(0.2306) , time:  0.0026695728302001953
iter:  7500 , fixed pt res:  tensor(0.0420) , ls:  1 , rel_err:  tensor(0.2351) , time:  0.0026400089263916016
iter:  7550 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(0.2418) , time:  0.0026137828826904297
iter:  7600 , fixed pt res:  tensor(0.0434) , ls:  1 , rel_err:  tensor(0.2372) , time:  0.0026285648345947266
iter:  7650 , fixed pt res:  tensor(0.0435) , ls:  1 , rel_err:  tensor(0.2312) , time:  0.002652406692504883
iter:  7700 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(0.2385) , time:  0.0028412342071533203
iter:  7750 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.2259) , time:  0.002637147903442383
iter:  7800 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(0.2300) , time:  0.0027627944946289062
iter:  7850 , fixed pt res:  tensor(0.0431) , ls:  1 , rel_err:  tensor(0.2300) , time:  0.0026209354400634766
iter:  7900 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(0.2359) , time:  0.0027120113372802734
iter:  7950 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.2265) , time:  0.002808094024658203
iter:  8000 , fixed pt res:  tensor(0.0429) , ls:  1 , rel_err:  tensor(0.2234) , time:  0.002773761749267578
iter:  8050 , fixed pt res:  tensor(0.0433) , ls:  1 , rel_err:  tensor(0.2249) , time:  0.0026445388793945312
iter:  8100 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(0.2226) , time:  0.0027518272399902344
iter:  8150 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.2228) , time:  0.0026426315307617188
iter:  8200 , fixed pt res:  tensor(0.0466) , ls:  1 , rel_err:  tensor(0.2248) , time:  0.002750873565673828
iter:  8250 , fixed pt res:  tensor(0.0421) , ls:  1 , rel_err:  tensor(0.2283) , time:  0.002641439437866211
iter:  8300 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(0.2306) , time:  0.0026345252990722656
iter:  8350 , fixed pt res:  tensor(0.0442) , ls:  1 , rel_err:  tensor(0.2301) , time:  0.002625703811645508
iter:  8400 , fixed pt res:  tensor(0.0427) , ls:  1 , rel_err:  tensor(0.2244) , time:  0.0026750564575195312
iter:  8450 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(0.2286) , time:  0.0026092529296875
iter:  8500 , fixed pt res:  tensor(0.0449) , ls:  1 , rel_err:  tensor(0.2295) , time:  0.0026237964630126953
iter:  8550 , fixed pt res:  tensor(0.0447) , ls:  1 , rel_err:  tensor(0.2288) , time:  0.0026187896728515625
iter:  8600 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(0.2285) , time:  0.002825498580932617
iter:  8650 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(0.2294) , time:  0.002656698226928711
iter:  8700 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(0.2260) , time:  0.002588033676147461
iter:  8750 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(0.2314) , time:  0.0026361942291259766
iter:  8800 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(0.2353) , time:  0.0026760101318359375
iter:  8850 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2314) , time:  0.0026254653930664062
iter:  8900 , fixed pt res:  tensor(0.0447) , ls:  1 , rel_err:  tensor(0.2321) , time:  0.002634763717651367
iter:  8950 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.2350) , time:  0.0026259422302246094
iter:  9000 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.2309) , time:  0.002634763717651367
iter:  9050 , fixed pt res:  tensor(0.0431) , ls:  1 , rel_err:  tensor(0.2292) , time:  0.0026578903198242188
iter:  9100 , fixed pt res:  tensor(0.0447) , ls:  1 , rel_err:  tensor(0.2276) , time:  0.002710580825805664
iter:  9150 , fixed pt res:  tensor(0.0434) , ls:  1 , rel_err:  tensor(0.2342) , time:  0.002606630325317383
iter:  9200 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.2367) , time:  0.0026199817657470703
iter:  9250 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(0.2335) , time:  0.002606630325317383
iter:  9300 , fixed pt res:  tensor(0.0465) , ls:  1 , rel_err:  tensor(0.2380) , time:  0.0026226043701171875
iter:  9350 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(0.2328) , time:  0.002691030502319336
iter:  9400 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(0.2285) , time:  0.0028655529022216797
iter:  9450 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(0.2304) , time:  0.0026679039001464844
iter:  9500 , fixed pt res:  tensor(0.0452) , ls:  1 , rel_err:  tensor(0.2202) , time:  0.002732515335083008
iter:  9550 , fixed pt res:  tensor(0.0455) , ls:  1 , rel_err:  tensor(0.2295) , time:  0.0026450157165527344
iter:  9600 , fixed pt res:  tensor(0.0433) , ls:  1 , rel_err:  tensor(0.2291) , time:  0.0028519630432128906
iter:  9650 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(0.2385) , time:  0.0026679039001464844
iter:  9700 , fixed pt res:  tensor(0.0431) , ls:  1 , rel_err:  tensor(0.2467) , time:  0.0026292800903320312
iter:  9750 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(0.2429) , time:  0.0026733875274658203
iter:  9800 , fixed pt res:  tensor(0.0428) , ls:  1 , rel_err:  tensor(0.2391) , time:  0.002670764923095703
iter:  9850 , fixed pt res:  tensor(0.0432) , ls:  1 , rel_err:  tensor(0.2349) , time:  0.0026319026947021484
iter:  9900 , fixed pt res:  tensor(0.0455) , ls:  1 , rel_err:  tensor(0.2392) , time:  0.0026323795318603516
iter:  9950 , fixed pt res:  tensor(0.0462) , ls:  1 , rel_err:  tensor(0.2395) , time:  0.002651691436767578
iter:  10000 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2342) , time:  0.0026276111602783203
trial  1  finished after 27.629093647003174




 CHANGE, delta =  10
norm(rel_err_hist_temp - rel_err_lmm_hist_delta_array[sample_index,:]) =  tensor(0.)
delta = 100
iter:  50 , fixed pt res:  tensor(0.6023) , ls:  1 , rel_err:  tensor(15.2866) , time:  0.0026459693908691406
iter:  100 , fixed pt res:  tensor(0.2743) , ls:  1 , rel_err:  tensor(14.2015) , time:  0.002714872360229492
iter:  150 , fixed pt res:  tensor(0.1567) , ls:  1 , rel_err:  tensor(13.2935) , time:  0.0026772022247314453
iter:  200 , fixed pt res:  tensor(0.1120) , ls:  1 , rel_err:  tensor(12.4750) , time:  0.002736330032348633
iter:  250 , fixed pt res:  tensor(0.1090) , ls:  1 , rel_err:  tensor(11.7074) , time:  0.0026776790618896484
iter:  300 , fixed pt res:  tensor(0.1033) , ls:  1 , rel_err:  tensor(10.9887) , time:  0.002933025360107422
iter:  350 , fixed pt res:  tensor(0.1008) , ls:  1 , rel_err:  tensor(10.3216) , time:  0.0027055740356445312
iter:  400 , fixed pt res:  tensor(0.0972) , ls:  1 , rel_err:  tensor(9.7369) , time:  0.0026187896728515625
iter:  450 , fixed pt res:  tensor(0.0987) , ls:  1 , rel_err:  tensor(9.1810) , time:  0.0026667118072509766
iter:  500 , fixed pt res:  tensor(0.1002) , ls:  1 , rel_err:  tensor(8.6461) , time:  0.0026848316192626953
iter:  550 , fixed pt res:  tensor(0.0993) , ls:  1 , rel_err:  tensor(8.1415) , time:  0.0026662349700927734
iter:  600 , fixed pt res:  tensor(0.1000) , ls:  1 , rel_err:  tensor(7.6925) , time:  0.0027320384979248047
iter:  650 , fixed pt res:  tensor(0.0998) , ls:  1 , rel_err:  tensor(7.2723) , time:  0.002678394317626953
iter:  700 , fixed pt res:  tensor(0.0997) , ls:  1 , rel_err:  tensor(6.8970) , time:  0.0026404857635498047
iter:  750 , fixed pt res:  tensor(0.0987) , ls:  1 , rel_err:  tensor(6.5439) , time:  0.0026535987854003906
iter:  800 , fixed pt res:  tensor(0.0996) , ls:  1 , rel_err:  tensor(6.2256) , time:  0.002608060836791992
iter:  850 , fixed pt res:  tensor(0.0970) , ls:  1 , rel_err:  tensor(5.9144) , time:  0.002660036087036133
iter:  900 , fixed pt res:  tensor(0.0996) , ls:  1 , rel_err:  tensor(5.6303) , time:  0.002655029296875
iter:  950 , fixed pt res:  tensor(0.0964) , ls:  1 , rel_err:  tensor(5.3909) , time:  0.0026865005493164062
iter:  1000 , fixed pt res:  tensor(0.0943) , ls:  1 , rel_err:  tensor(5.1599) , time:  0.002641916275024414
iter:  1050 , fixed pt res:  tensor(0.0981) , ls:  1 , rel_err:  tensor(4.9327) , time:  0.002622365951538086
iter:  1100 , fixed pt res:  tensor(0.0976) , ls:  1 , rel_err:  tensor(4.7290) , time:  0.002641916275024414
iter:  1150 , fixed pt res:  tensor(0.1019) , ls:  1 , rel_err:  tensor(4.5391) , time:  0.00260162353515625
iter:  1200 , fixed pt res:  tensor(0.0966) , ls:  1 , rel_err:  tensor(4.3715) , time:  0.002603292465209961
iter:  1250 , fixed pt res:  tensor(0.0969) , ls:  1 , rel_err:  tensor(4.1634) , time:  0.002655506134033203
iter:  1300 , fixed pt res:  tensor(0.0937) , ls:  1 , rel_err:  tensor(4.0231) , time:  0.00273895263671875
iter:  1350 , fixed pt res:  tensor(0.1010) , ls:  1 , rel_err:  tensor(3.8837) , time:  0.002590179443359375
iter:  1400 , fixed pt res:  tensor(0.0976) , ls:  1 , rel_err:  tensor(3.7580) , time:  0.0026481151580810547
iter:  1450 , fixed pt res:  tensor(0.0942) , ls:  1 , rel_err:  tensor(3.6261) , time:  0.0025975704193115234
iter:  1500 , fixed pt res:  tensor(0.0992) , ls:  1 , rel_err:  tensor(3.5287) , time:  0.0029494762420654297
iter:  1550 , fixed pt res:  tensor(0.0973) , ls:  1 , rel_err:  tensor(3.4125) , time:  0.002590179443359375
iter:  1600 , fixed pt res:  tensor(0.0981) , ls:  1 , rel_err:  tensor(3.2955) , time:  0.002639293670654297
iter:  1650 , fixed pt res:  tensor(0.1037) , ls:  1 , rel_err:  tensor(3.1531) , time:  0.002641439437866211
iter:  1700 , fixed pt res:  tensor(0.0973) , ls:  1 , rel_err:  tensor(3.0603) , time:  0.002615690231323242
iter:  1750 , fixed pt res:  tensor(0.0964) , ls:  1 , rel_err:  tensor(2.9629) , time:  0.002638578414916992
iter:  1800 , fixed pt res:  tensor(0.1004) , ls:  1 , rel_err:  tensor(2.8633) , time:  0.002737760543823242
iter:  1850 , fixed pt res:  tensor(0.1000) , ls:  1 , rel_err:  tensor(2.7754) , time:  0.002644777297973633
iter:  1900 , fixed pt res:  tensor(0.0986) , ls:  1 , rel_err:  tensor(2.7222) , time:  0.0027008056640625
iter:  1950 , fixed pt res:  tensor(0.0954) , ls:  1 , rel_err:  tensor(2.6393) , time:  0.002593517303466797
iter:  2000 , fixed pt res:  tensor(0.0967) , ls:  1 , rel_err:  tensor(2.5414) , time:  0.002609729766845703
iter:  2050 , fixed pt res:  tensor(0.0985) , ls:  1 , rel_err:  tensor(2.4807) , time:  0.0025920867919921875
iter:  2100 , fixed pt res:  tensor(0.1011) , ls:  1 , rel_err:  tensor(2.4069) , time:  0.002677440643310547
iter:  2150 , fixed pt res:  tensor(0.1029) , ls:  1 , rel_err:  tensor(2.3481) , time:  0.002691984176635742
iter:  2200 , fixed pt res:  tensor(0.0995) , ls:  1 , rel_err:  tensor(2.3009) , time:  0.002658843994140625
iter:  2250 , fixed pt res:  tensor(0.0974) , ls:  1 , rel_err:  tensor(2.2204) , time:  0.002644777297973633
iter:  2300 , fixed pt res:  tensor(0.0958) , ls:  1 , rel_err:  tensor(2.1699) , time:  0.0026183128356933594
iter:  2350 , fixed pt res:  tensor(0.0962) , ls:  1 , rel_err:  tensor(2.1140) , time:  0.0026688575744628906
iter:  2400 , fixed pt res:  tensor(0.0949) , ls:  1 , rel_err:  tensor(2.0510) , time:  0.002635955810546875
iter:  2450 , fixed pt res:  tensor(0.1010) , ls:  1 , rel_err:  tensor(1.9629) , time:  0.002660036087036133
iter:  2500 , fixed pt res:  tensor(0.0966) , ls:  1 , rel_err:  tensor(1.9201) , time:  0.0028488636016845703
iter:  2550 , fixed pt res:  tensor(0.0983) , ls:  1 , rel_err:  tensor(1.8688) , time:  0.0026946067810058594
iter:  2600 , fixed pt res:  tensor(0.0928) , ls:  1 , rel_err:  tensor(1.8419) , time:  0.0028378963470458984
iter:  2650 , fixed pt res:  tensor(0.1003) , ls:  1 , rel_err:  tensor(1.7858) , time:  0.0026476383209228516
iter:  2700 , fixed pt res:  tensor(0.0988) , ls:  1 , rel_err:  tensor(1.7670) , time:  0.0026540756225585938
iter:  2750 , fixed pt res:  tensor(0.1000) , ls:  1 , rel_err:  tensor(1.7203) , time:  0.002648591995239258
iter:  2800 , fixed pt res:  tensor(0.0993) , ls:  1 , rel_err:  tensor(1.6839) , time:  0.002696990966796875
iter:  2850 , fixed pt res:  tensor(0.0967) , ls:  1 , rel_err:  tensor(1.6716) , time:  0.0027446746826171875
iter:  2900 , fixed pt res:  tensor(0.0982) , ls:  1 , rel_err:  tensor(1.6258) , time:  0.002668619155883789
iter:  2950 , fixed pt res:  tensor(0.0991) , ls:  1 , rel_err:  tensor(1.5811) , time:  0.0026657581329345703
iter:  3000 , fixed pt res:  tensor(0.1003) , ls:  1 , rel_err:  tensor(1.5758) , time:  0.002733945846557617
iter:  3050 , fixed pt res:  tensor(0.0959) , ls:  1 , rel_err:  tensor(1.5285) , time:  0.0026404857635498047
iter:  3100 , fixed pt res:  tensor(0.0985) , ls:  1 , rel_err:  tensor(1.4862) , time:  0.0026612281799316406
iter:  3150 , fixed pt res:  tensor(0.0980) , ls:  1 , rel_err:  tensor(1.4581) , time:  0.002658843994140625
iter:  3200 , fixed pt res:  tensor(0.1008) , ls:  1 , rel_err:  tensor(1.4271) , time:  0.002880096435546875
iter:  3250 , fixed pt res:  tensor(0.0999) , ls:  1 , rel_err:  tensor(1.4019) , time:  0.002668619155883789
iter:  3300 , fixed pt res:  tensor(0.1013) , ls:  1 , rel_err:  tensor(1.3836) , time:  0.0028009414672851562
iter:  3350 , fixed pt res:  tensor(0.0952) , ls:  1 , rel_err:  tensor(1.3714) , time:  0.0027768611907958984
iter:  3400 , fixed pt res:  tensor(0.0960) , ls:  1 , rel_err:  tensor(1.3542) , time:  0.002624988555908203
iter:  3450 , fixed pt res:  tensor(0.0976) , ls:  1 , rel_err:  tensor(1.3573) , time:  0.002659320831298828
iter:  3500 , fixed pt res:  tensor(0.0967) , ls:  1 , rel_err:  tensor(1.3376) , time:  0.002629518508911133
iter:  3550 , fixed pt res:  tensor(0.0979) , ls:  1 , rel_err:  tensor(1.3192) , time:  0.0026197433471679688
iter:  3600 , fixed pt res:  tensor(0.1008) , ls:  1 , rel_err:  tensor(1.3276) , time:  0.0026237964630126953
iter:  3650 , fixed pt res:  tensor(0.0979) , ls:  1 , rel_err:  tensor(1.3151) , time:  0.002643108367919922
iter:  3700 , fixed pt res:  tensor(0.0990) , ls:  1 , rel_err:  tensor(1.2869) , time:  0.0025787353515625
iter:  3750 , fixed pt res:  tensor(0.0993) , ls:  1 , rel_err:  tensor(1.2659) , time:  0.0026044845581054688
iter:  3800 , fixed pt res:  tensor(0.0971) , ls:  1 , rel_err:  tensor(1.2559) , time:  0.0027234554290771484
iter:  3850 , fixed pt res:  tensor(0.0954) , ls:  1 , rel_err:  tensor(1.2633) , time:  0.0026369094848632812
iter:  3900 , fixed pt res:  tensor(0.0973) , ls:  1 , rel_err:  tensor(1.2327) , time:  0.0026540756225585938
iter:  3950 , fixed pt res:  tensor(0.0986) , ls:  1 , rel_err:  tensor(1.1850) , time:  0.002602815628051758
iter:  4000 , fixed pt res:  tensor(0.0990) , ls:  1 , rel_err:  tensor(1.1821) , time:  0.0025894641876220703
iter:  4050 , fixed pt res:  tensor(0.0981) , ls:  1 , rel_err:  tensor(1.1858) , time:  0.002626657485961914
iter:  4100 , fixed pt res:  tensor(0.0947) , ls:  1 , rel_err:  tensor(1.1439) , time:  0.002600431442260742
iter:  4150 , fixed pt res:  tensor(0.0993) , ls:  1 , rel_err:  tensor(1.1387) , time:  0.0026493072509765625
iter:  4200 , fixed pt res:  tensor(0.0996) , ls:  1 , rel_err:  tensor(1.1192) , time:  0.0028383731842041016
iter:  4250 , fixed pt res:  tensor(0.0949) , ls:  1 , rel_err:  tensor(1.0881) , time:  0.0027513504028320312
iter:  4300 , fixed pt res:  tensor(0.0958) , ls:  1 , rel_err:  tensor(1.0803) , time:  0.0026311874389648438
iter:  4350 , fixed pt res:  tensor(0.0997) , ls:  1 , rel_err:  tensor(1.0660) , time:  0.0026144981384277344
iter:  4400 , fixed pt res:  tensor(0.0955) , ls:  1 , rel_err:  tensor(1.0327) , time:  0.0026459693908691406
iter:  4450 , fixed pt res:  tensor(0.0964) , ls:  1 , rel_err:  tensor(1.0417) , time:  0.002648591995239258
iter:  4500 , fixed pt res:  tensor(0.0976) , ls:  1 , rel_err:  tensor(1.0445) , time:  0.002627849578857422
iter:  4550 , fixed pt res:  tensor(0.0935) , ls:  1 , rel_err:  tensor(1.0299) , time:  0.002614736557006836
iter:  4600 , fixed pt res:  tensor(0.0952) , ls:  1 , rel_err:  tensor(1.0165) , time:  0.0026810169219970703
iter:  4650 , fixed pt res:  tensor(0.0977) , ls:  1 , rel_err:  tensor(1.0161) , time:  0.0026397705078125
iter:  4700 , fixed pt res:  tensor(0.0997) , ls:  1 , rel_err:  tensor(1.0329) , time:  0.0026242733001708984
iter:  4750 , fixed pt res:  tensor(0.0987) , ls:  1 , rel_err:  tensor(1.0263) , time:  0.002673625946044922
iter:  4800 , fixed pt res:  tensor(0.0957) , ls:  1 , rel_err:  tensor(1.0179) , time:  0.0026826858520507812
iter:  4850 , fixed pt res:  tensor(0.0989) , ls:  1 , rel_err:  tensor(1.0152) , time:  0.002739429473876953
iter:  4900 , fixed pt res:  tensor(0.0958) , ls:  1 , rel_err:  tensor(1.0190) , time:  0.0026226043701171875
iter:  4950 , fixed pt res:  tensor(0.1008) , ls:  1 , rel_err:  tensor(1.0083) , time:  0.0027060508728027344
iter:  5000 , fixed pt res:  tensor(0.0958) , ls:  1 , rel_err:  tensor(0.9981) , time:  0.0026297569274902344
iter:  5050 , fixed pt res:  tensor(0.0980) , ls:  1 , rel_err:  tensor(0.9669) , time:  0.0026192665100097656
iter:  5100 , fixed pt res:  tensor(0.0957) , ls:  1 , rel_err:  tensor(0.9647) , time:  0.002637624740600586
iter:  5150 , fixed pt res:  tensor(0.0973) , ls:  1 , rel_err:  tensor(0.9442) , time:  0.002691030502319336
iter:  5200 , fixed pt res:  tensor(0.0995) , ls:  1 , rel_err:  tensor(0.9389) , time:  0.002645730972290039
iter:  5250 , fixed pt res:  tensor(0.1021) , ls:  1 , rel_err:  tensor(0.9361) , time:  0.0026426315307617188
iter:  5300 , fixed pt res:  tensor(0.0970) , ls:  1 , rel_err:  tensor(0.9394) , time:  0.0026667118072509766
iter:  5350 , fixed pt res:  tensor(0.0947) , ls:  1 , rel_err:  tensor(0.9366) , time:  0.0026564598083496094
iter:  5400 , fixed pt res:  tensor(0.0977) , ls:  1 , rel_err:  tensor(0.9341) , time:  0.002665281295776367
iter:  5450 , fixed pt res:  tensor(0.0952) , ls:  1 , rel_err:  tensor(0.9331) , time:  0.002624988555908203
iter:  5500 , fixed pt res:  tensor(0.0955) , ls:  1 , rel_err:  tensor(0.9150) , time:  0.0027475357055664062
iter:  5550 , fixed pt res:  tensor(0.1011) , ls:  1 , rel_err:  tensor(0.9156) , time:  0.0026793479919433594
iter:  5600 , fixed pt res:  tensor(0.0967) , ls:  1 , rel_err:  tensor(0.9214) , time:  0.003243684768676758
iter:  5650 , fixed pt res:  tensor(0.0969) , ls:  1 , rel_err:  tensor(0.9045) , time:  0.0027420520782470703
iter:  5700 , fixed pt res:  tensor(0.0970) , ls:  1 , rel_err:  tensor(0.9005) , time:  0.0028600692749023438
iter:  5750 , fixed pt res:  tensor(0.0975) , ls:  1 , rel_err:  tensor(0.8744) , time:  0.0026273727416992188
iter:  5800 , fixed pt res:  tensor(0.1004) , ls:  1 , rel_err:  tensor(0.8787) , time:  0.0026781558990478516
iter:  5850 , fixed pt res:  tensor(0.0981) , ls:  1 , rel_err:  tensor(0.8809) , time:  0.0026412010192871094
iter:  5900 , fixed pt res:  tensor(0.1017) , ls:  1 , rel_err:  tensor(0.8876) , time:  0.0027167797088623047
iter:  5950 , fixed pt res:  tensor(0.0959) , ls:  1 , rel_err:  tensor(0.9126) , time:  0.0028395652770996094
iter:  6000 , fixed pt res:  tensor(0.0977) , ls:  1 , rel_err:  tensor(0.9226) , time:  0.0026731491088867188
iter:  6050 , fixed pt res:  tensor(0.1001) , ls:  1 , rel_err:  tensor(0.9003) , time:  0.002645730972290039
iter:  6100 , fixed pt res:  tensor(0.0969) , ls:  1 , rel_err:  tensor(0.8834) , time:  0.0027151107788085938
iter:  6150 , fixed pt res:  tensor(0.0941) , ls:  1 , rel_err:  tensor(0.8768) , time:  0.002652406692504883
iter:  6200 , fixed pt res:  tensor(0.0940) , ls:  1 , rel_err:  tensor(0.8559) , time:  0.002678394317626953
iter:  6250 , fixed pt res:  tensor(0.1001) , ls:  1 , rel_err:  tensor(0.8983) , time:  0.002643585205078125
iter:  6300 , fixed pt res:  tensor(0.0964) , ls:  1 , rel_err:  tensor(0.8749) , time:  0.0027146339416503906
iter:  6350 , fixed pt res:  tensor(0.1014) , ls:  1 , rel_err:  tensor(0.8756) , time:  0.0026204586029052734
iter:  6400 , fixed pt res:  tensor(0.0987) , ls:  1 , rel_err:  tensor(0.8709) , time:  0.003339052200317383
iter:  6450 , fixed pt res:  tensor(0.0978) , ls:  1 , rel_err:  tensor(0.8542) , time:  0.002626180648803711
iter:  6500 , fixed pt res:  tensor(0.0923) , ls:  1 , rel_err:  tensor(0.8736) , time:  0.0028307437896728516
iter:  6550 , fixed pt res:  tensor(0.1011) , ls:  1 , rel_err:  tensor(0.8672) , time:  0.00263214111328125
iter:  6600 , fixed pt res:  tensor(0.1008) , ls:  1 , rel_err:  tensor(0.8565) , time:  0.002622842788696289
iter:  6650 , fixed pt res:  tensor(0.0956) , ls:  1 , rel_err:  tensor(0.8584) , time:  0.002637147903442383
iter:  6700 , fixed pt res:  tensor(0.1013) , ls:  1 , rel_err:  tensor(0.8428) , time:  0.0026862621307373047
iter:  6750 , fixed pt res:  tensor(0.0975) , ls:  1 , rel_err:  tensor(0.8281) , time:  0.0026366710662841797
iter:  6800 , fixed pt res:  tensor(0.1006) , ls:  1 , rel_err:  tensor(0.8326) , time:  0.002646207809448242
iter:  6850 , fixed pt res:  tensor(0.0989) , ls:  1 , rel_err:  tensor(0.8473) , time:  0.002773284912109375
iter:  6900 , fixed pt res:  tensor(0.0977) , ls:  1 , rel_err:  tensor(0.8323) , time:  0.002663850784301758
iter:  6950 , fixed pt res:  tensor(0.0964) , ls:  1 , rel_err:  tensor(0.8157) , time:  0.0027344226837158203
iter:  7000 , fixed pt res:  tensor(0.0948) , ls:  1 , rel_err:  tensor(0.7917) , time:  0.0026133060455322266
iter:  7050 , fixed pt res:  tensor(0.0980) , ls:  1 , rel_err:  tensor(0.7897) , time:  0.002618551254272461
iter:  7100 , fixed pt res:  tensor(0.0989) , ls:  1 , rel_err:  tensor(0.8002) , time:  0.0026454925537109375
iter:  7150 , fixed pt res:  tensor(0.0995) , ls:  1 , rel_err:  tensor(0.7800) , time:  0.0026922225952148438
iter:  7200 , fixed pt res:  tensor(0.0969) , ls:  1 , rel_err:  tensor(0.7724) , time:  0.002665281295776367
iter:  7250 , fixed pt res:  tensor(0.0959) , ls:  1 , rel_err:  tensor(0.7787) , time:  0.0026602745056152344
iter:  7300 , fixed pt res:  tensor(0.0984) , ls:  1 , rel_err:  tensor(0.7685) , time:  0.002656698226928711
iter:  7350 , fixed pt res:  tensor(0.0981) , ls:  1 , rel_err:  tensor(0.7541) , time:  0.0026764869689941406
iter:  7400 , fixed pt res:  tensor(0.0967) , ls:  1 , rel_err:  tensor(0.7528) , time:  0.0027091503143310547
iter:  7450 , fixed pt res:  tensor(0.0977) , ls:  1 , rel_err:  tensor(0.7571) , time:  0.0026578903198242188
iter:  7500 , fixed pt res:  tensor(0.1019) , ls:  1 , rel_err:  tensor(0.7803) , time:  0.002628326416015625
iter:  7550 , fixed pt res:  tensor(0.0989) , ls:  1 , rel_err:  tensor(0.7855) , time:  0.002618551254272461
iter:  7600 , fixed pt res:  tensor(0.0959) , ls:  1 , rel_err:  tensor(0.7927) , time:  0.0026383399963378906
iter:  7650 , fixed pt res:  tensor(0.0991) , ls:  1 , rel_err:  tensor(0.8163) , time:  0.0029609203338623047
iter:  7700 , fixed pt res:  tensor(0.0986) , ls:  1 , rel_err:  tensor(0.8069) , time:  0.0026159286499023438
iter:  7750 , fixed pt res:  tensor(0.0966) , ls:  1 , rel_err:  tensor(0.8106) , time:  0.002648591995239258
iter:  7800 , fixed pt res:  tensor(0.0994) , ls:  1 , rel_err:  tensor(0.8216) , time:  0.002741098403930664
iter:  7850 , fixed pt res:  tensor(0.0960) , ls:  1 , rel_err:  tensor(0.8073) , time:  0.0026519298553466797
iter:  7900 , fixed pt res:  tensor(0.0990) , ls:  1 , rel_err:  tensor(0.7909) , time:  0.0027015209197998047
iter:  7950 , fixed pt res:  tensor(0.0944) , ls:  1 , rel_err:  tensor(0.7850) , time:  0.0027010440826416016
iter:  8000 , fixed pt res:  tensor(0.0987) , ls:  1 , rel_err:  tensor(0.7983) , time:  0.0026674270629882812
iter:  8050 , fixed pt res:  tensor(0.0946) , ls:  1 , rel_err:  tensor(0.8124) , time:  0.002661466598510742
iter:  8100 , fixed pt res:  tensor(0.0938) , ls:  1 , rel_err:  tensor(0.8342) , time:  0.0027573108673095703
iter:  8150 , fixed pt res:  tensor(0.0981) , ls:  1 , rel_err:  tensor(0.8350) , time:  0.0026941299438476562
iter:  8200 , fixed pt res:  tensor(0.0981) , ls:  1 , rel_err:  tensor(0.8280) , time:  0.0026400089263916016
iter:  8250 , fixed pt res:  tensor(0.0975) , ls:  1 , rel_err:  tensor(0.8259) , time:  0.0026302337646484375
iter:  8300 , fixed pt res:  tensor(0.1005) , ls:  1 , rel_err:  tensor(0.8353) , time:  0.0026967525482177734
iter:  8350 , fixed pt res:  tensor(0.0974) , ls:  1 , rel_err:  tensor(0.8341) , time:  0.0026950836181640625
iter:  8400 , fixed pt res:  tensor(0.0965) , ls:  1 , rel_err:  tensor(0.8350) , time:  0.00262451171875
iter:  8450 , fixed pt res:  tensor(0.0988) , ls:  1 , rel_err:  tensor(0.8283) , time:  0.002725839614868164
iter:  8500 , fixed pt res:  tensor(0.0975) , ls:  1 , rel_err:  tensor(0.8191) , time:  0.002711772918701172
iter:  8550 , fixed pt res:  tensor(0.0952) , ls:  1 , rel_err:  tensor(0.7979) , time:  0.002776622772216797
iter:  8600 , fixed pt res:  tensor(0.0939) , ls:  1 , rel_err:  tensor(0.8140) , time:  0.002679109573364258
iter:  8650 , fixed pt res:  tensor(0.0999) , ls:  1 , rel_err:  tensor(0.8250) , time:  0.0026099681854248047
iter:  8700 , fixed pt res:  tensor(0.0997) , ls:  1 , rel_err:  tensor(0.8035) , time:  0.002637147903442383
iter:  8750 , fixed pt res:  tensor(0.0993) , ls:  1 , rel_err:  tensor(0.7881) , time:  0.0026390552520751953
iter:  8800 , fixed pt res:  tensor(0.0932) , ls:  1 , rel_err:  tensor(0.8042) , time:  0.0029790401458740234
iter:  8850 , fixed pt res:  tensor(0.1018) , ls:  1 , rel_err:  tensor(0.7971) , time:  0.0026328563690185547
iter:  8900 , fixed pt res:  tensor(0.0981) , ls:  1 , rel_err:  tensor(0.7499) , time:  0.002802133560180664
iter:  8950 , fixed pt res:  tensor(0.0980) , ls:  1 , rel_err:  tensor(0.7480) , time:  0.0026078224182128906
iter:  9000 , fixed pt res:  tensor(0.0976) , ls:  1 , rel_err:  tensor(0.7709) , time:  0.002763032913208008
iter:  9050 , fixed pt res:  tensor(0.1001) , ls:  1 , rel_err:  tensor(0.7505) , time:  0.002651214599609375
iter:  9100 , fixed pt res:  tensor(0.0995) , ls:  1 , rel_err:  tensor(0.7414) , time:  0.0027337074279785156
iter:  9150 , fixed pt res:  tensor(0.1007) , ls:  1 , rel_err:  tensor(0.7507) , time:  0.0026307106018066406
iter:  9200 , fixed pt res:  tensor(0.0970) , ls:  1 , rel_err:  tensor(0.7687) , time:  0.0026504993438720703
iter:  9250 , fixed pt res:  tensor(0.0963) , ls:  1 , rel_err:  tensor(0.7512) , time:  0.0026214122772216797
iter:  9300 , fixed pt res:  tensor(0.1027) , ls:  1 , rel_err:  tensor(0.7331) , time:  0.0026297569274902344
iter:  9350 , fixed pt res:  tensor(0.1021) , ls:  1 , rel_err:  tensor(0.7549) , time:  0.0026335716247558594
iter:  9400 , fixed pt res:  tensor(0.0972) , ls:  1 , rel_err:  tensor(0.7413) , time:  0.0026237964630126953
iter:  9450 , fixed pt res:  tensor(0.1009) , ls:  1 , rel_err:  tensor(0.7578) , time:  0.0026702880859375
iter:  9500 , fixed pt res:  tensor(0.0990) , ls:  1 , rel_err:  tensor(0.7343) , time:  0.0026204586029052734
iter:  9550 , fixed pt res:  tensor(0.0995) , ls:  1 , rel_err:  tensor(0.7433) , time:  0.0026760101318359375
iter:  9600 , fixed pt res:  tensor(0.0988) , ls:  1 , rel_err:  tensor(0.7255) , time:  0.002622842788696289
iter:  9650 , fixed pt res:  tensor(0.0958) , ls:  1 , rel_err:  tensor(0.7154) , time:  0.002593994140625
iter:  9700 , fixed pt res:  tensor(0.1000) , ls:  1 , rel_err:  tensor(0.7179) , time:  0.0026009082794189453
iter:  9750 , fixed pt res:  tensor(0.0998) , ls:  1 , rel_err:  tensor(0.7104) , time:  0.0026895999908447266
iter:  9800 , fixed pt res:  tensor(0.0993) , ls:  1 , rel_err:  tensor(0.7094) , time:  0.002779722213745117
iter:  9850 , fixed pt res:  tensor(0.0990) , ls:  1 , rel_err:  tensor(0.6891) , time:  0.0026383399963378906
iter:  9900 , fixed pt res:  tensor(0.0974) , ls:  1 , rel_err:  tensor(0.7056) , time:  0.002614736557006836
iter:  9950 , fixed pt res:  tensor(0.1008) , ls:  1 , rel_err:  tensor(0.7106) , time:  0.002745389938354492
iter:  10000 , fixed pt res:  tensor(0.0996) , ls:  1 , rel_err:  tensor(0.7004) , time:  0.0027778148651123047
trial  1  finished after 27.69773554801941




 CHANGE, delta =  100
norm(rel_err_hist_temp - rel_err_lmm_hist_delta_array[sample_index,:]) =  tensor(0.)

Plot varying $\delta$ results¶

In [ ]:

Copied!





title_fontsize = 22
fontsize       = 20
fig1 = plt.figure()
my_blue = '#1f77b4'
my_orange = '#F97306'

fig1 = plt.figure()
plt.style.use('seaborn-whitegrid')
ax = plt.axes()

ax.semilogy(rel_err_lmm_hist_delta_array[0,:], linewidth=5, color=my_blue)
ax.semilogy(rel_err_lmm_hist_delta_array[1,:], linewidth=5, color='tab:green')
ax.semilogy(rel_err_lmm_hist_delta_array[2,:], linewidth=5, color='tab:purple')
ax.semilogy(rel_err_lmm_hist_delta_array[3,:], linewidth=5, color='tab:cyan')
ax.semilogy(rel_err_lmm_hist_delta_array[4,:], linewidth=5, color='tab:brown')

# ax.set_xlabel("x-axis", fontsize=title_fontsize)
# ax.legend(['HJ-ADMM $\delta=0.001$', 'HJ-ADMM $\delta=0.01$', 'HJ-ADMM $\delta=0.1$', 'HJ-ADMM $\delta=1$', 'HJ-ADMM $\delta=10$'],fontsize=fontsize, bbox_to_anchor=(1.0, 1.0), loc='upper left')
ax.legend(['HJ-MM $\delta=0.01$', 'HJ-MM $\delta=0.1$', 'HJ-MM $\delta=1$', 'HJ-MM $\delta=10$', 'HJ-MM $\delta=100$'],fontsize=fontsize, bbox_to_anchor=(1.0, 1.0), loc='upper left')
ax.tick_params(labelsize=fontsize, which='both', direction='in')
save_str = 'ADMM_noisy_delta_comparison.pdf'
fig1.savefig(save_str, dpi=300 , bbox_inches="tight", pad_inches=0.0)

title_str = 'samples =  ' + str(samples_for_varying_delta)
ax.set_title(title_str, fontsize=fontsize)

title_fontsize = 22
fontsize       = 20
fig1 = plt.figure()
my_blue = '#1f77b4'
my_orange = '#F97306'

fig1 = plt.figure()
plt.style.use('seaborn-whitegrid')
ax = plt.axes()

ax.semilogy(rel_err_lmm_hist_delta_array[0,:], linewidth=5, color=my_blue)
ax.semilogy(rel_err_lmm_hist_delta_array[1,:], linewidth=5, color='tab:green')
ax.semilogy(rel_err_lmm_hist_delta_array[2,:], linewidth=5, color='tab:purple')
ax.semilogy(rel_err_lmm_hist_delta_array[3,:], linewidth=5, color='tab:cyan')
ax.semilogy(rel_err_lmm_hist_delta_array[4,:], linewidth=5, color='tab:brown')

# ax.set_xlabel("x-axis", fontsize=title_fontsize)
# ax.legend(['HJ-ADMM $\delta=0.001$', 'HJ-ADMM $\delta=0.01$', 'HJ-ADMM $\delta=0.1$', 'HJ-ADMM $\delta=1$', 'HJ-ADMM $\delta=10$'],fontsize=fontsize, bbox_to_anchor=(1.0, 1.0), loc='upper left')
ax.legend(['HJ-MM $\delta=0.01$', 'HJ-MM $\delta=0.1$', 'HJ-MM $\delta=1$', 'HJ-MM $\delta=10$', 'HJ-MM $\delta=100$'],fontsize=fontsize, bbox_to_anchor=(1.0, 1.0), loc='upper left')
ax.tick_params(labelsize=fontsize, which='both', direction='in')
save_str = 'ADMM_noisy_delta_comparison.pdf'
fig1.savefig(save_str, dpi=300 , bbox_inches="tight", pad_inches=0.0)

title_str = 'samples =  ' + str(samples_for_varying_delta)
ax.set_title(title_str, fontsize=fontsize)

<ipython-input-8-3304b89c7bc1>:8: MatplotlibDeprecationWarning: The seaborn styles shipped by Matplotlib are deprecated since 3.6, as they no longer correspond to the styles shipped by seaborn. However, they will remain available as 'seaborn-v0_8-<style>'. Alternatively, directly use the seaborn API instead.
  plt.style.use('seaborn-whitegrid')

Out[ ]:

Text(0.5, 1.0, 'samples =  10000')

<Figure size 640x480 with 0 Axes>

No description has been provided for this image

Blackbox Oracle HJ Prox Experiment 2: Fix best $\delta$ and vary number of samples¶

In [ ]:

Copied!





sample_array = [int(1), int(1e2), int(1e3), int(1e4)]

xopt_HJ_lmm_sample_array = torch.zeros(len(sample_array), x_true.shape[0], x_true.shape[1])
fk_HJ_lmm_hist_sample_array = torch.zeros(len(sample_array), max_iters)
norm_diff_HJ_lmm_hist_sample_array = torch.zeros(len(sample_array), max_iters)
rel_err_lmm_hist_sample_array = torch.zeros(len(sample_array), max_iters)
ls_iters_lmm_hist_sample_array = torch.zeros(len(sample_array), max_iters)


delta_for_varying_samples = 10
alpha_for_varying_samples = alpha_array[3]

for sample_index in range(len(sample_array)):
  print('Number of Samples =', sample_array[sample_index])
  for i in range(n_trials):
    start_time_hj_lmm = time.time()
    current_samples = sample_array[sample_index]
    xopt_HJ_lmm_temp, fk_HJ_lmm_hist_temp, norm_diff_HJ_lmm_hist_temp, rel_err_hist_temp, ls_iters_hist_temp = hj_mm(x0, W, A, b, max_iters, 
                                                                                                  int_samples=current_samples, 
                                                                                                  delta=delta_for_varying_samples, 
                                                                                                  step_size=step_size,
                                                                                                  lambd = lambd, 
                                                                                                  verbose=True, 
                                                                                                  print_freq=50,
                                                                                                  x_true=x_true,
                                                                                                  alpha=alpha_for_varying_samples)
    end_time_hj_lmm = time.time()
    time_hj_lmm = end_time_hj_lmm - start_time_hj_lmm
    print('trial ', i+1, ' finished after', time_hj_lmm)
      
    if i==0:
      xopt_HJ_lmm = xopt_HJ_lmm_temp
      fk_HJ_lmm_hist = fk_HJ_lmm_hist_temp
      norm_diff_HJ_lmm_hist = norm_diff_HJ_lmm_hist_temp
      rel_err_hist = rel_err_hist_temp
      ls_iters_hist = ls_iters_hist_temp
    else:
      xopt_HJ_lmm = xopt_HJ_lmm + xopt_HJ_lmm_temp
      fk_HJ_lmm_hist = fk_HJ_lmm_hist + fk_HJ_lmm_hist_temp
      norm_diff_HJ_lmm_hist = norm_diff_HJ_lmm_hist + norm_diff_HJ_lmm_hist_temp
      rel_err_hist = rel_err_hist + rel_err_hist_temp
      ls_iters_hist = ls_iters_hist + ls_iters_hist_temp


  xopt_HJ_lmm_sample_array[sample_index, :, :] = xopt_HJ_lmm/n_trials
  fk_HJ_lmm_hist_sample_array[sample_index, :] = fk_HJ_lmm_hist/n_trials
  norm_diff_HJ_lmm_hist_sample_array[sample_index, :] = norm_diff_HJ_lmm_hist/n_trials
  rel_err_lmm_hist_sample_array[sample_index, :] = rel_err_hist/n_trials 
  ls_iters_lmm_hist_sample_array[sample_index,:] = ls_iters_hist/n_trials
  print('norm(rel_err_hist_temp - rel_err_lmm_hist_sample_array[sample_index,:]) = ', torch.norm(rel_err_hist_temp - rel_err_lmm_hist_sample_array[sample_index,:]))
sample_array = [int(1), int(1e2), int(1e3), int(1e4)]

xopt_HJ_lmm_sample_array = torch.zeros(len(sample_array), x_true.shape[0], x_true.shape[1])
fk_HJ_lmm_hist_sample_array = torch.zeros(len(sample_array), max_iters)
norm_diff_HJ_lmm_hist_sample_array = torch.zeros(len(sample_array), max_iters)
rel_err_lmm_hist_sample_array = torch.zeros(len(sample_array), max_iters)
ls_iters_lmm_hist_sample_array = torch.zeros(len(sample_array), max_iters)


delta_for_varying_samples = 10
alpha_for_varying_samples = alpha_array[3]

for sample_index in range(len(sample_array)):
  print('Number of Samples =', sample_array[sample_index])
  for i in range(n_trials):
    start_time_hj_lmm = time.time()
    current_samples = sample_array[sample_index]
    xopt_HJ_lmm_temp, fk_HJ_lmm_hist_temp, norm_diff_HJ_lmm_hist_temp, rel_err_hist_temp, ls_iters_hist_temp = hj_mm(x0, W, A, b, max_iters, 
                                                                                                  int_samples=current_samples, 
                                                                                                  delta=delta_for_varying_samples, 
                                                                                                  step_size=step_size,
                                                                                                  lambd = lambd, 
                                                                                                  verbose=True, 
                                                                                                  print_freq=50,
                                                                                                  x_true=x_true,
                                                                                                  alpha=alpha_for_varying_samples)
    end_time_hj_lmm = time.time()
    time_hj_lmm = end_time_hj_lmm - start_time_hj_lmm
    print('trial ', i+1, ' finished after', time_hj_lmm)
      
    if i==0:
      xopt_HJ_lmm = xopt_HJ_lmm_temp
      fk_HJ_lmm_hist = fk_HJ_lmm_hist_temp
      norm_diff_HJ_lmm_hist = norm_diff_HJ_lmm_hist_temp
      rel_err_hist = rel_err_hist_temp
      ls_iters_hist = ls_iters_hist_temp
    else:
      xopt_HJ_lmm = xopt_HJ_lmm + xopt_HJ_lmm_temp
      fk_HJ_lmm_hist = fk_HJ_lmm_hist + fk_HJ_lmm_hist_temp
      norm_diff_HJ_lmm_hist = norm_diff_HJ_lmm_hist + norm_diff_HJ_lmm_hist_temp
      rel_err_hist = rel_err_hist + rel_err_hist_temp
      ls_iters_hist = ls_iters_hist + ls_iters_hist_temp


  xopt_HJ_lmm_sample_array[sample_index, :, :] = xopt_HJ_lmm/n_trials
  fk_HJ_lmm_hist_sample_array[sample_index, :] = fk_HJ_lmm_hist/n_trials
  norm_diff_HJ_lmm_hist_sample_array[sample_index, :] = norm_diff_HJ_lmm_hist/n_trials
  rel_err_lmm_hist_sample_array[sample_index, :] = rel_err_hist/n_trials 
  ls_iters_lmm_hist_sample_array[sample_index,:] = ls_iters_hist/n_trials
  print('norm(rel_err_hist_temp - rel_err_lmm_hist_sample_array[sample_index,:]) = ', torch.norm(rel_err_hist_temp - rel_err_lmm_hist_sample_array[sample_index,:]))

Number of Samples = 1
iter:  50 , fixed pt res:  tensor(4.3492) , ls:  1 , rel_err:  tensor(22.0593) , time:  0.0012357234954833984
iter:  100 , fixed pt res:  tensor(4.2398) , ls:  1 , rel_err:  tensor(26.0788) , time:  0.0013756752014160156
iter:  150 , fixed pt res:  tensor(4.3122) , ls:  1 , rel_err:  tensor(29.7923) , time:  0.0012924671173095703
iter:  200 , fixed pt res:  tensor(4.2578) , ls:  1 , rel_err:  tensor(32.9060) , time:  0.001222848892211914
iter:  250 , fixed pt res:  tensor(4.2742) , ls:  1 , rel_err:  tensor(36.0638) , time:  0.001218557357788086
iter:  300 , fixed pt res:  tensor(4.2656) , ls:  1 , rel_err:  tensor(38.0925) , time:  0.001238107681274414
iter:  350 , fixed pt res:  tensor(4.3729) , ls:  1 , rel_err:  tensor(40.2829) , time:  0.0012290477752685547
iter:  400 , fixed pt res:  tensor(4.4421) , ls:  1 , rel_err:  tensor(42.0254) , time:  0.001237630844116211
iter:  450 , fixed pt res:  tensor(4.4611) , ls:  1 , rel_err:  tensor(43.8565) , time:  0.002003908157348633
iter:  500 , fixed pt res:  tensor(4.3654) , ls:  1 , rel_err:  tensor(46.0011) , time:  0.0012235641479492188
iter:  550 , fixed pt res:  tensor(4.4026) , ls:  1 , rel_err:  tensor(48.4161) , time:  0.0012052059173583984
iter:  600 , fixed pt res:  tensor(4.3892) , ls:  1 , rel_err:  tensor(50.3180) , time:  0.0013031959533691406
iter:  650 , fixed pt res:  tensor(4.4085) , ls:  1 , rel_err:  tensor(51.9762) , time:  0.0017383098602294922
iter:  700 , fixed pt res:  tensor(4.2235) , ls:  1 , rel_err:  tensor(53.8161) , time:  0.001260519027709961
iter:  750 , fixed pt res:  tensor(4.3299) , ls:  1 , rel_err:  tensor(55.5659) , time:  0.0012469291687011719
iter:  800 , fixed pt res:  tensor(4.3728) , ls:  1 , rel_err:  tensor(58.0084) , time:  0.0012285709381103516
iter:  850 , fixed pt res:  tensor(4.5146) , ls:  1 , rel_err:  tensor(59.5839) , time:  0.001222372055053711
iter:  900 , fixed pt res:  tensor(4.4009) , ls:  1 , rel_err:  tensor(61.2522) , time:  0.0012252330780029297
iter:  950 , fixed pt res:  tensor(4.3857) , ls:  1 , rel_err:  tensor(63.1436) , time:  0.0012290477752685547
iter:  1000 , fixed pt res:  tensor(4.3707) , ls:  1 , rel_err:  tensor(64.9713) , time:  0.001203775405883789
iter:  1050 , fixed pt res:  tensor(4.2621) , ls:  1 , rel_err:  tensor(66.0495) , time:  0.0012392997741699219
iter:  1100 , fixed pt res:  tensor(4.5444) , ls:  1 , rel_err:  tensor(67.6864) , time:  0.001298666000366211
iter:  1150 , fixed pt res:  tensor(4.2174) , ls:  1 , rel_err:  tensor(69.0120) , time:  0.0012073516845703125
iter:  1200 , fixed pt res:  tensor(4.3168) , ls:  1 , rel_err:  tensor(70.2632) , time:  0.0012478828430175781
iter:  1250 , fixed pt res:  tensor(4.5151) , ls:  1 , rel_err:  tensor(72.0217) , time:  0.0012123584747314453
iter:  1300 , fixed pt res:  tensor(4.4318) , ls:  1 , rel_err:  tensor(72.8153) , time:  0.0013761520385742188
iter:  1350 , fixed pt res:  tensor(4.4913) , ls:  1 , rel_err:  tensor(75.4307) , time:  0.0012369155883789062
iter:  1400 , fixed pt res:  tensor(4.3189) , ls:  1 , rel_err:  tensor(76.6560) , time:  0.001226186752319336
iter:  1450 , fixed pt res:  tensor(4.3683) , ls:  1 , rel_err:  tensor(76.9220) , time:  0.0021958351135253906
iter:  1500 , fixed pt res:  tensor(4.5245) , ls:  1 , rel_err:  tensor(78.3533) , time:  0.0012638568878173828
iter:  1550 , fixed pt res:  tensor(4.4940) , ls:  1 , rel_err:  tensor(79.2072) , time:  0.001207113265991211
iter:  1600 , fixed pt res:  tensor(4.4487) , ls:  1 , rel_err:  tensor(80.8733) , time:  0.0015423297882080078
iter:  1650 , fixed pt res:  tensor(4.5145) , ls:  1 , rel_err:  tensor(81.4984) , time:  0.0012214183807373047
iter:  1700 , fixed pt res:  tensor(4.3175) , ls:  1 , rel_err:  tensor(83.0772) , time:  0.0011968612670898438
iter:  1750 , fixed pt res:  tensor(4.4626) , ls:  1 , rel_err:  tensor(84.6571) , time:  0.0012164115905761719
iter:  1800 , fixed pt res:  tensor(4.5192) , ls:  1 , rel_err:  tensor(86.1078) , time:  0.001241445541381836
iter:  1850 , fixed pt res:  tensor(4.3068) , ls:  1 , rel_err:  tensor(87.2647) , time:  0.0012059211730957031
iter:  1900 , fixed pt res:  tensor(4.2968) , ls:  1 , rel_err:  tensor(88.8182) , time:  0.0011935234069824219
iter:  1950 , fixed pt res:  tensor(4.2597) , ls:  1 , rel_err:  tensor(90.1199) , time:  0.0012166500091552734
iter:  2000 , fixed pt res:  tensor(4.2799) , ls:  1 , rel_err:  tensor(91.2706) , time:  0.0015642642974853516
iter:  2050 , fixed pt res:  tensor(4.3968) , ls:  1 , rel_err:  tensor(93.1120) , time:  0.001207590103149414
iter:  2100 , fixed pt res:  tensor(4.2294) , ls:  1 , rel_err:  tensor(94.6409) , time:  0.0012409687042236328
iter:  2150 , fixed pt res:  tensor(4.1860) , ls:  1 , rel_err:  tensor(95.0320) , time:  0.0012357234954833984
iter:  2200 , fixed pt res:  tensor(4.3426) , ls:  1 , rel_err:  tensor(96.5111) , time:  0.0012042522430419922
iter:  2250 , fixed pt res:  tensor(4.3826) , ls:  1 , rel_err:  tensor(97.7978) , time:  0.001207590103149414
iter:  2300 , fixed pt res:  tensor(4.4400) , ls:  1 , rel_err:  tensor(99.7797) , time:  0.0012328624725341797
iter:  2350 , fixed pt res:  tensor(4.4654) , ls:  1 , rel_err:  tensor(101.4514) , time:  0.0012099742889404297
iter:  2400 , fixed pt res:  tensor(4.3189) , ls:  1 , rel_err:  tensor(102.1042) , time:  0.0012078285217285156
iter:  2450 , fixed pt res:  tensor(4.2428) , ls:  1 , rel_err:  tensor(103.2482) , time:  0.0012216567993164062
iter:  2500 , fixed pt res:  tensor(4.3893) , ls:  1 , rel_err:  tensor(103.7875) , time:  0.00119781494140625
iter:  2550 , fixed pt res:  tensor(4.3765) , ls:  1 , rel_err:  tensor(104.7699) , time:  0.0012710094451904297
iter:  2600 , fixed pt res:  tensor(4.3278) , ls:  1 , rel_err:  tensor(105.2396) , time:  0.001291513442993164
iter:  2650 , fixed pt res:  tensor(4.2415) , ls:  1 , rel_err:  tensor(105.4326) , time:  0.001216888427734375
iter:  2700 , fixed pt res:  tensor(4.1311) , ls:  1 , rel_err:  tensor(106.0025) , time:  0.001210927963256836
iter:  2750 , fixed pt res:  tensor(4.4455) , ls:  1 , rel_err:  tensor(106.8205) , time:  0.0012440681457519531
iter:  2800 , fixed pt res:  tensor(4.3838) , ls:  1 , rel_err:  tensor(108.9093) , time:  0.0015323162078857422
iter:  2850 , fixed pt res:  tensor(4.1581) , ls:  1 , rel_err:  tensor(110.2157) , time:  0.0012116432189941406
iter:  2900 , fixed pt res:  tensor(4.3265) , ls:  1 , rel_err:  tensor(110.9226) , time:  0.0012247562408447266
iter:  2950 , fixed pt res:  tensor(4.2206) , ls:  1 , rel_err:  tensor(111.6767) , time:  0.0012023448944091797
iter:  3000 , fixed pt res:  tensor(4.3502) , ls:  1 , rel_err:  tensor(112.5193) , time:  0.0012269020080566406
iter:  3050 , fixed pt res:  tensor(4.3318) , ls:  1 , rel_err:  tensor(112.9838) , time:  0.0012021064758300781
iter:  3100 , fixed pt res:  tensor(4.4128) , ls:  1 , rel_err:  tensor(114.0561) , time:  0.0012402534484863281
iter:  3150 , fixed pt res:  tensor(4.1856) , ls:  1 , rel_err:  tensor(114.3409) , time:  0.0012400150299072266
iter:  3200 , fixed pt res:  tensor(4.3584) , ls:  1 , rel_err:  tensor(115.2822) , time:  0.001224517822265625
iter:  3250 , fixed pt res:  tensor(4.1931) , ls:  1 , rel_err:  tensor(115.8058) , time:  0.0012483596801757812
iter:  3300 , fixed pt res:  tensor(4.3942) , ls:  1 , rel_err:  tensor(115.2320) , time:  0.0012700557708740234
iter:  3350 , fixed pt res:  tensor(4.4909) , ls:  1 , rel_err:  tensor(116.2544) , time:  0.0013701915740966797
iter:  3400 , fixed pt res:  tensor(4.3221) , ls:  1 , rel_err:  tensor(116.9406) , time:  0.0012536048889160156
iter:  3450 , fixed pt res:  tensor(4.2316) , ls:  1 , rel_err:  tensor(118.2200) , time:  0.001245260238647461
iter:  3500 , fixed pt res:  tensor(4.3480) , ls:  1 , rel_err:  tensor(119.6118) , time:  0.0012774467468261719
iter:  3550 , fixed pt res:  tensor(4.4444) , ls:  1 , rel_err:  tensor(120.3955) , time:  0.0012316703796386719
iter:  3600 , fixed pt res:  tensor(4.2887) , ls:  1 , rel_err:  tensor(121.2983) , time:  0.0012226104736328125
iter:  3650 , fixed pt res:  tensor(4.3580) , ls:  1 , rel_err:  tensor(122.7465) , time:  0.001253366470336914
iter:  3700 , fixed pt res:  tensor(4.3071) , ls:  1 , rel_err:  tensor(122.0843) , time:  0.001230478286743164
iter:  3750 , fixed pt res:  tensor(4.4815) , ls:  1 , rel_err:  tensor(122.9040) , time:  0.0012171268463134766
iter:  3800 , fixed pt res:  tensor(4.1557) , ls:  1 , rel_err:  tensor(123.1052) , time:  0.0012226104736328125
iter:  3850 , fixed pt res:  tensor(4.2987) , ls:  1 , rel_err:  tensor(123.6549) , time:  0.001209259033203125
iter:  3900 , fixed pt res:  tensor(4.3005) , ls:  1 , rel_err:  tensor(123.6455) , time:  0.0012559890747070312
iter:  3950 , fixed pt res:  tensor(4.4685) , ls:  1 , rel_err:  tensor(124.7504) , time:  0.0014450550079345703
iter:  4000 , fixed pt res:  tensor(4.3504) , ls:  1 , rel_err:  tensor(125.1232) , time:  0.0012197494506835938
iter:  4050 , fixed pt res:  tensor(4.2964) , ls:  1 , rel_err:  tensor(124.8335) , time:  0.0012235641479492188
iter:  4100 , fixed pt res:  tensor(4.4660) , ls:  1 , rel_err:  tensor(126.1516) , time:  0.001220703125
iter:  4150 , fixed pt res:  tensor(4.5462) , ls:  1 , rel_err:  tensor(127.2025) , time:  0.0012118816375732422
iter:  4200 , fixed pt res:  tensor(4.3123) , ls:  1 , rel_err:  tensor(126.9322) , time:  0.0012183189392089844
iter:  4250 , fixed pt res:  tensor(4.3521) , ls:  1 , rel_err:  tensor(127.3104) , time:  0.0012164115905761719
iter:  4300 , fixed pt res:  tensor(4.2757) , ls:  1 , rel_err:  tensor(128.1557) , time:  0.001264810562133789
iter:  4350 , fixed pt res:  tensor(4.3231) , ls:  1 , rel_err:  tensor(129.0909) , time:  0.0012233257293701172
iter:  4400 , fixed pt res:  tensor(4.3421) , ls:  1 , rel_err:  tensor(129.4120) , time:  0.0013189315795898438
iter:  4450 , fixed pt res:  tensor(4.5757) , ls:  1 , rel_err:  tensor(129.6437) , time:  0.0012159347534179688
iter:  4500 , fixed pt res:  tensor(4.3383) , ls:  1 , rel_err:  tensor(130.6270) , time:  0.0012118816375732422
iter:  4550 , fixed pt res:  tensor(4.2816) , ls:  1 , rel_err:  tensor(130.9382) , time:  0.0012242794036865234
iter:  4600 , fixed pt res:  tensor(4.5565) , ls:  1 , rel_err:  tensor(131.5833) , time:  0.001222848892211914
iter:  4650 , fixed pt res:  tensor(4.3938) , ls:  1 , rel_err:  tensor(132.9198) , time:  0.0012316703796386719
iter:  4700 , fixed pt res:  tensor(4.1798) , ls:  1 , rel_err:  tensor(134.1844) , time:  0.0013203620910644531
iter:  4750 , fixed pt res:  tensor(4.2626) , ls:  1 , rel_err:  tensor(134.6126) , time:  0.0012254714965820312
iter:  4800 , fixed pt res:  tensor(4.3756) , ls:  1 , rel_err:  tensor(135.4791) , time:  0.0012345314025878906
iter:  4850 , fixed pt res:  tensor(4.2080) , ls:  1 , rel_err:  tensor(137.6846) , time:  0.0015499591827392578
iter:  4900 , fixed pt res:  tensor(4.5486) , ls:  1 , rel_err:  tensor(137.8125) , time:  0.0012052059173583984
iter:  4950 , fixed pt res:  tensor(4.4195) , ls:  1 , rel_err:  tensor(138.8419) , time:  0.0012307167053222656
iter:  5000 , fixed pt res:  tensor(4.4447) , ls:  1 , rel_err:  tensor(139.9588) , time:  0.0012805461883544922
iter:  5050 , fixed pt res:  tensor(4.4724) , ls:  1 , rel_err:  tensor(140.3710) , time:  0.0012123584747314453
iter:  5100 , fixed pt res:  tensor(4.3861) , ls:  1 , rel_err:  tensor(141.2079) , time:  0.0012962818145751953
iter:  5150 , fixed pt res:  tensor(4.2526) , ls:  1 , rel_err:  tensor(143.0183) , time:  0.0012211799621582031
iter:  5200 , fixed pt res:  tensor(4.4316) , ls:  1 , rel_err:  tensor(143.8309) , time:  0.0012254714965820312
iter:  5250 , fixed pt res:  tensor(4.3292) , ls:  1 , rel_err:  tensor(144.8065) , time:  0.0012087821960449219
iter:  5300 , fixed pt res:  tensor(4.1978) , ls:  1 , rel_err:  tensor(145.1760) , time:  0.0012331008911132812
iter:  5350 , fixed pt res:  tensor(4.3954) , ls:  1 , rel_err:  tensor(145.9453) , time:  0.0012140274047851562
iter:  5400 , fixed pt res:  tensor(4.5053) , ls:  1 , rel_err:  tensor(148.0731) , time:  0.0012247562408447266
iter:  5450 , fixed pt res:  tensor(4.4561) , ls:  1 , rel_err:  tensor(148.3923) , time:  0.0014989376068115234
iter:  5500 , fixed pt res:  tensor(4.5518) , ls:  1 , rel_err:  tensor(147.9458) , time:  0.001608133316040039
iter:  5550 , fixed pt res:  tensor(4.3144) , ls:  1 , rel_err:  tensor(148.8063) , time:  0.0012576580047607422
iter:  5600 , fixed pt res:  tensor(4.3206) , ls:  1 , rel_err:  tensor(149.4920) , time:  0.0016512870788574219
iter:  5650 , fixed pt res:  tensor(4.4863) , ls:  1 , rel_err:  tensor(149.8737) , time:  0.001215219497680664
iter:  5700 , fixed pt res:  tensor(4.4126) , ls:  1 , rel_err:  tensor(150.2984) , time:  0.0012290477752685547
iter:  5750 , fixed pt res:  tensor(4.5650) , ls:  1 , rel_err:  tensor(151.4765) , time:  0.0012562274932861328
iter:  5800 , fixed pt res:  tensor(4.2554) , ls:  1 , rel_err:  tensor(150.7308) , time:  0.0012602806091308594
iter:  5850 , fixed pt res:  tensor(4.2354) , ls:  1 , rel_err:  tensor(151.0924) , time:  0.0012059211730957031
iter:  5900 , fixed pt res:  tensor(4.4356) , ls:  1 , rel_err:  tensor(152.0713) , time:  0.0012755393981933594
iter:  5950 , fixed pt res:  tensor(4.4027) , ls:  1 , rel_err:  tensor(152.7037) , time:  0.001271963119506836
iter:  6000 , fixed pt res:  tensor(4.1286) , ls:  1 , rel_err:  tensor(154.0682) , time:  0.0012352466583251953
iter:  6050 , fixed pt res:  tensor(4.4555) , ls:  1 , rel_err:  tensor(155.5801) , time:  0.001233816146850586
iter:  6100 , fixed pt res:  tensor(4.1264) , ls:  1 , rel_err:  tensor(155.5533) , time:  0.0012335777282714844
iter:  6150 , fixed pt res:  tensor(4.3133) , ls:  1 , rel_err:  tensor(155.6941) , time:  0.0012552738189697266
iter:  6200 , fixed pt res:  tensor(4.1991) , ls:  1 , rel_err:  tensor(155.8660) , time:  0.0012211799621582031
iter:  6250 , fixed pt res:  tensor(4.1847) , ls:  1 , rel_err:  tensor(155.8810) , time:  0.0012273788452148438
iter:  6300 , fixed pt res:  tensor(4.3835) , ls:  1 , rel_err:  tensor(157.5197) , time:  0.0012090206146240234
iter:  6350 , fixed pt res:  tensor(4.4773) , ls:  1 , rel_err:  tensor(158.0135) , time:  0.0012416839599609375
iter:  6400 , fixed pt res:  tensor(4.2756) , ls:  1 , rel_err:  tensor(158.2304) , time:  0.001211404800415039
iter:  6450 , fixed pt res:  tensor(4.3285) , ls:  1 , rel_err:  tensor(158.2862) , time:  0.0012767314910888672
iter:  6500 , fixed pt res:  tensor(4.3430) , ls:  1 , rel_err:  tensor(158.4957) , time:  0.0012328624725341797
iter:  6550 , fixed pt res:  tensor(4.4024) , ls:  1 , rel_err:  tensor(158.7941) , time:  0.001210927963256836
iter:  6600 , fixed pt res:  tensor(4.3154) , ls:  1 , rel_err:  tensor(160.0966) , time:  0.0012059211730957031
iter:  6650 , fixed pt res:  tensor(4.3362) , ls:  1 , rel_err:  tensor(161.1296) , time:  0.0012753009796142578
iter:  6700 , fixed pt res:  tensor(4.2554) , ls:  1 , rel_err:  tensor(161.2044) , time:  0.0012421607971191406
iter:  6750 , fixed pt res:  tensor(4.3534) , ls:  1 , rel_err:  tensor(161.5621) , time:  0.0020461082458496094
iter:  6800 , fixed pt res:  tensor(4.4202) , ls:  1 , rel_err:  tensor(162.2499) , time:  0.0012614727020263672
iter:  6850 , fixed pt res:  tensor(4.3849) , ls:  1 , rel_err:  tensor(162.1226) , time:  0.0012538433074951172
iter:  6900 , fixed pt res:  tensor(4.4920) , ls:  1 , rel_err:  tensor(163.0170) , time:  0.001508951187133789
iter:  6950 , fixed pt res:  tensor(4.3237) , ls:  1 , rel_err:  tensor(163.7699) , time:  0.001219034194946289
iter:  7000 , fixed pt res:  tensor(4.3307) , ls:  1 , rel_err:  tensor(164.7055) , time:  0.001251220703125
iter:  7050 , fixed pt res:  tensor(4.3875) , ls:  1 , rel_err:  tensor(166.1488) , time:  0.001219034194946289
iter:  7100 , fixed pt res:  tensor(4.2454) , ls:  1 , rel_err:  tensor(166.5173) , time:  0.0012180805206298828
iter:  7150 , fixed pt res:  tensor(4.5403) , ls:  1 , rel_err:  tensor(166.6101) , time:  0.0012042522430419922
iter:  7200 , fixed pt res:  tensor(4.3249) , ls:  1 , rel_err:  tensor(166.7866) , time:  0.0011980533599853516
iter:  7250 , fixed pt res:  tensor(4.2379) , ls:  1 , rel_err:  tensor(166.9070) , time:  0.001207113265991211
iter:  7300 , fixed pt res:  tensor(4.1400) , ls:  1 , rel_err:  tensor(166.6404) , time:  0.001207113265991211
iter:  7350 , fixed pt res:  tensor(4.3063) , ls:  1 , rel_err:  tensor(167.2403) , time:  0.0011966228485107422
iter:  7400 , fixed pt res:  tensor(4.3218) , ls:  1 , rel_err:  tensor(167.5425) , time:  0.001191854476928711
iter:  7450 , fixed pt res:  tensor(4.4948) , ls:  1 , rel_err:  tensor(169.0356) , time:  0.0011944770812988281
iter:  7500 , fixed pt res:  tensor(4.3100) , ls:  1 , rel_err:  tensor(168.2851) , time:  0.001407623291015625
iter:  7550 , fixed pt res:  tensor(4.2834) , ls:  1 , rel_err:  tensor(169.4251) , time:  0.0013039112091064453
iter:  7600 , fixed pt res:  tensor(4.3020) , ls:  1 , rel_err:  tensor(170.3145) , time:  0.001260519027709961
iter:  7650 , fixed pt res:  tensor(4.1987) , ls:  1 , rel_err:  tensor(170.9172) , time:  0.0012359619140625
iter:  7700 , fixed pt res:  tensor(4.2722) , ls:  1 , rel_err:  tensor(171.9099) , time:  0.0012178421020507812
iter:  7750 , fixed pt res:  tensor(4.3431) , ls:  1 , rel_err:  tensor(172.2238) , time:  0.0012478828430175781
iter:  7800 , fixed pt res:  tensor(4.4537) , ls:  1 , rel_err:  tensor(172.2792) , time:  0.0012288093566894531
iter:  7850 , fixed pt res:  tensor(4.2815) , ls:  1 , rel_err:  tensor(172.2619) , time:  0.0012218952178955078
iter:  7900 , fixed pt res:  tensor(4.4923) , ls:  1 , rel_err:  tensor(171.8362) , time:  0.001215219497680664
iter:  7950 , fixed pt res:  tensor(4.1887) , ls:  1 , rel_err:  tensor(172.1641) , time:  0.0014383792877197266
iter:  8000 , fixed pt res:  tensor(4.1604) , ls:  1 , rel_err:  tensor(171.9559) , time:  0.0012226104736328125
iter:  8050 , fixed pt res:  tensor(4.5022) , ls:  1 , rel_err:  tensor(172.6269) , time:  0.0012369155883789062
iter:  8100 , fixed pt res:  tensor(4.3863) , ls:  1 , rel_err:  tensor(172.6648) , time:  0.0012516975402832031
iter:  8150 , fixed pt res:  tensor(4.5391) , ls:  1 , rel_err:  tensor(172.7342) , time:  0.0012407302856445312
iter:  8200 , fixed pt res:  tensor(4.3726) , ls:  1 , rel_err:  tensor(174.2308) , time:  0.0012116432189941406
iter:  8250 , fixed pt res:  tensor(4.4113) , ls:  1 , rel_err:  tensor(174.1882) , time:  0.001323699951171875
iter:  8300 , fixed pt res:  tensor(4.3925) , ls:  1 , rel_err:  tensor(174.1840) , time:  0.0012085437774658203
iter:  8350 , fixed pt res:  tensor(4.4766) , ls:  1 , rel_err:  tensor(174.4624) , time:  0.0012044906616210938
iter:  8400 , fixed pt res:  tensor(4.3198) , ls:  1 , rel_err:  tensor(174.3825) , time:  0.0013380050659179688
iter:  8450 , fixed pt res:  tensor(4.2532) , ls:  1 , rel_err:  tensor(175.6964) , time:  0.001226663589477539
iter:  8500 , fixed pt res:  tensor(4.1973) , ls:  1 , rel_err:  tensor(175.1025) , time:  0.0020666122436523438
iter:  8550 , fixed pt res:  tensor(4.2542) , ls:  1 , rel_err:  tensor(175.3418) , time:  0.0012395381927490234
iter:  8600 , fixed pt res:  tensor(4.3963) , ls:  1 , rel_err:  tensor(175.6121) , time:  0.001481771469116211
iter:  8650 , fixed pt res:  tensor(4.5053) , ls:  1 , rel_err:  tensor(175.6718) , time:  0.0012290477752685547
iter:  8700 , fixed pt res:  tensor(4.5122) , ls:  1 , rel_err:  tensor(175.7919) , time:  0.0012845993041992188
iter:  8750 , fixed pt res:  tensor(4.5139) , ls:  1 , rel_err:  tensor(177.0038) , time:  0.001233816146850586
iter:  8800 , fixed pt res:  tensor(4.4545) , ls:  1 , rel_err:  tensor(177.6089) , time:  0.001232147216796875
iter:  8850 , fixed pt res:  tensor(4.4173) , ls:  1 , rel_err:  tensor(178.6840) , time:  0.001413106918334961
iter:  8900 , fixed pt res:  tensor(4.3516) , ls:  1 , rel_err:  tensor(179.5216) , time:  0.0012238025665283203
iter:  8950 , fixed pt res:  tensor(4.4677) , ls:  1 , rel_err:  tensor(180.7708) , time:  0.0012068748474121094
iter:  9000 , fixed pt res:  tensor(4.1992) , ls:  1 , rel_err:  tensor(180.7370) , time:  0.0012822151184082031
iter:  9050 , fixed pt res:  tensor(4.5322) , ls:  1 , rel_err:  tensor(181.2392) , time:  0.0012969970703125
iter:  9100 , fixed pt res:  tensor(4.4024) , ls:  1 , rel_err:  tensor(181.9343) , time:  0.0012459754943847656
iter:  9150 , fixed pt res:  tensor(4.2635) , ls:  1 , rel_err:  tensor(183.3682) , time:  0.0014431476593017578
iter:  9200 , fixed pt res:  tensor(4.3862) , ls:  1 , rel_err:  tensor(183.9166) , time:  0.0012183189392089844
iter:  9250 , fixed pt res:  tensor(4.4897) , ls:  1 , rel_err:  tensor(185.5516) , time:  0.0012104511260986328
iter:  9300 , fixed pt res:  tensor(4.3664) , ls:  1 , rel_err:  tensor(186.3082) , time:  0.001222848892211914
iter:  9350 , fixed pt res:  tensor(4.2656) , ls:  1 , rel_err:  tensor(186.6163) , time:  0.0011990070343017578
iter:  9400 , fixed pt res:  tensor(4.2133) , ls:  1 , rel_err:  tensor(187.3418) , time:  0.001234292984008789
iter:  9450 , fixed pt res:  tensor(4.4000) , ls:  1 , rel_err:  tensor(188.1650) , time:  0.0012516975402832031
iter:  9500 , fixed pt res:  tensor(4.3824) , ls:  1 , rel_err:  tensor(188.9037) , time:  0.0012440681457519531
iter:  9550 , fixed pt res:  tensor(4.3504) , ls:  1 , rel_err:  tensor(189.1673) , time:  0.00121307373046875
iter:  9600 , fixed pt res:  tensor(4.5031) , ls:  1 , rel_err:  tensor(190.2241) , time:  0.0012161731719970703
iter:  9650 , fixed pt res:  tensor(4.3399) , ls:  1 , rel_err:  tensor(190.3523) , time:  0.0012121200561523438
iter:  9700 , fixed pt res:  tensor(4.4984) , ls:  1 , rel_err:  tensor(191.1164) , time:  0.0013153553009033203
iter:  9750 , fixed pt res:  tensor(4.3649) , ls:  1 , rel_err:  tensor(192.1975) , time:  0.001260995864868164
iter:  9800 , fixed pt res:  tensor(4.3691) , ls:  1 , rel_err:  tensor(191.9936) , time:  0.0012807846069335938
iter:  9850 , fixed pt res:  tensor(4.4353) , ls:  1 , rel_err:  tensor(192.7367) , time:  0.0012423992156982422
iter:  9900 , fixed pt res:  tensor(4.4677) , ls:  1 , rel_err:  tensor(193.7720) , time:  0.0012576580047607422
iter:  9950 , fixed pt res:  tensor(4.3881) , ls:  1 , rel_err:  tensor(193.7817) , time:  0.0018002986907958984
iter:  10000 , fixed pt res:  tensor(4.2635) , ls:  1 , rel_err:  tensor(194.5954) , time:  0.0012183189392089844
trial  1  finished after 13.358325004577637
norm(rel_err_hist_temp - rel_err_lmm_hist_sample_array[sample_index,:]) =  tensor(0.)
Number of Samples = 100
iter:  50 , fixed pt res:  tensor(0.7745) , ls:  1 , rel_err:  tensor(15.3021) , time:  0.001310110092163086
iter:  100 , fixed pt res:  tensor(0.5291) , ls:  1 , rel_err:  tensor(14.1823) , time:  0.001262664794921875
iter:  150 , fixed pt res:  tensor(0.4858) , ls:  1 , rel_err:  tensor(13.4265) , time:  0.001257181167602539
iter:  200 , fixed pt res:  tensor(0.4850) , ls:  1 , rel_err:  tensor(12.6152) , time:  0.0014407634735107422
iter:  250 , fixed pt res:  tensor(0.4468) , ls:  1 , rel_err:  tensor(11.8486) , time:  0.00128173828125
iter:  300 , fixed pt res:  tensor(0.4460) , ls:  1 , rel_err:  tensor(11.1918) , time:  0.0014660358428955078
iter:  350 , fixed pt res:  tensor(0.4499) , ls:  1 , rel_err:  tensor(10.4513) , time:  0.0012519359588623047
iter:  400 , fixed pt res:  tensor(0.4664) , ls:  1 , rel_err:  tensor(10.0025) , time:  0.0012450218200683594
iter:  450 , fixed pt res:  tensor(0.4562) , ls:  1 , rel_err:  tensor(9.3706) , time:  0.0014846324920654297
iter:  500 , fixed pt res:  tensor(0.4679) , ls:  1 , rel_err:  tensor(8.8386) , time:  0.0012602806091308594
iter:  550 , fixed pt res:  tensor(0.4514) , ls:  1 , rel_err:  tensor(8.2698) , time:  0.001310586929321289
iter:  600 , fixed pt res:  tensor(0.4350) , ls:  1 , rel_err:  tensor(7.8019) , time:  0.0014019012451171875
iter:  650 , fixed pt res:  tensor(0.4584) , ls:  1 , rel_err:  tensor(7.3331) , time:  0.001260519027709961
iter:  700 , fixed pt res:  tensor(0.4525) , ls:  1 , rel_err:  tensor(7.0093) , time:  0.0012497901916503906
iter:  750 , fixed pt res:  tensor(0.4378) , ls:  1 , rel_err:  tensor(6.7379) , time:  0.002296924591064453
iter:  800 , fixed pt res:  tensor(0.4329) , ls:  1 , rel_err:  tensor(6.3145) , time:  0.0013031959533691406
iter:  850 , fixed pt res:  tensor(0.4491) , ls:  1 , rel_err:  tensor(5.9935) , time:  0.0012462139129638672
iter:  900 , fixed pt res:  tensor(0.4482) , ls:  1 , rel_err:  tensor(5.6870) , time:  0.0012958049774169922
iter:  950 , fixed pt res:  tensor(0.4573) , ls:  1 , rel_err:  tensor(5.4872) , time:  0.0013692378997802734
iter:  1000 , fixed pt res:  tensor(0.4454) , ls:  1 , rel_err:  tensor(5.2836) , time:  0.001262664794921875
iter:  1050 , fixed pt res:  tensor(0.4282) , ls:  1 , rel_err:  tensor(5.1239) , time:  0.002058744430541992
iter:  1100 , fixed pt res:  tensor(0.4601) , ls:  1 , rel_err:  tensor(4.8779) , time:  0.0012831687927246094
iter:  1150 , fixed pt res:  tensor(0.4569) , ls:  1 , rel_err:  tensor(4.7101) , time:  0.0012745857238769531
iter:  1200 , fixed pt res:  tensor(0.4447) , ls:  1 , rel_err:  tensor(4.3678) , time:  0.0012636184692382812
iter:  1250 , fixed pt res:  tensor(0.4372) , ls:  1 , rel_err:  tensor(4.1949) , time:  0.0012631416320800781
iter:  1300 , fixed pt res:  tensor(0.4548) , ls:  1 , rel_err:  tensor(4.1029) , time:  0.0012526512145996094
iter:  1350 , fixed pt res:  tensor(0.4440) , ls:  1 , rel_err:  tensor(3.9807) , time:  0.0013229846954345703
iter:  1400 , fixed pt res:  tensor(0.4486) , ls:  1 , rel_err:  tensor(3.8899) , time:  0.0012824535369873047
iter:  1450 , fixed pt res:  tensor(0.4442) , ls:  1 , rel_err:  tensor(3.7541) , time:  0.00128936767578125
iter:  1500 , fixed pt res:  tensor(0.4513) , ls:  1 , rel_err:  tensor(3.5777) , time:  0.0012822151184082031
iter:  1550 , fixed pt res:  tensor(0.4171) , ls:  1 , rel_err:  tensor(3.4925) , time:  0.0012974739074707031
iter:  1600 , fixed pt res:  tensor(0.4389) , ls:  1 , rel_err:  tensor(3.2979) , time:  0.0012981891632080078
iter:  1650 , fixed pt res:  tensor(0.4353) , ls:  1 , rel_err:  tensor(3.1825) , time:  0.0013427734375
iter:  1700 , fixed pt res:  tensor(0.4445) , ls:  1 , rel_err:  tensor(3.1297) , time:  0.0012636184692382812
iter:  1750 , fixed pt res:  tensor(0.4608) , ls:  1 , rel_err:  tensor(3.1012) , time:  0.002191781997680664
iter:  1800 , fixed pt res:  tensor(0.4407) , ls:  1 , rel_err:  tensor(2.9704) , time:  0.0013363361358642578
iter:  1850 , fixed pt res:  tensor(0.4530) , ls:  1 , rel_err:  tensor(3.0412) , time:  0.0012786388397216797
iter:  1900 , fixed pt res:  tensor(0.4420) , ls:  1 , rel_err:  tensor(2.9323) , time:  0.0013391971588134766
iter:  1950 , fixed pt res:  tensor(0.4544) , ls:  1 , rel_err:  tensor(2.9978) , time:  0.0013275146484375
iter:  2000 , fixed pt res:  tensor(0.4489) , ls:  1 , rel_err:  tensor(2.9866) , time:  0.0012645721435546875
iter:  2050 , fixed pt res:  tensor(0.4557) , ls:  1 , rel_err:  tensor(2.8863) , time:  0.0012743473052978516
iter:  2100 , fixed pt res:  tensor(0.4534) , ls:  1 , rel_err:  tensor(2.8482) , time:  0.001253366470336914
iter:  2150 , fixed pt res:  tensor(0.4365) , ls:  1 , rel_err:  tensor(2.8268) , time:  0.0012631416320800781
iter:  2200 , fixed pt res:  tensor(0.4519) , ls:  1 , rel_err:  tensor(2.8113) , time:  0.002457141876220703
iter:  2250 , fixed pt res:  tensor(0.4497) , ls:  1 , rel_err:  tensor(2.8356) , time:  0.0012688636779785156
iter:  2300 , fixed pt res:  tensor(0.4532) , ls:  1 , rel_err:  tensor(2.8252) , time:  0.0012612342834472656
iter:  2350 , fixed pt res:  tensor(0.4499) , ls:  1 , rel_err:  tensor(2.7863) , time:  0.0012767314910888672
iter:  2400 , fixed pt res:  tensor(0.4548) , ls:  1 , rel_err:  tensor(2.6750) , time:  0.0012617111206054688
iter:  2450 , fixed pt res:  tensor(0.4557) , ls:  1 , rel_err:  tensor(2.6320) , time:  0.001291513442993164
iter:  2500 , fixed pt res:  tensor(0.4548) , ls:  1 , rel_err:  tensor(2.5933) , time:  0.0012500286102294922
iter:  2550 , fixed pt res:  tensor(0.4298) , ls:  1 , rel_err:  tensor(2.5141) , time:  0.00127410888671875
iter:  2600 , fixed pt res:  tensor(0.4316) , ls:  1 , rel_err:  tensor(2.5273) , time:  0.0012772083282470703
iter:  2650 , fixed pt res:  tensor(0.4303) , ls:  1 , rel_err:  tensor(2.5612) , time:  0.0012912750244140625
iter:  2700 , fixed pt res:  tensor(0.4467) , ls:  1 , rel_err:  tensor(2.4864) , time:  0.0012738704681396484
iter:  2750 , fixed pt res:  tensor(0.4455) , ls:  1 , rel_err:  tensor(2.5092) , time:  0.001268625259399414
iter:  2800 , fixed pt res:  tensor(0.4444) , ls:  1 , rel_err:  tensor(2.4727) , time:  0.0013165473937988281
iter:  2850 , fixed pt res:  tensor(0.4379) , ls:  1 , rel_err:  tensor(2.5988) , time:  0.0015835762023925781
iter:  2900 , fixed pt res:  tensor(0.4466) , ls:  1 , rel_err:  tensor(2.4992) , time:  0.0012874603271484375
iter:  2950 , fixed pt res:  tensor(0.4663) , ls:  1 , rel_err:  tensor(2.4649) , time:  0.0013229846954345703
iter:  3000 , fixed pt res:  tensor(0.4610) , ls:  1 , rel_err:  tensor(2.3884) , time:  0.0015511512756347656
iter:  3050 , fixed pt res:  tensor(0.4496) , ls:  1 , rel_err:  tensor(2.4362) , time:  0.0012707710266113281
iter:  3100 , fixed pt res:  tensor(0.4613) , ls:  1 , rel_err:  tensor(2.4918) , time:  0.0012714862823486328
iter:  3150 , fixed pt res:  tensor(0.4323) , ls:  1 , rel_err:  tensor(2.5321) , time:  0.0016469955444335938
iter:  3200 , fixed pt res:  tensor(0.4321) , ls:  1 , rel_err:  tensor(2.5537) , time:  0.001306772232055664
iter:  3250 , fixed pt res:  tensor(0.4411) , ls:  1 , rel_err:  tensor(2.5134) , time:  0.0013120174407958984
iter:  3300 , fixed pt res:  tensor(0.4662) , ls:  1 , rel_err:  tensor(2.4812) , time:  0.0013124942779541016
iter:  3350 , fixed pt res:  tensor(0.4527) , ls:  1 , rel_err:  tensor(2.5300) , time:  0.0012688636779785156
iter:  3400 , fixed pt res:  tensor(0.4342) , ls:  1 , rel_err:  tensor(2.5346) , time:  0.0013089179992675781
iter:  3450 , fixed pt res:  tensor(0.4434) , ls:  1 , rel_err:  tensor(2.5757) , time:  0.001308441162109375
iter:  3500 , fixed pt res:  tensor(0.4403) , ls:  1 , rel_err:  tensor(2.5628) , time:  0.0012698173522949219
iter:  3550 , fixed pt res:  tensor(0.4538) , ls:  1 , rel_err:  tensor(2.5536) , time:  0.0012412071228027344
iter:  3600 , fixed pt res:  tensor(0.4477) , ls:  1 , rel_err:  tensor(2.5987) , time:  0.001482248306274414
iter:  3650 , fixed pt res:  tensor(0.4369) , ls:  1 , rel_err:  tensor(2.5586) , time:  0.0013012886047363281
iter:  3700 , fixed pt res:  tensor(0.4333) , ls:  1 , rel_err:  tensor(2.5921) , time:  0.001262664794921875
iter:  3750 , fixed pt res:  tensor(0.4608) , ls:  1 , rel_err:  tensor(2.6525) , time:  0.0014340877532958984
iter:  3800 , fixed pt res:  tensor(0.4732) , ls:  1 , rel_err:  tensor(2.5588) , time:  0.0012569427490234375
iter:  3850 , fixed pt res:  tensor(0.4544) , ls:  1 , rel_err:  tensor(2.4464) , time:  0.001260995864868164
iter:  3900 , fixed pt res:  tensor(0.4411) , ls:  1 , rel_err:  tensor(2.4579) , time:  0.0020532608032226562
iter:  3950 , fixed pt res:  tensor(0.4562) , ls:  1 , rel_err:  tensor(2.4835) , time:  0.0012850761413574219
iter:  4000 , fixed pt res:  tensor(0.4672) , ls:  1 , rel_err:  tensor(2.5577) , time:  0.0012614727020263672
iter:  4050 , fixed pt res:  tensor(0.4457) , ls:  1 , rel_err:  tensor(2.6432) , time:  0.0012993812561035156
iter:  4100 , fixed pt res:  tensor(0.4429) , ls:  1 , rel_err:  tensor(2.6017) , time:  0.0014777183532714844
iter:  4150 , fixed pt res:  tensor(0.4509) , ls:  1 , rel_err:  tensor(2.6780) , time:  0.0012485980987548828
iter:  4200 , fixed pt res:  tensor(0.4659) , ls:  1 , rel_err:  tensor(2.7382) , time:  0.0015103816986083984
iter:  4250 , fixed pt res:  tensor(0.4457) , ls:  1 , rel_err:  tensor(2.8506) , time:  0.0012941360473632812
iter:  4300 , fixed pt res:  tensor(0.4542) , ls:  1 , rel_err:  tensor(2.8595) , time:  0.0012731552124023438
iter:  4350 , fixed pt res:  tensor(0.4522) , ls:  1 , rel_err:  tensor(2.7833) , time:  0.0013074874877929688
iter:  4400 , fixed pt res:  tensor(0.4330) , ls:  1 , rel_err:  tensor(2.7281) , time:  0.0012753009796142578
iter:  4450 , fixed pt res:  tensor(0.4533) , ls:  1 , rel_err:  tensor(2.6754) , time:  0.0012965202331542969
iter:  4500 , fixed pt res:  tensor(0.4516) , ls:  1 , rel_err:  tensor(2.7161) , time:  0.0012755393981933594
iter:  4550 , fixed pt res:  tensor(0.4267) , ls:  1 , rel_err:  tensor(2.6371) , time:  0.001272439956665039
iter:  4600 , fixed pt res:  tensor(0.4389) , ls:  1 , rel_err:  tensor(2.6671) , time:  0.0013155937194824219
iter:  4650 , fixed pt res:  tensor(0.4410) , ls:  1 , rel_err:  tensor(2.7035) , time:  0.002562284469604492
iter:  4700 , fixed pt res:  tensor(0.4573) , ls:  1 , rel_err:  tensor(2.7827) , time:  0.0012750625610351562
iter:  4750 , fixed pt res:  tensor(0.4588) , ls:  1 , rel_err:  tensor(2.7491) , time:  0.001306295394897461
iter:  4800 , fixed pt res:  tensor(0.4312) , ls:  1 , rel_err:  tensor(2.7217) , time:  0.001256704330444336
iter:  4850 , fixed pt res:  tensor(0.4261) , ls:  1 , rel_err:  tensor(2.6792) , time:  0.0013720989227294922
iter:  4900 , fixed pt res:  tensor(0.4446) , ls:  1 , rel_err:  tensor(2.6511) , time:  0.0012812614440917969
iter:  4950 , fixed pt res:  tensor(0.4327) , ls:  1 , rel_err:  tensor(2.6247) , time:  0.0013003349304199219
iter:  5000 , fixed pt res:  tensor(0.4393) , ls:  1 , rel_err:  tensor(2.7084) , time:  0.0012722015380859375
iter:  5050 , fixed pt res:  tensor(0.4415) , ls:  1 , rel_err:  tensor(2.6556) , time:  0.0012764930725097656
iter:  5100 , fixed pt res:  tensor(0.4376) , ls:  1 , rel_err:  tensor(2.6374) , time:  0.00128936767578125
iter:  5150 , fixed pt res:  tensor(0.4451) , ls:  1 , rel_err:  tensor(2.6773) , time:  0.0012967586517333984
iter:  5200 , fixed pt res:  tensor(0.4329) , ls:  1 , rel_err:  tensor(2.6425) , time:  0.0012586116790771484
iter:  5250 , fixed pt res:  tensor(0.4457) , ls:  1 , rel_err:  tensor(2.6239) , time:  0.0014810562133789062
iter:  5300 , fixed pt res:  tensor(0.4464) , ls:  1 , rel_err:  tensor(2.5870) , time:  0.0012464523315429688
iter:  5350 , fixed pt res:  tensor(0.4333) , ls:  1 , rel_err:  tensor(2.5976) , time:  0.0012700557708740234
iter:  5400 , fixed pt res:  tensor(0.4657) , ls:  1 , rel_err:  tensor(2.6095) , time:  0.002093791961669922
iter:  5450 , fixed pt res:  tensor(0.4228) , ls:  1 , rel_err:  tensor(2.6575) , time:  0.0012738704681396484
iter:  5500 , fixed pt res:  tensor(0.4472) , ls:  1 , rel_err:  tensor(2.6417) , time:  0.001249551773071289
iter:  5550 , fixed pt res:  tensor(0.4502) , ls:  1 , rel_err:  tensor(2.6688) , time:  0.0013189315795898438
iter:  5600 , fixed pt res:  tensor(0.4398) , ls:  1 , rel_err:  tensor(2.5957) , time:  0.0012636184692382812
iter:  5650 , fixed pt res:  tensor(0.4403) , ls:  1 , rel_err:  tensor(2.7314) , time:  0.0015289783477783203
iter:  5700 , fixed pt res:  tensor(0.4475) , ls:  1 , rel_err:  tensor(2.7141) , time:  0.0012578964233398438
iter:  5750 , fixed pt res:  tensor(0.4478) , ls:  1 , rel_err:  tensor(2.5837) , time:  0.0012476444244384766
iter:  5800 , fixed pt res:  tensor(0.4301) , ls:  1 , rel_err:  tensor(2.5068) , time:  0.0012676715850830078
iter:  5850 , fixed pt res:  tensor(0.4321) , ls:  1 , rel_err:  tensor(2.5192) , time:  0.0021185874938964844
iter:  5900 , fixed pt res:  tensor(0.4248) , ls:  1 , rel_err:  tensor(2.5974) , time:  0.0012679100036621094
iter:  5950 , fixed pt res:  tensor(0.4331) , ls:  1 , rel_err:  tensor(2.5246) , time:  0.0012791156768798828
iter:  6000 , fixed pt res:  tensor(0.4590) , ls:  1 , rel_err:  tensor(2.4553) , time:  0.0012676715850830078
iter:  6050 , fixed pt res:  tensor(0.4304) , ls:  1 , rel_err:  tensor(2.4796) , time:  0.0013103485107421875
iter:  6100 , fixed pt res:  tensor(0.4495) , ls:  1 , rel_err:  tensor(2.4144) , time:  0.001794576644897461
iter:  6150 , fixed pt res:  tensor(0.4649) , ls:  1 , rel_err:  tensor(2.4198) , time:  0.0014438629150390625
iter:  6200 , fixed pt res:  tensor(0.4218) , ls:  1 , rel_err:  tensor(2.4917) , time:  0.0012674331665039062
iter:  6250 , fixed pt res:  tensor(0.4481) , ls:  1 , rel_err:  tensor(2.3632) , time:  0.0012516975402832031
iter:  6300 , fixed pt res:  tensor(0.4243) , ls:  1 , rel_err:  tensor(2.4195) , time:  0.0013699531555175781
iter:  6350 , fixed pt res:  tensor(0.4402) , ls:  1 , rel_err:  tensor(2.4351) , time:  0.0012640953063964844
iter:  6400 , fixed pt res:  tensor(0.4558) , ls:  1 , rel_err:  tensor(2.4017) , time:  0.0012509822845458984
iter:  6450 , fixed pt res:  tensor(0.4455) , ls:  1 , rel_err:  tensor(2.4357) , time:  0.0012545585632324219
iter:  6500 , fixed pt res:  tensor(0.4350) , ls:  1 , rel_err:  tensor(2.5407) , time:  0.0012562274932861328
iter:  6550 , fixed pt res:  tensor(0.4334) , ls:  1 , rel_err:  tensor(2.5926) , time:  0.0013165473937988281
iter:  6600 , fixed pt res:  tensor(0.4355) , ls:  1 , rel_err:  tensor(2.5967) , time:  0.0013005733489990234
iter:  6650 , fixed pt res:  tensor(0.4348) , ls:  1 , rel_err:  tensor(2.5459) , time:  0.001241922378540039
iter:  6700 , fixed pt res:  tensor(0.4418) , ls:  1 , rel_err:  tensor(2.4853) , time:  0.0012357234954833984
iter:  6750 , fixed pt res:  tensor(0.4555) , ls:  1 , rel_err:  tensor(2.4952) , time:  0.002100229263305664
iter:  6800 , fixed pt res:  tensor(0.4577) , ls:  1 , rel_err:  tensor(2.5541) , time:  0.001260519027709961
iter:  6850 , fixed pt res:  tensor(0.4522) , ls:  1 , rel_err:  tensor(2.5946) , time:  0.0012519359588623047
iter:  6900 , fixed pt res:  tensor(0.4356) , ls:  1 , rel_err:  tensor(2.6092) , time:  0.0017650127410888672
iter:  6950 , fixed pt res:  tensor(0.4543) , ls:  1 , rel_err:  tensor(2.7244) , time:  0.0012505054473876953
iter:  7000 , fixed pt res:  tensor(0.4458) , ls:  1 , rel_err:  tensor(2.6772) , time:  0.0012483596801757812
iter:  7050 , fixed pt res:  tensor(0.4605) , ls:  1 , rel_err:  tensor(2.6311) , time:  0.0019910335540771484
iter:  7100 , fixed pt res:  tensor(0.4489) , ls:  1 , rel_err:  tensor(2.6301) , time:  0.001264333724975586
iter:  7150 , fixed pt res:  tensor(0.4489) , ls:  1 , rel_err:  tensor(2.6015) , time:  0.0012781620025634766
iter:  7200 , fixed pt res:  tensor(0.4505) , ls:  1 , rel_err:  tensor(2.5872) , time:  0.0012555122375488281
iter:  7250 , fixed pt res:  tensor(0.4641) , ls:  1 , rel_err:  tensor(2.5395) , time:  0.0012531280517578125
iter:  7300 , fixed pt res:  tensor(0.4527) , ls:  1 , rel_err:  tensor(2.5581) , time:  0.001360177993774414
iter:  7350 , fixed pt res:  tensor(0.4420) , ls:  1 , rel_err:  tensor(2.4711) , time:  0.0014126300811767578
iter:  7400 , fixed pt res:  tensor(0.4186) , ls:  1 , rel_err:  tensor(2.4286) , time:  0.001264810562133789
iter:  7450 , fixed pt res:  tensor(0.4505) , ls:  1 , rel_err:  tensor(2.4745) , time:  0.0012433528900146484
iter:  7500 , fixed pt res:  tensor(0.4584) , ls:  1 , rel_err:  tensor(2.4456) , time:  0.0012326240539550781
iter:  7550 , fixed pt res:  tensor(0.4400) , ls:  1 , rel_err:  tensor(2.4178) , time:  0.0012383460998535156
iter:  7600 , fixed pt res:  tensor(0.4476) , ls:  1 , rel_err:  tensor(2.5011) , time:  0.0012450218200683594
iter:  7650 , fixed pt res:  tensor(0.4579) , ls:  1 , rel_err:  tensor(2.6409) , time:  0.0012357234954833984
iter:  7700 , fixed pt res:  tensor(0.4439) , ls:  1 , rel_err:  tensor(2.6166) , time:  0.0012264251708984375
iter:  7750 , fixed pt res:  tensor(0.4443) , ls:  1 , rel_err:  tensor(2.5579) , time:  0.0012443065643310547
iter:  7800 , fixed pt res:  tensor(0.4565) , ls:  1 , rel_err:  tensor(2.5625) , time:  0.0012531280517578125
iter:  7850 , fixed pt res:  tensor(0.4612) , ls:  1 , rel_err:  tensor(2.5585) , time:  0.0012328624725341797
iter:  7900 , fixed pt res:  tensor(0.4464) , ls:  1 , rel_err:  tensor(2.5860) , time:  0.0012888908386230469
iter:  7950 , fixed pt res:  tensor(0.4675) , ls:  1 , rel_err:  tensor(2.6191) , time:  0.0012793540954589844
iter:  8000 , fixed pt res:  tensor(0.4551) , ls:  1 , rel_err:  tensor(2.6183) , time:  0.0012764930725097656
iter:  8050 , fixed pt res:  tensor(0.4404) , ls:  1 , rel_err:  tensor(2.5896) , time:  0.001256704330444336
iter:  8100 , fixed pt res:  tensor(0.4297) , ls:  1 , rel_err:  tensor(2.5343) , time:  0.001226186752319336
iter:  8150 , fixed pt res:  tensor(0.4416) , ls:  1 , rel_err:  tensor(2.5102) , time:  0.0018901824951171875
iter:  8200 , fixed pt res:  tensor(0.4591) , ls:  1 , rel_err:  tensor(2.5327) , time:  0.0012922286987304688
iter:  8250 , fixed pt res:  tensor(0.4611) , ls:  1 , rel_err:  tensor(2.5747) , time:  0.0012788772583007812
iter:  8300 , fixed pt res:  tensor(0.4653) , ls:  1 , rel_err:  tensor(2.5335) , time:  0.0013616085052490234
iter:  8350 , fixed pt res:  tensor(0.4489) , ls:  1 , rel_err:  tensor(2.4418) , time:  0.0012996196746826172
iter:  8400 , fixed pt res:  tensor(0.4395) , ls:  1 , rel_err:  tensor(2.4722) , time:  0.0012898445129394531
iter:  8450 , fixed pt res:  tensor(0.4424) , ls:  1 , rel_err:  tensor(2.6027) , time:  0.0013608932495117188
iter:  8500 , fixed pt res:  tensor(0.4472) , ls:  1 , rel_err:  tensor(2.6112) , time:  0.0013065338134765625
iter:  8550 , fixed pt res:  tensor(0.4407) , ls:  1 , rel_err:  tensor(2.5731) , time:  0.0012345314025878906
iter:  8600 , fixed pt res:  tensor(0.4375) , ls:  1 , rel_err:  tensor(2.5217) , time:  0.0013403892517089844
iter:  8650 , fixed pt res:  tensor(0.4432) , ls:  1 , rel_err:  tensor(2.4493) , time:  0.0012629032135009766
iter:  8700 , fixed pt res:  tensor(0.4439) , ls:  1 , rel_err:  tensor(2.4075) , time:  0.0012497901916503906
iter:  8750 , fixed pt res:  tensor(0.4542) , ls:  1 , rel_err:  tensor(2.5005) , time:  0.0014042854309082031
iter:  8800 , fixed pt res:  tensor(0.4657) , ls:  1 , rel_err:  tensor(2.4351) , time:  0.001285552978515625
iter:  8850 , fixed pt res:  tensor(0.4505) , ls:  1 , rel_err:  tensor(2.4636) , time:  0.0013206005096435547
iter:  8900 , fixed pt res:  tensor(0.4628) , ls:  1 , rel_err:  tensor(2.4797) , time:  0.0013439655303955078
iter:  8950 , fixed pt res:  tensor(0.4455) , ls:  1 , rel_err:  tensor(2.5193) , time:  0.0012857913970947266
iter:  9000 , fixed pt res:  tensor(0.4604) , ls:  1 , rel_err:  tensor(2.4887) , time:  0.0012402534484863281
iter:  9050 , fixed pt res:  tensor(0.4510) , ls:  1 , rel_err:  tensor(2.4719) , time:  0.0016787052154541016
iter:  9100 , fixed pt res:  tensor(0.4549) , ls:  1 , rel_err:  tensor(2.5755) , time:  0.0013275146484375
iter:  9150 , fixed pt res:  tensor(0.4400) , ls:  1 , rel_err:  tensor(2.5101) , time:  0.0013310909271240234
iter:  9200 , fixed pt res:  tensor(0.4531) , ls:  1 , rel_err:  tensor(2.5554) , time:  0.001753091812133789
iter:  9250 , fixed pt res:  tensor(0.4451) , ls:  1 , rel_err:  tensor(2.5776) , time:  0.0015501976013183594
iter:  9300 , fixed pt res:  tensor(0.4418) , ls:  1 , rel_err:  tensor(2.5312) , time:  0.0012629032135009766
iter:  9350 , fixed pt res:  tensor(0.4620) , ls:  1 , rel_err:  tensor(2.5556) , time:  0.0012884140014648438
iter:  9400 , fixed pt res:  tensor(0.4415) , ls:  1 , rel_err:  tensor(2.6350) , time:  0.0013172626495361328
iter:  9450 , fixed pt res:  tensor(0.4341) , ls:  1 , rel_err:  tensor(2.7121) , time:  0.0012619495391845703
iter:  9500 , fixed pt res:  tensor(0.4464) , ls:  1 , rel_err:  tensor(2.7205) , time:  0.001325845718383789
iter:  9550 , fixed pt res:  tensor(0.4657) , ls:  1 , rel_err:  tensor(2.6309) , time:  0.0012462139129638672
iter:  9600 , fixed pt res:  tensor(0.4424) , ls:  1 , rel_err:  tensor(2.5812) , time:  0.0012547969818115234
iter:  9650 , fixed pt res:  tensor(0.4499) , ls:  1 , rel_err:  tensor(2.5472) , time:  0.0013298988342285156
iter:  9700 , fixed pt res:  tensor(0.4457) , ls:  1 , rel_err:  tensor(2.5958) , time:  0.0012683868408203125
iter:  9750 , fixed pt res:  tensor(0.4490) , ls:  1 , rel_err:  tensor(2.5856) , time:  0.0012929439544677734
iter:  9800 , fixed pt res:  tensor(0.4406) , ls:  1 , rel_err:  tensor(2.5817) , time:  0.0021517276763916016
iter:  9850 , fixed pt res:  tensor(0.4568) , ls:  1 , rel_err:  tensor(2.5975) , time:  0.0012955665588378906
iter:  9900 , fixed pt res:  tensor(0.4475) , ls:  1 , rel_err:  tensor(2.5970) , time:  0.0013790130615234375
iter:  9950 , fixed pt res:  tensor(0.4553) , ls:  1 , rel_err:  tensor(2.6232) , time:  0.001971006393432617
iter:  10000 , fixed pt res:  tensor(0.4416) , ls:  1 , rel_err:  tensor(2.7439) , time:  0.0013000965118408203
trial  1  finished after 13.767562627792358
norm(rel_err_hist_temp - rel_err_lmm_hist_sample_array[sample_index,:]) =  tensor(0.)
Number of Samples = 1000
iter:  50 , fixed pt res:  tensor(0.6200) , ls:  1 , rel_err:  tensor(15.2804) , time:  0.0013267993927001953
iter:  100 , fixed pt res:  tensor(0.2911) , ls:  1 , rel_err:  tensor(14.1417) , time:  0.001394510269165039
iter:  150 , fixed pt res:  tensor(0.1899) , ls:  1 , rel_err:  tensor(13.2434) , time:  0.0013072490692138672
iter:  200 , fixed pt res:  tensor(0.1576) , ls:  1 , rel_err:  tensor(12.3547) , time:  0.0013120174407958984
iter:  250 , fixed pt res:  tensor(0.1532) , ls:  1 , rel_err:  tensor(11.5270) , time:  0.0013427734375
iter:  300 , fixed pt res:  tensor(0.1436) , ls:  1 , rel_err:  tensor(10.7576) , time:  0.0013353824615478516
iter:  350 , fixed pt res:  tensor(0.1472) , ls:  1 , rel_err:  tensor(10.0357) , time:  0.0013246536254882812
iter:  400 , fixed pt res:  tensor(0.1415) , ls:  1 , rel_err:  tensor(9.3400) , time:  0.0013191699981689453
iter:  450 , fixed pt res:  tensor(0.1464) , ls:  1 , rel_err:  tensor(8.6775) , time:  0.0013461112976074219
iter:  500 , fixed pt res:  tensor(0.1451) , ls:  1 , rel_err:  tensor(8.0616) , time:  0.0013186931610107422
iter:  550 , fixed pt res:  tensor(0.1401) , ls:  1 , rel_err:  tensor(7.5455) , time:  0.0013468265533447266
iter:  600 , fixed pt res:  tensor(0.1464) , ls:  1 , rel_err:  tensor(7.0520) , time:  0.0013208389282226562
iter:  650 , fixed pt res:  tensor(0.1461) , ls:  1 , rel_err:  tensor(6.6107) , time:  0.0013234615325927734
iter:  700 , fixed pt res:  tensor(0.1416) , ls:  1 , rel_err:  tensor(6.1561) , time:  0.0013158321380615234
iter:  750 , fixed pt res:  tensor(0.1482) , ls:  1 , rel_err:  tensor(5.7341) , time:  0.0013077259063720703
iter:  800 , fixed pt res:  tensor(0.1493) , ls:  1 , rel_err:  tensor(5.3443) , time:  0.0013120174407958984
iter:  850 , fixed pt res:  tensor(0.1430) , ls:  1 , rel_err:  tensor(4.9980) , time:  0.001325845718383789
iter:  900 , fixed pt res:  tensor(0.1422) , ls:  1 , rel_err:  tensor(4.6833) , time:  0.0013217926025390625
iter:  950 , fixed pt res:  tensor(0.1458) , ls:  1 , rel_err:  tensor(4.3519) , time:  0.0013098716735839844
iter:  1000 , fixed pt res:  tensor(0.1456) , ls:  1 , rel_err:  tensor(4.0684) , time:  0.0013206005096435547
iter:  1050 , fixed pt res:  tensor(0.1501) , ls:  1 , rel_err:  tensor(3.8478) , time:  0.0013120174407958984
iter:  1100 , fixed pt res:  tensor(0.1433) , ls:  1 , rel_err:  tensor(3.5528) , time:  0.0013175010681152344
iter:  1150 , fixed pt res:  tensor(0.1389) , ls:  1 , rel_err:  tensor(3.3633) , time:  0.0013332366943359375
iter:  1200 , fixed pt res:  tensor(0.1424) , ls:  1 , rel_err:  tensor(3.1406) , time:  0.0016374588012695312
iter:  1250 , fixed pt res:  tensor(0.1448) , ls:  1 , rel_err:  tensor(2.9578) , time:  0.0013070106506347656
iter:  1300 , fixed pt res:  tensor(0.1426) , ls:  1 , rel_err:  tensor(2.8013) , time:  0.001321554183959961
iter:  1350 , fixed pt res:  tensor(0.1404) , ls:  1 , rel_err:  tensor(2.6524) , time:  0.0013594627380371094
iter:  1400 , fixed pt res:  tensor(0.1420) , ls:  1 , rel_err:  tensor(2.4768) , time:  0.0013468265533447266
iter:  1450 , fixed pt res:  tensor(0.1411) , ls:  1 , rel_err:  tensor(2.3400) , time:  0.0013494491577148438
iter:  1500 , fixed pt res:  tensor(0.1338) , ls:  1 , rel_err:  tensor(2.1864) , time:  0.0013415813446044922
iter:  1550 , fixed pt res:  tensor(0.1412) , ls:  1 , rel_err:  tensor(2.0526) , time:  0.0012879371643066406
iter:  1600 , fixed pt res:  tensor(0.1430) , ls:  1 , rel_err:  tensor(1.9605) , time:  0.0013277530670166016
iter:  1650 , fixed pt res:  tensor(0.1379) , ls:  1 , rel_err:  tensor(1.8693) , time:  0.0013251304626464844
iter:  1700 , fixed pt res:  tensor(0.1475) , ls:  1 , rel_err:  tensor(1.7646) , time:  0.0013089179992675781
iter:  1750 , fixed pt res:  tensor(0.1407) , ls:  1 , rel_err:  tensor(1.6703) , time:  0.0013682842254638672
iter:  1800 , fixed pt res:  tensor(0.1371) , ls:  1 , rel_err:  tensor(1.5463) , time:  0.0013401508331298828
iter:  1850 , fixed pt res:  tensor(0.1418) , ls:  1 , rel_err:  tensor(1.4633) , time:  0.0013346672058105469
iter:  1900 , fixed pt res:  tensor(0.1449) , ls:  1 , rel_err:  tensor(1.3959) , time:  0.001359701156616211
iter:  1950 , fixed pt res:  tensor(0.1457) , ls:  1 , rel_err:  tensor(1.3546) , time:  0.0013265609741210938
iter:  2000 , fixed pt res:  tensor(0.1362) , ls:  1 , rel_err:  tensor(1.2826) , time:  0.0013346672058105469
iter:  2050 , fixed pt res:  tensor(0.1465) , ls:  1 , rel_err:  tensor(1.2340) , time:  0.00138092041015625
iter:  2100 , fixed pt res:  tensor(0.1402) , ls:  1 , rel_err:  tensor(1.1610) , time:  0.0013661384582519531
iter:  2150 , fixed pt res:  tensor(0.1354) , ls:  1 , rel_err:  tensor(1.1367) , time:  0.0012798309326171875
iter:  2200 , fixed pt res:  tensor(0.1405) , ls:  1 , rel_err:  tensor(1.1296) , time:  0.001314401626586914
iter:  2250 , fixed pt res:  tensor(0.1388) , ls:  1 , rel_err:  tensor(1.0809) , time:  0.0013225078582763672
iter:  2300 , fixed pt res:  tensor(0.1445) , ls:  1 , rel_err:  tensor(1.0908) , time:  0.0013470649719238281
iter:  2350 , fixed pt res:  tensor(0.1458) , ls:  1 , rel_err:  tensor(1.0808) , time:  0.0013251304626464844
iter:  2400 , fixed pt res:  tensor(0.1373) , ls:  1 , rel_err:  tensor(1.0571) , time:  0.0016121864318847656
iter:  2450 , fixed pt res:  tensor(0.1444) , ls:  1 , rel_err:  tensor(1.0480) , time:  0.0013060569763183594
iter:  2500 , fixed pt res:  tensor(0.1383) , ls:  1 , rel_err:  tensor(1.0257) , time:  0.0013158321380615234
iter:  2550 , fixed pt res:  tensor(0.1422) , ls:  1 , rel_err:  tensor(0.9971) , time:  0.0013117790222167969
iter:  2600 , fixed pt res:  tensor(0.1411) , ls:  1 , rel_err:  tensor(0.9783) , time:  0.001316070556640625
iter:  2650 , fixed pt res:  tensor(0.1514) , ls:  1 , rel_err:  tensor(0.9526) , time:  0.0013360977172851562
iter:  2700 , fixed pt res:  tensor(0.1430) , ls:  1 , rel_err:  tensor(0.9029) , time:  0.001321554183959961
iter:  2750 , fixed pt res:  tensor(0.1389) , ls:  1 , rel_err:  tensor(0.8754) , time:  0.0013165473937988281
iter:  2800 , fixed pt res:  tensor(0.1417) , ls:  1 , rel_err:  tensor(0.8571) , time:  0.0013239383697509766
iter:  2850 , fixed pt res:  tensor(0.1427) , ls:  1 , rel_err:  tensor(0.8450) , time:  0.0015385150909423828
iter:  2900 , fixed pt res:  tensor(0.1392) , ls:  1 , rel_err:  tensor(0.8497) , time:  0.0013189315795898438
iter:  2950 , fixed pt res:  tensor(0.1374) , ls:  1 , rel_err:  tensor(0.8491) , time:  0.0013012886047363281
iter:  3000 , fixed pt res:  tensor(0.1415) , ls:  1 , rel_err:  tensor(0.8551) , time:  0.0013537406921386719
iter:  3050 , fixed pt res:  tensor(0.1446) , ls:  1 , rel_err:  tensor(0.8554) , time:  0.0013756752014160156
iter:  3100 , fixed pt res:  tensor(0.1413) , ls:  1 , rel_err:  tensor(0.8232) , time:  0.0013523101806640625
iter:  3150 , fixed pt res:  tensor(0.1374) , ls:  1 , rel_err:  tensor(0.7949) , time:  0.0013320446014404297
iter:  3200 , fixed pt res:  tensor(0.1393) , ls:  1 , rel_err:  tensor(0.8140) , time:  0.0013515949249267578
iter:  3250 , fixed pt res:  tensor(0.1373) , ls:  1 , rel_err:  tensor(0.7867) , time:  0.0013611316680908203
iter:  3300 , fixed pt res:  tensor(0.1399) , ls:  1 , rel_err:  tensor(0.7978) , time:  0.0013480186462402344
iter:  3350 , fixed pt res:  tensor(0.1426) , ls:  1 , rel_err:  tensor(0.7789) , time:  0.0013136863708496094
iter:  3400 , fixed pt res:  tensor(0.1410) , ls:  1 , rel_err:  tensor(0.7430) , time:  0.0014259815216064453
iter:  3450 , fixed pt res:  tensor(0.1390) , ls:  1 , rel_err:  tensor(0.7700) , time:  0.001317739486694336
iter:  3500 , fixed pt res:  tensor(0.1388) , ls:  1 , rel_err:  tensor(0.7716) , time:  0.0013110637664794922
iter:  3550 , fixed pt res:  tensor(0.1428) , ls:  1 , rel_err:  tensor(0.7730) , time:  0.0013706684112548828
iter:  3600 , fixed pt res:  tensor(0.1406) , ls:  1 , rel_err:  tensor(0.7427) , time:  0.0013256072998046875
iter:  3650 , fixed pt res:  tensor(0.1371) , ls:  1 , rel_err:  tensor(0.7262) , time:  0.001313924789428711
iter:  3700 , fixed pt res:  tensor(0.1399) , ls:  1 , rel_err:  tensor(0.7157) , time:  0.0013587474822998047
iter:  3750 , fixed pt res:  tensor(0.1349) , ls:  1 , rel_err:  tensor(0.7001) , time:  0.0013885498046875
iter:  3800 , fixed pt res:  tensor(0.1397) , ls:  1 , rel_err:  tensor(0.7020) , time:  0.001420736312866211
iter:  3850 , fixed pt res:  tensor(0.1456) , ls:  1 , rel_err:  tensor(0.7235) , time:  0.0013663768768310547
iter:  3900 , fixed pt res:  tensor(0.1397) , ls:  1 , rel_err:  tensor(0.7402) , time:  0.0013508796691894531
iter:  3950 , fixed pt res:  tensor(0.1425) , ls:  1 , rel_err:  tensor(0.7389) , time:  0.0018417835235595703
iter:  4000 , fixed pt res:  tensor(0.1436) , ls:  1 , rel_err:  tensor(0.7703) , time:  0.0014348030090332031
iter:  4050 , fixed pt res:  tensor(0.1337) , ls:  1 , rel_err:  tensor(0.7420) , time:  0.0013256072998046875
iter:  4100 , fixed pt res:  tensor(0.1397) , ls:  1 , rel_err:  tensor(0.7171) , time:  0.0014235973358154297
iter:  4150 , fixed pt res:  tensor(0.1453) , ls:  1 , rel_err:  tensor(0.7282) , time:  0.0013887882232666016
iter:  4200 , fixed pt res:  tensor(0.1385) , ls:  1 , rel_err:  tensor(0.7112) , time:  0.0013659000396728516
iter:  4250 , fixed pt res:  tensor(0.1352) , ls:  1 , rel_err:  tensor(0.7160) , time:  0.0013580322265625
iter:  4300 , fixed pt res:  tensor(0.1392) , ls:  1 , rel_err:  tensor(0.7100) , time:  0.0013353824615478516
iter:  4350 , fixed pt res:  tensor(0.1354) , ls:  1 , rel_err:  tensor(0.6963) , time:  0.001379251480102539
iter:  4400 , fixed pt res:  tensor(0.1355) , ls:  1 , rel_err:  tensor(0.7218) , time:  0.0014262199401855469
iter:  4450 , fixed pt res:  tensor(0.1435) , ls:  1 , rel_err:  tensor(0.7312) , time:  0.0013265609741210938
iter:  4500 , fixed pt res:  tensor(0.1412) , ls:  1 , rel_err:  tensor(0.7417) , time:  0.001306295394897461
iter:  4550 , fixed pt res:  tensor(0.1381) , ls:  1 , rel_err:  tensor(0.7293) , time:  0.001348257064819336
iter:  4600 , fixed pt res:  tensor(0.1438) , ls:  1 , rel_err:  tensor(0.7312) , time:  0.0012989044189453125
iter:  4650 , fixed pt res:  tensor(0.1411) , ls:  1 , rel_err:  tensor(0.7336) , time:  0.001322031021118164
iter:  4700 , fixed pt res:  tensor(0.1406) , ls:  1 , rel_err:  tensor(0.7303) , time:  0.0013480186462402344
iter:  4750 , fixed pt res:  tensor(0.1405) , ls:  1 , rel_err:  tensor(0.7269) , time:  0.0015642642974853516
iter:  4800 , fixed pt res:  tensor(0.1401) , ls:  1 , rel_err:  tensor(0.7372) , time:  0.0012898445129394531
iter:  4850 , fixed pt res:  tensor(0.1392) , ls:  1 , rel_err:  tensor(0.7530) , time:  0.0013241767883300781
iter:  4900 , fixed pt res:  tensor(0.1426) , ls:  1 , rel_err:  tensor(0.7443) , time:  0.0013282299041748047
iter:  4950 , fixed pt res:  tensor(0.1415) , ls:  1 , rel_err:  tensor(0.7270) , time:  0.0013480186462402344
iter:  5000 , fixed pt res:  tensor(0.1400) , ls:  1 , rel_err:  tensor(0.7463) , time:  0.002387523651123047
iter:  5050 , fixed pt res:  tensor(0.1389) , ls:  1 , rel_err:  tensor(0.7224) , time:  0.001508951187133789
iter:  5100 , fixed pt res:  tensor(0.1411) , ls:  1 , rel_err:  tensor(0.7328) , time:  0.0013210773468017578
iter:  5150 , fixed pt res:  tensor(0.1442) , ls:  1 , rel_err:  tensor(0.7383) , time:  0.0013167858123779297
iter:  5200 , fixed pt res:  tensor(0.1433) , ls:  1 , rel_err:  tensor(0.7134) , time:  0.0013079643249511719
iter:  5250 , fixed pt res:  tensor(0.1407) , ls:  1 , rel_err:  tensor(0.7117) , time:  0.0013206005096435547
iter:  5300 , fixed pt res:  tensor(0.1411) , ls:  1 , rel_err:  tensor(0.7247) , time:  0.001340627670288086
iter:  5350 , fixed pt res:  tensor(0.1418) , ls:  1 , rel_err:  tensor(0.7424) , time:  0.0013110637664794922
iter:  5400 , fixed pt res:  tensor(0.1416) , ls:  1 , rel_err:  tensor(0.7371) , time:  0.0013244152069091797
iter:  5450 , fixed pt res:  tensor(0.1425) , ls:  1 , rel_err:  tensor(0.7402) , time:  0.0013208389282226562
iter:  5500 , fixed pt res:  tensor(0.1421) , ls:  1 , rel_err:  tensor(0.7102) , time:  0.0013213157653808594
iter:  5550 , fixed pt res:  tensor(0.1387) , ls:  1 , rel_err:  tensor(0.7229) , time:  0.0013065338134765625
iter:  5600 , fixed pt res:  tensor(0.1406) , ls:  1 , rel_err:  tensor(0.6984) , time:  0.001306772232055664
iter:  5650 , fixed pt res:  tensor(0.1352) , ls:  1 , rel_err:  tensor(0.6988) , time:  0.001354217529296875
iter:  5700 , fixed pt res:  tensor(0.1405) , ls:  1 , rel_err:  tensor(0.7207) , time:  0.0013785362243652344
iter:  5750 , fixed pt res:  tensor(0.1384) , ls:  1 , rel_err:  tensor(0.7091) , time:  0.0013396739959716797
iter:  5800 , fixed pt res:  tensor(0.1351) , ls:  1 , rel_err:  tensor(0.7164) , time:  0.001318216323852539
iter:  5850 , fixed pt res:  tensor(0.1448) , ls:  1 , rel_err:  tensor(0.7072) , time:  0.0013074874877929688
iter:  5900 , fixed pt res:  tensor(0.1397) , ls:  1 , rel_err:  tensor(0.7139) , time:  0.0013289451599121094
iter:  5950 , fixed pt res:  tensor(0.1380) , ls:  1 , rel_err:  tensor(0.7105) , time:  0.0013780593872070312
iter:  6000 , fixed pt res:  tensor(0.1332) , ls:  1 , rel_err:  tensor(0.7013) , time:  0.0013089179992675781
iter:  6050 , fixed pt res:  tensor(0.1376) , ls:  1 , rel_err:  tensor(0.7175) , time:  0.0012922286987304688
iter:  6100 , fixed pt res:  tensor(0.1384) , ls:  1 , rel_err:  tensor(0.7122) , time:  0.0013475418090820312
iter:  6150 , fixed pt res:  tensor(0.1434) , ls:  1 , rel_err:  tensor(0.7273) , time:  0.0012962818145751953
iter:  6200 , fixed pt res:  tensor(0.1402) , ls:  1 , rel_err:  tensor(0.7510) , time:  0.001360177993774414
iter:  6250 , fixed pt res:  tensor(0.1441) , ls:  1 , rel_err:  tensor(0.7371) , time:  0.001276254653930664
iter:  6300 , fixed pt res:  tensor(0.1403) , ls:  1 , rel_err:  tensor(0.7318) , time:  0.0013804435729980469
iter:  6350 , fixed pt res:  tensor(0.1406) , ls:  1 , rel_err:  tensor(0.7090) , time:  0.0012996196746826172
iter:  6400 , fixed pt res:  tensor(0.1324) , ls:  1 , rel_err:  tensor(0.7121) , time:  0.001295328140258789
iter:  6450 , fixed pt res:  tensor(0.1376) , ls:  1 , rel_err:  tensor(0.7380) , time:  0.0013594627380371094
iter:  6500 , fixed pt res:  tensor(0.1328) , ls:  1 , rel_err:  tensor(0.7118) , time:  0.001314401626586914
iter:  6550 , fixed pt res:  tensor(0.1424) , ls:  1 , rel_err:  tensor(0.6888) , time:  0.0013668537139892578
iter:  6600 , fixed pt res:  tensor(0.1386) , ls:  1 , rel_err:  tensor(0.6796) , time:  0.0013508796691894531
iter:  6650 , fixed pt res:  tensor(0.1440) , ls:  1 , rel_err:  tensor(0.6752) , time:  0.0013365745544433594
iter:  6700 , fixed pt res:  tensor(0.1395) , ls:  1 , rel_err:  tensor(0.6984) , time:  0.0013523101806640625
iter:  6750 , fixed pt res:  tensor(0.1383) , ls:  1 , rel_err:  tensor(0.6993) , time:  0.0013263225555419922
iter:  6800 , fixed pt res:  tensor(0.1401) , ls:  1 , rel_err:  tensor(0.6909) , time:  0.0013306140899658203
iter:  6850 , fixed pt res:  tensor(0.1416) , ls:  1 , rel_err:  tensor(0.6914) , time:  0.0013840198516845703
iter:  6900 , fixed pt res:  tensor(0.1435) , ls:  1 , rel_err:  tensor(0.6869) , time:  0.0013513565063476562
iter:  6950 , fixed pt res:  tensor(0.1434) , ls:  1 , rel_err:  tensor(0.7111) , time:  0.0012936592102050781
iter:  7000 , fixed pt res:  tensor(0.1376) , ls:  1 , rel_err:  tensor(0.6947) , time:  0.0013315677642822266
iter:  7050 , fixed pt res:  tensor(0.1412) , ls:  1 , rel_err:  tensor(0.6983) , time:  0.0012993812561035156
iter:  7100 , fixed pt res:  tensor(0.1369) , ls:  1 , rel_err:  tensor(0.7071) , time:  0.0013473033905029297
iter:  7150 , fixed pt res:  tensor(0.1388) , ls:  1 , rel_err:  tensor(0.7188) , time:  0.0013604164123535156
iter:  7200 , fixed pt res:  tensor(0.1369) , ls:  1 , rel_err:  tensor(0.7448) , time:  0.0013170242309570312
iter:  7250 , fixed pt res:  tensor(0.1319) , ls:  1 , rel_err:  tensor(0.7368) , time:  0.0013206005096435547
iter:  7300 , fixed pt res:  tensor(0.1384) , ls:  1 , rel_err:  tensor(0.7549) , time:  0.0013265609741210938
iter:  7350 , fixed pt res:  tensor(0.1407) , ls:  1 , rel_err:  tensor(0.7553) , time:  0.0013527870178222656
iter:  7400 , fixed pt res:  tensor(0.1432) , ls:  1 , rel_err:  tensor(0.7565) , time:  0.0013544559478759766
iter:  7450 , fixed pt res:  tensor(0.1406) , ls:  1 , rel_err:  tensor(0.7565) , time:  0.0013697147369384766
iter:  7500 , fixed pt res:  tensor(0.1371) , ls:  1 , rel_err:  tensor(0.7860) , time:  0.0013203620910644531
iter:  7550 , fixed pt res:  tensor(0.1376) , ls:  1 , rel_err:  tensor(0.7820) , time:  0.0013303756713867188
iter:  7600 , fixed pt res:  tensor(0.1436) , ls:  1 , rel_err:  tensor(0.7663) , time:  0.0013909339904785156
iter:  7650 , fixed pt res:  tensor(0.1442) , ls:  1 , rel_err:  tensor(0.7851) , time:  0.0013649463653564453
iter:  7700 , fixed pt res:  tensor(0.1420) , ls:  1 , rel_err:  tensor(0.7762) , time:  0.0013232231140136719
iter:  7750 , fixed pt res:  tensor(0.1393) , ls:  1 , rel_err:  tensor(0.7632) , time:  0.0013110637664794922
iter:  7800 , fixed pt res:  tensor(0.1449) , ls:  1 , rel_err:  tensor(0.7567) , time:  0.0013320446014404297
iter:  7850 , fixed pt res:  tensor(0.1384) , ls:  1 , rel_err:  tensor(0.7296) , time:  0.001306772232055664
iter:  7900 , fixed pt res:  tensor(0.1419) , ls:  1 , rel_err:  tensor(0.7354) , time:  0.0013387203216552734
iter:  7950 , fixed pt res:  tensor(0.1383) , ls:  1 , rel_err:  tensor(0.7613) , time:  0.0013384819030761719
iter:  8000 , fixed pt res:  tensor(0.1419) , ls:  1 , rel_err:  tensor(0.7467) , time:  0.0013229846954345703
iter:  8050 , fixed pt res:  tensor(0.1452) , ls:  1 , rel_err:  tensor(0.7224) , time:  0.0013377666473388672
iter:  8100 , fixed pt res:  tensor(0.1356) , ls:  1 , rel_err:  tensor(0.7609) , time:  0.0013251304626464844
iter:  8150 , fixed pt res:  tensor(0.1378) , ls:  1 , rel_err:  tensor(0.7480) , time:  0.0013353824615478516
iter:  8200 , fixed pt res:  tensor(0.1463) , ls:  1 , rel_err:  tensor(0.7545) , time:  0.001306772232055664
iter:  8250 , fixed pt res:  tensor(0.1358) , ls:  1 , rel_err:  tensor(0.7275) , time:  0.0013210773468017578
iter:  8300 , fixed pt res:  tensor(0.1417) , ls:  1 , rel_err:  tensor(0.7226) , time:  0.0013229846954345703
iter:  8350 , fixed pt res:  tensor(0.1443) , ls:  1 , rel_err:  tensor(0.7175) , time:  0.0012788772583007812
iter:  8400 , fixed pt res:  tensor(0.1343) , ls:  1 , rel_err:  tensor(0.6982) , time:  0.001321554183959961
iter:  8450 , fixed pt res:  tensor(0.1443) , ls:  1 , rel_err:  tensor(0.6652) , time:  0.0013155937194824219
iter:  8500 , fixed pt res:  tensor(0.1454) , ls:  1 , rel_err:  tensor(0.6660) , time:  0.0013110637664794922
iter:  8550 , fixed pt res:  tensor(0.1424) , ls:  1 , rel_err:  tensor(0.6580) , time:  0.001312255859375
iter:  8600 , fixed pt res:  tensor(0.1389) , ls:  1 , rel_err:  tensor(0.6790) , time:  0.0013172626495361328
iter:  8650 , fixed pt res:  tensor(0.1384) , ls:  1 , rel_err:  tensor(0.6680) , time:  0.0013356208801269531
iter:  8700 , fixed pt res:  tensor(0.1440) , ls:  1 , rel_err:  tensor(0.6871) , time:  0.0013518333435058594
iter:  8750 , fixed pt res:  tensor(0.1415) , ls:  1 , rel_err:  tensor(0.6838) , time:  0.001293182373046875
iter:  8800 , fixed pt res:  tensor(0.1370) , ls:  1 , rel_err:  tensor(0.7017) , time:  0.0013251304626464844
iter:  8850 , fixed pt res:  tensor(0.1383) , ls:  1 , rel_err:  tensor(0.6953) , time:  0.001463174819946289
iter:  8900 , fixed pt res:  tensor(0.1370) , ls:  1 , rel_err:  tensor(0.6880) , time:  0.0013523101806640625
iter:  8950 , fixed pt res:  tensor(0.1372) , ls:  1 , rel_err:  tensor(0.7067) , time:  0.0013394355773925781
iter:  9000 , fixed pt res:  tensor(0.1405) , ls:  1 , rel_err:  tensor(0.6999) , time:  0.0013720989227294922
iter:  9050 , fixed pt res:  tensor(0.1458) , ls:  1 , rel_err:  tensor(0.7255) , time:  0.0013670921325683594
iter:  9100 , fixed pt res:  tensor(0.1417) , ls:  1 , rel_err:  tensor(0.7175) , time:  0.001369476318359375
iter:  9150 , fixed pt res:  tensor(0.1353) , ls:  1 , rel_err:  tensor(0.7578) , time:  0.0013086795806884766
iter:  9200 , fixed pt res:  tensor(0.1428) , ls:  1 , rel_err:  tensor(0.7319) , time:  0.0013167858123779297
iter:  9250 , fixed pt res:  tensor(0.1454) , ls:  1 , rel_err:  tensor(0.7346) , time:  0.00135040283203125
iter:  9300 , fixed pt res:  tensor(0.1510) , ls:  1 , rel_err:  tensor(0.7130) , time:  0.0013244152069091797
iter:  9350 , fixed pt res:  tensor(0.1436) , ls:  1 , rel_err:  tensor(0.7370) , time:  0.0013294219970703125
iter:  9400 , fixed pt res:  tensor(0.1460) , ls:  1 , rel_err:  tensor(0.7486) , time:  0.0013434886932373047
iter:  9450 , fixed pt res:  tensor(0.1476) , ls:  1 , rel_err:  tensor(0.7702) , time:  0.0013148784637451172
iter:  9500 , fixed pt res:  tensor(0.1363) , ls:  1 , rel_err:  tensor(0.7755) , time:  0.0013265609741210938
iter:  9550 , fixed pt res:  tensor(0.1386) , ls:  1 , rel_err:  tensor(0.8153) , time:  0.0013353824615478516
iter:  9600 , fixed pt res:  tensor(0.1428) , ls:  1 , rel_err:  tensor(0.8055) , time:  0.0013208389282226562
iter:  9650 , fixed pt res:  tensor(0.1402) , ls:  1 , rel_err:  tensor(0.8265) , time:  0.0013794898986816406
iter:  9700 , fixed pt res:  tensor(0.1369) , ls:  1 , rel_err:  tensor(0.8177) , time:  0.00130462646484375
iter:  9750 , fixed pt res:  tensor(0.1397) , ls:  1 , rel_err:  tensor(0.8112) , time:  0.0013453960418701172
iter:  9800 , fixed pt res:  tensor(0.1359) , ls:  1 , rel_err:  tensor(0.7822) , time:  0.0012867450714111328
iter:  9850 , fixed pt res:  tensor(0.1455) , ls:  1 , rel_err:  tensor(0.7920) , time:  0.0013346672058105469
iter:  9900 , fixed pt res:  tensor(0.1358) , ls:  1 , rel_err:  tensor(0.7939) , time:  0.0013093948364257812
iter:  9950 , fixed pt res:  tensor(0.1455) , ls:  1 , rel_err:  tensor(0.7798) , time:  0.0013077259063720703
iter:  10000 , fixed pt res:  tensor(0.1357) , ls:  1 , rel_err:  tensor(0.7740) , time:  0.0013074874877929688
trial  1  finished after 14.293195247650146
norm(rel_err_hist_temp - rel_err_lmm_hist_sample_array[sample_index,:]) =  tensor(0.)
Number of Samples = 10000
iter:  50 , fixed pt res:  tensor(0.5998) , ls:  1 , rel_err:  tensor(15.2757) , time:  0.0027074813842773438
iter:  100 , fixed pt res:  tensor(0.2594) , ls:  1 , rel_err:  tensor(14.1686) , time:  0.0026099681854248047
iter:  150 , fixed pt res:  tensor(0.1279) , ls:  1 , rel_err:  tensor(13.2236) , time:  0.002675771713256836
iter:  200 , fixed pt res:  tensor(0.0737) , ls:  1 , rel_err:  tensor(12.3519) , time:  0.0026006698608398438
iter:  250 , fixed pt res:  tensor(0.0603) , ls:  1 , rel_err:  tensor(11.5406) , time:  0.0026094913482666016
iter:  300 , fixed pt res:  tensor(0.0584) , ls:  1 , rel_err:  tensor(10.7629) , time:  0.0026221275329589844
iter:  350 , fixed pt res:  tensor(0.0541) , ls:  1 , rel_err:  tensor(10.0456) , time:  0.0026645660400390625
iter:  400 , fixed pt res:  tensor(0.0505) , ls:  1 , rel_err:  tensor(9.3683) , time:  0.0026504993438720703
iter:  450 , fixed pt res:  tensor(0.0519) , ls:  1 , rel_err:  tensor(8.7420) , time:  0.002627849578857422
iter:  500 , fixed pt res:  tensor(0.0506) , ls:  1 , rel_err:  tensor(8.1479) , time:  0.0025856494903564453
iter:  550 , fixed pt res:  tensor(0.0503) , ls:  1 , rel_err:  tensor(7.5810) , time:  0.002635955810546875
iter:  600 , fixed pt res:  tensor(0.0478) , ls:  1 , rel_err:  tensor(7.0501) , time:  0.0025839805603027344
iter:  650 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(6.5615) , time:  0.002607583999633789
iter:  700 , fixed pt res:  tensor(0.0459) , ls:  1 , rel_err:  tensor(6.1033) , time:  0.002591848373413086
iter:  750 , fixed pt res:  tensor(0.0472) , ls:  1 , rel_err:  tensor(5.6785) , time:  0.0026273727416992188
iter:  800 , fixed pt res:  tensor(0.0452) , ls:  1 , rel_err:  tensor(5.2926) , time:  0.0025899410247802734
iter:  850 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(4.9309) , time:  0.002669811248779297
iter:  900 , fixed pt res:  tensor(0.0449) , ls:  1 , rel_err:  tensor(4.5934) , time:  0.0026297569274902344
iter:  950 , fixed pt res:  tensor(0.0456) , ls:  1 , rel_err:  tensor(4.2891) , time:  0.002602815628051758
iter:  1000 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(4.0050) , time:  0.0025856494903564453
iter:  1050 , fixed pt res:  tensor(0.0463) , ls:  1 , rel_err:  tensor(3.7418) , time:  0.002600431442260742
iter:  1100 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(3.4833) , time:  0.0026674270629882812
iter:  1150 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(3.2542) , time:  0.0025968551635742188
iter:  1200 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(3.0267) , time:  0.0026426315307617188
iter:  1250 , fixed pt res:  tensor(0.0428) , ls:  1 , rel_err:  tensor(2.8332) , time:  0.0026497840881347656
iter:  1300 , fixed pt res:  tensor(0.0433) , ls:  1 , rel_err:  tensor(2.6491) , time:  0.0026335716247558594
iter:  1350 , fixed pt res:  tensor(0.0447) , ls:  1 , rel_err:  tensor(2.4943) , time:  0.0026102066040039062
iter:  1400 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(2.3482) , time:  0.002626180648803711
iter:  1450 , fixed pt res:  tensor(0.0456) , ls:  1 , rel_err:  tensor(2.1982) , time:  0.0025713443756103516
iter:  1500 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(2.0771) , time:  0.002585172653198242
iter:  1550 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(1.9496) , time:  0.002601146697998047
iter:  1600 , fixed pt res:  tensor(0.0462) , ls:  1 , rel_err:  tensor(1.8299) , time:  0.002844095230102539
iter:  1650 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(1.7146) , time:  0.002629995346069336
iter:  1700 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(1.6179) , time:  0.0026674270629882812
iter:  1750 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(1.5289) , time:  0.0026276111602783203
iter:  1800 , fixed pt res:  tensor(0.0455) , ls:  1 , rel_err:  tensor(1.4434) , time:  0.0029823780059814453
iter:  1850 , fixed pt res:  tensor(0.0435) , ls:  1 , rel_err:  tensor(1.3554) , time:  0.0026099681854248047
iter:  1900 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(1.2854) , time:  0.002686023712158203
iter:  1950 , fixed pt res:  tensor(0.0456) , ls:  1 , rel_err:  tensor(1.2185) , time:  0.0026140213012695312
iter:  2000 , fixed pt res:  tensor(0.0439) , ls:  1 , rel_err:  tensor(1.1540) , time:  0.002638578414916992
iter:  2050 , fixed pt res:  tensor(0.0430) , ls:  1 , rel_err:  tensor(1.0946) , time:  0.002608060836791992
iter:  2100 , fixed pt res:  tensor(0.0458) , ls:  1 , rel_err:  tensor(1.0330) , time:  0.00270843505859375
iter:  2150 , fixed pt res:  tensor(0.0421) , ls:  1 , rel_err:  tensor(0.9753) , time:  0.002655506134033203
iter:  2200 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(0.9208) , time:  0.002630949020385742
iter:  2250 , fixed pt res:  tensor(0.0432) , ls:  1 , rel_err:  tensor(0.8866) , time:  0.0026464462280273438
iter:  2300 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(0.8417) , time:  0.0026397705078125
iter:  2350 , fixed pt res:  tensor(0.0447) , ls:  1 , rel_err:  tensor(0.8105) , time:  0.002640962600708008
iter:  2400 , fixed pt res:  tensor(0.0445) , ls:  1 , rel_err:  tensor(0.7749) , time:  0.0026361942291259766
iter:  2450 , fixed pt res:  tensor(0.0456) , ls:  1 , rel_err:  tensor(0.7270) , time:  0.002605915069580078
iter:  2500 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(0.6931) , time:  0.0027260780334472656
iter:  2550 , fixed pt res:  tensor(0.0433) , ls:  1 , rel_err:  tensor(0.6712) , time:  0.0026073455810546875
iter:  2600 , fixed pt res:  tensor(0.0459) , ls:  1 , rel_err:  tensor(0.6482) , time:  0.0026667118072509766
iter:  2650 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(0.6209) , time:  0.0026335716247558594
iter:  2700 , fixed pt res:  tensor(0.0433) , ls:  1 , rel_err:  tensor(0.5894) , time:  0.0026459693908691406
iter:  2750 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.5727) , time:  0.0026636123657226562
iter:  2800 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.5474) , time:  0.0026328563690185547
iter:  2850 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(0.5290) , time:  0.0026111602783203125
iter:  2900 , fixed pt res:  tensor(0.0439) , ls:  1 , rel_err:  tensor(0.5104) , time:  0.002604961395263672
iter:  2950 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.4945) , time:  0.002595186233520508
iter:  3000 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.4735) , time:  0.002607107162475586
iter:  3050 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.4701) , time:  0.0026390552520751953
iter:  3100 , fixed pt res:  tensor(0.0452) , ls:  1 , rel_err:  tensor(0.4575) , time:  0.0027000904083251953
iter:  3150 , fixed pt res:  tensor(0.0436) , ls:  1 , rel_err:  tensor(0.4560) , time:  0.002655506134033203
iter:  3200 , fixed pt res:  tensor(0.0429) , ls:  1 , rel_err:  tensor(0.4378) , time:  0.0026350021362304688
iter:  3250 , fixed pt res:  tensor(0.0468) , ls:  1 , rel_err:  tensor(0.4312) , time:  0.0026488304138183594
iter:  3300 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.4260) , time:  0.002607583999633789
iter:  3350 , fixed pt res:  tensor(0.0435) , ls:  1 , rel_err:  tensor(0.4119) , time:  0.002643585205078125
iter:  3400 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.4009) , time:  0.0026383399963378906
iter:  3450 , fixed pt res:  tensor(0.0436) , ls:  1 , rel_err:  tensor(0.4003) , time:  0.002610921859741211
iter:  3500 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.3972) , time:  0.0026311874389648438
iter:  3550 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(0.3895) , time:  0.002611398696899414
iter:  3600 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.3673) , time:  0.0025801658630371094
iter:  3650 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.3589) , time:  0.002653837203979492
iter:  3700 , fixed pt res:  tensor(0.0442) , ls:  1 , rel_err:  tensor(0.3513) , time:  0.002604961395263672
iter:  3750 , fixed pt res:  tensor(0.0445) , ls:  1 , rel_err:  tensor(0.3462) , time:  0.00260162353515625
iter:  3800 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.3258) , time:  0.002612590789794922
iter:  3850 , fixed pt res:  tensor(0.0466) , ls:  1 , rel_err:  tensor(0.3177) , time:  0.0026407241821289062
iter:  3900 , fixed pt res:  tensor(0.0452) , ls:  1 , rel_err:  tensor(0.3112) , time:  0.0027043819427490234
iter:  3950 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.3034) , time:  0.0026886463165283203
iter:  4000 , fixed pt res:  tensor(0.0465) , ls:  1 , rel_err:  tensor(0.3008) , time:  0.002656698226928711
iter:  4050 , fixed pt res:  tensor(0.0429) , ls:  1 , rel_err:  tensor(0.2942) , time:  0.0026063919067382812
iter:  4100 , fixed pt res:  tensor(0.0462) , ls:  1 , rel_err:  tensor(0.2901) , time:  0.002576112747192383
iter:  4150 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(0.2870) , time:  0.002616405487060547
iter:  4200 , fixed pt res:  tensor(0.0442) , ls:  1 , rel_err:  tensor(0.2925) , time:  0.0025975704193115234
iter:  4250 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(0.2890) , time:  0.002580404281616211
iter:  4300 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(0.2853) , time:  0.0026237964630126953
iter:  4350 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(0.2807) , time:  0.0026149749755859375
iter:  4400 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2652) , time:  0.002762317657470703
iter:  4450 , fixed pt res:  tensor(0.0449) , ls:  1 , rel_err:  tensor(0.2658) , time:  0.0026183128356933594
iter:  4500 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(0.2650) , time:  0.0026106834411621094
iter:  4550 , fixed pt res:  tensor(0.0455) , ls:  1 , rel_err:  tensor(0.2632) , time:  0.0026085376739501953
iter:  4600 , fixed pt res:  tensor(0.0426) , ls:  1 , rel_err:  tensor(0.2559) , time:  0.002616405487060547
iter:  4650 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(0.2557) , time:  0.0026111602783203125
iter:  4700 , fixed pt res:  tensor(0.0439) , ls:  1 , rel_err:  tensor(0.2527) , time:  0.0027801990509033203
iter:  4750 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.2439) , time:  0.0026085376739501953
iter:  4800 , fixed pt res:  tensor(0.0438) , ls:  1 , rel_err:  tensor(0.2454) , time:  0.0027692317962646484
iter:  4850 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.2379) , time:  0.0026547908782958984
iter:  4900 , fixed pt res:  tensor(0.0420) , ls:  1 , rel_err:  tensor(0.2425) , time:  0.002617359161376953
iter:  4950 , fixed pt res:  tensor(0.0461) , ls:  1 , rel_err:  tensor(0.2371) , time:  0.0026412010192871094
iter:  5000 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.2339) , time:  0.0026237964630126953
iter:  5050 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2248) , time:  0.0026144981384277344
iter:  5100 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(0.2118) , time:  0.002590179443359375
iter:  5150 , fixed pt res:  tensor(0.0447) , ls:  1 , rel_err:  tensor(0.2202) , time:  0.0026094913482666016
iter:  5200 , fixed pt res:  tensor(0.0449) , ls:  1 , rel_err:  tensor(0.2171) , time:  0.002602100372314453
iter:  5250 , fixed pt res:  tensor(0.0436) , ls:  1 , rel_err:  tensor(0.2239) , time:  0.0026252269744873047
iter:  5300 , fixed pt res:  tensor(0.0428) , ls:  1 , rel_err:  tensor(0.2304) , time:  0.0026252269744873047
iter:  5350 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(0.2316) , time:  0.0027251243591308594
iter:  5400 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.2359) , time:  0.0026140213012695312
iter:  5450 , fixed pt res:  tensor(0.0434) , ls:  1 , rel_err:  tensor(0.2331) , time:  0.0025742053985595703
iter:  5500 , fixed pt res:  tensor(0.0427) , ls:  1 , rel_err:  tensor(0.2311) , time:  0.002604246139526367
iter:  5550 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.2384) , time:  0.002608060836791992
iter:  5600 , fixed pt res:  tensor(0.0442) , ls:  1 , rel_err:  tensor(0.2446) , time:  0.002630949020385742
iter:  5650 , fixed pt res:  tensor(0.0475) , ls:  1 , rel_err:  tensor(0.2419) , time:  0.002635955810546875
iter:  5700 , fixed pt res:  tensor(0.0433) , ls:  1 , rel_err:  tensor(0.2349) , time:  0.0026977062225341797
iter:  5750 , fixed pt res:  tensor(0.0447) , ls:  1 , rel_err:  tensor(0.2445) , time:  0.0026857852935791016
iter:  5800 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.2331) , time:  0.0026388168334960938
iter:  5850 , fixed pt res:  tensor(0.0459) , ls:  1 , rel_err:  tensor(0.2479) , time:  0.0027358531951904297
iter:  5900 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(0.2453) , time:  0.0026693344116210938
iter:  5950 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(0.2445) , time:  0.002724885940551758
iter:  6000 , fixed pt res:  tensor(0.0449) , ls:  1 , rel_err:  tensor(0.2450) , time:  0.002702474594116211
iter:  6050 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.2435) , time:  0.0026938915252685547
iter:  6100 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(0.2424) , time:  0.002642393112182617
iter:  6150 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.2368) , time:  0.002652883529663086
iter:  6200 , fixed pt res:  tensor(0.0432) , ls:  1 , rel_err:  tensor(0.2373) , time:  0.002664804458618164
iter:  6250 , fixed pt res:  tensor(0.0432) , ls:  1 , rel_err:  tensor(0.2439) , time:  0.0026662349700927734
iter:  6300 , fixed pt res:  tensor(0.0447) , ls:  1 , rel_err:  tensor(0.2335) , time:  0.002669095993041992
iter:  6350 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.2266) , time:  0.0026252269744873047
iter:  6400 , fixed pt res:  tensor(0.0447) , ls:  1 , rel_err:  tensor(0.2422) , time:  0.0026700496673583984
iter:  6450 , fixed pt res:  tensor(0.0449) , ls:  1 , rel_err:  tensor(0.2494) , time:  0.002619028091430664
iter:  6500 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.2411) , time:  0.002673625946044922
iter:  6550 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(0.2380) , time:  0.0026128292083740234
iter:  6600 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2402) , time:  0.0026497840881347656
iter:  6650 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.2401) , time:  0.002610921859741211
iter:  6700 , fixed pt res:  tensor(0.0461) , ls:  1 , rel_err:  tensor(0.2256) , time:  0.0026390552520751953
iter:  6750 , fixed pt res:  tensor(0.0429) , ls:  1 , rel_err:  tensor(0.2239) , time:  0.002607107162475586
iter:  6800 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.2263) , time:  0.002607583999633789
iter:  6850 , fixed pt res:  tensor(0.0439) , ls:  1 , rel_err:  tensor(0.2380) , time:  0.002641916275024414
iter:  6900 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.2354) , time:  0.0027132034301757812
iter:  6950 , fixed pt res:  tensor(0.0434) , ls:  1 , rel_err:  tensor(0.2344) , time:  0.0027217864990234375
iter:  7000 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.2373) , time:  0.002673625946044922
iter:  7050 , fixed pt res:  tensor(0.0425) , ls:  1 , rel_err:  tensor(0.2289) , time:  0.002610921859741211
iter:  7100 , fixed pt res:  tensor(0.0433) , ls:  1 , rel_err:  tensor(0.2301) , time:  0.002599477767944336
iter:  7150 , fixed pt res:  tensor(0.0442) , ls:  1 , rel_err:  tensor(0.2306) , time:  0.002654552459716797
iter:  7200 , fixed pt res:  tensor(0.0465) , ls:  1 , rel_err:  tensor(0.2249) , time:  0.002593994140625
iter:  7250 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.2225) , time:  0.0026268959045410156
iter:  7300 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.2336) , time:  0.0026633739471435547
iter:  7350 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.2387) , time:  0.0026025772094726562
iter:  7400 , fixed pt res:  tensor(0.0445) , ls:  1 , rel_err:  tensor(0.2361) , time:  0.0027959346771240234
iter:  7450 , fixed pt res:  tensor(0.0447) , ls:  1 , rel_err:  tensor(0.2402) , time:  0.0026085376739501953
iter:  7500 , fixed pt res:  tensor(0.0437) , ls:  1 , rel_err:  tensor(0.2379) , time:  0.002661466598510742
iter:  7550 , fixed pt res:  tensor(0.0419) , ls:  1 , rel_err:  tensor(0.2360) , time:  0.0026352405548095703
iter:  7600 , fixed pt res:  tensor(0.0429) , ls:  1 , rel_err:  tensor(0.2377) , time:  0.0026412010192871094
iter:  7650 , fixed pt res:  tensor(0.0430) , ls:  1 , rel_err:  tensor(0.2376) , time:  0.00267791748046875
iter:  7700 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2417) , time:  0.0025963783264160156
iter:  7750 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(0.2421) , time:  0.002657651901245117
iter:  7800 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.2403) , time:  0.0026264190673828125
iter:  7850 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.2342) , time:  0.002609729766845703
iter:  7900 , fixed pt res:  tensor(0.0445) , ls:  1 , rel_err:  tensor(0.2360) , time:  0.0026090145111083984
iter:  7950 , fixed pt res:  tensor(0.0449) , ls:  1 , rel_err:  tensor(0.2312) , time:  0.002658843994140625
iter:  8000 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.2386) , time:  0.0026226043701171875
iter:  8050 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(0.2389) , time:  0.002712249755859375
iter:  8100 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.2428) , time:  0.0026378631591796875
iter:  8150 , fixed pt res:  tensor(0.0463) , ls:  1 , rel_err:  tensor(0.2545) , time:  0.0025916099548339844
iter:  8200 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(0.2502) , time:  0.002643585205078125
iter:  8250 , fixed pt res:  tensor(0.0441) , ls:  1 , rel_err:  tensor(0.2474) , time:  0.0025789737701416016
iter:  8300 , fixed pt res:  tensor(0.0450) , ls:  1 , rel_err:  tensor(0.2367) , time:  0.0025835037231445312
iter:  8350 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.2376) , time:  0.0025784969329833984
iter:  8400 , fixed pt res:  tensor(0.0435) , ls:  1 , rel_err:  tensor(0.2338) , time:  0.0026235580444335938
iter:  8450 , fixed pt res:  tensor(0.0427) , ls:  1 , rel_err:  tensor(0.2361) , time:  0.002642393112182617
iter:  8500 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.2343) , time:  0.0026869773864746094
iter:  8550 , fixed pt res:  tensor(0.0440) , ls:  1 , rel_err:  tensor(0.2282) , time:  0.0026705265045166016
iter:  8600 , fixed pt res:  tensor(0.0432) , ls:  1 , rel_err:  tensor(0.2310) , time:  0.002638578414916992
iter:  8650 , fixed pt res:  tensor(0.0433) , ls:  1 , rel_err:  tensor(0.2331) , time:  0.002623319625854492
iter:  8700 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.2308) , time:  0.0025844573974609375
iter:  8750 , fixed pt res:  tensor(0.0449) , ls:  1 , rel_err:  tensor(0.2349) , time:  0.0025954246520996094
iter:  8800 , fixed pt res:  tensor(0.0431) , ls:  1 , rel_err:  tensor(0.2365) , time:  0.002594470977783203
iter:  8850 , fixed pt res:  tensor(0.0460) , ls:  1 , rel_err:  tensor(0.2305) , time:  0.0025708675384521484
iter:  8900 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.2322) , time:  0.002640247344970703
iter:  8950 , fixed pt res:  tensor(0.0439) , ls:  1 , rel_err:  tensor(0.2321) , time:  0.00260162353515625
iter:  9000 , fixed pt res:  tensor(0.0455) , ls:  1 , rel_err:  tensor(0.2332) , time:  0.002644062042236328
iter:  9050 , fixed pt res:  tensor(0.0439) , ls:  1 , rel_err:  tensor(0.2342) , time:  0.002650022506713867
iter:  9100 , fixed pt res:  tensor(0.0431) , ls:  1 , rel_err:  tensor(0.2352) , time:  0.0026323795318603516
iter:  9150 , fixed pt res:  tensor(0.0419) , ls:  1 , rel_err:  tensor(0.2377) , time:  0.0025916099548339844
iter:  9200 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2383) , time:  0.002780437469482422
iter:  9250 , fixed pt res:  tensor(0.0446) , ls:  1 , rel_err:  tensor(0.2482) , time:  0.0026121139526367188
iter:  9300 , fixed pt res:  tensor(0.0454) , ls:  1 , rel_err:  tensor(0.2536) , time:  0.002669095993041992
iter:  9350 , fixed pt res:  tensor(0.0449) , ls:  1 , rel_err:  tensor(0.2583) , time:  0.0026013851165771484
iter:  9400 , fixed pt res:  tensor(0.0443) , ls:  1 , rel_err:  tensor(0.2538) , time:  0.002645730972290039
iter:  9450 , fixed pt res:  tensor(0.0442) , ls:  1 , rel_err:  tensor(0.2458) , time:  0.0026335716247558594
iter:  9500 , fixed pt res:  tensor(0.0445) , ls:  1 , rel_err:  tensor(0.2508) , time:  0.0026450157165527344
iter:  9550 , fixed pt res:  tensor(0.0448) , ls:  1 , rel_err:  tensor(0.2519) , time:  0.0026204586029052734
iter:  9600 , fixed pt res:  tensor(0.0453) , ls:  1 , rel_err:  tensor(0.2448) , time:  0.0026938915252685547
iter:  9650 , fixed pt res:  tensor(0.0452) , ls:  1 , rel_err:  tensor(0.2445) , time:  0.002580404281616211
iter:  9700 , fixed pt res:  tensor(0.0435) , ls:  1 , rel_err:  tensor(0.2445) , time:  0.0026569366455078125
iter:  9750 , fixed pt res:  tensor(0.0444) , ls:  1 , rel_err:  tensor(0.2461) , time:  0.0026009082794189453
iter:  9800 , fixed pt res:  tensor(0.0457) , ls:  1 , rel_err:  tensor(0.2492) , time:  0.0026717185974121094
iter:  9850 , fixed pt res:  tensor(0.0439) , ls:  1 , rel_err:  tensor(0.2451) , time:  0.002624034881591797
iter:  9900 , fixed pt res:  tensor(0.0445) , ls:  1 , rel_err:  tensor(0.2470) , time:  0.0027098655700683594
iter:  9950 , fixed pt res:  tensor(0.0451) , ls:  1 , rel_err:  tensor(0.2420) , time:  0.002642393112182617
iter:  10000 , fixed pt res:  tensor(0.0422) , ls:  1 , rel_err:  tensor(0.2330) , time:  0.0026102066040039062
trial  1  finished after 27.398865699768066
norm(rel_err_hist_temp - rel_err_lmm_hist_sample_array[sample_index,:]) =  tensor(0.)

In [ ]:

Copied!





# plotting parameters
title_fontsize = 22
fontsize       = 20
fig1 = plt.figure()
my_blue = '#1f77b4'
my_orange = '#F97306'

fig1 = plt.figure()
plt.style.use('seaborn-whitegrid')
ax = plt.axes()

ax.semilogy(rel_err_lmm_hist_sample_array[0,:], '-',linewidth=5, color=my_blue)
ax.semilogy(rel_err_lmm_hist_sample_array[1,:], '-', linewidth=5, color='tab:green')
ax.semilogy(rel_err_lmm_hist_sample_array[2,:], linewidth=5, color='tab:purple')
ax.semilogy(rel_err_lmm_hist_sample_array[3,:], linewidth=5, color='tab:cyan')
ax.semilogy(rel_err_hist_GD, linewidth=5, color='k')

# ax.set_xlabel("x-axis", fontsize=title_fontsize)
ax.legend(['HJ-MM 1 Sample', 'HJ-MM 100 Samples', 'HJ-MM 1K Samples', 'HJ-MM 10K Samples', 'Gradient Descent'],fontsize=fontsize, bbox_to_anchor=(1.0, 1.0), loc='upper left')
ax.tick_params(labelsize=fontsize, which='both', direction='in')
save_str = 'ADMM_noisy_sample_comparison.pdf'
fig1.savefig(save_str, dpi=300 , bbox_inches="tight", pad_inches=0.0)

title_str = 'delta =  ' + str(delta_for_varying_samples)
ax.set_title(title_str, fontsize=fontsize)
# plotting parameters
title_fontsize = 22
fontsize       = 20
fig1 = plt.figure()
my_blue = '#1f77b4'
my_orange = '#F97306'

fig1 = plt.figure()
plt.style.use('seaborn-whitegrid')
ax = plt.axes()

ax.semilogy(rel_err_lmm_hist_sample_array[0,:], '-',linewidth=5, color=my_blue)
ax.semilogy(rel_err_lmm_hist_sample_array[1,:], '-', linewidth=5, color='tab:green')
ax.semilogy(rel_err_lmm_hist_sample_array[2,:], linewidth=5, color='tab:purple')
ax.semilogy(rel_err_lmm_hist_sample_array[3,:], linewidth=5, color='tab:cyan')
ax.semilogy(rel_err_hist_GD, linewidth=5, color='k')

# ax.set_xlabel("x-axis", fontsize=title_fontsize)
ax.legend(['HJ-MM 1 Sample', 'HJ-MM 100 Samples', 'HJ-MM 1K Samples', 'HJ-MM 10K Samples', 'Gradient Descent'],fontsize=fontsize, bbox_to_anchor=(1.0, 1.0), loc='upper left')
ax.tick_params(labelsize=fontsize, which='both', direction='in')
save_str = 'ADMM_noisy_sample_comparison.pdf'
fig1.savefig(save_str, dpi=300 , bbox_inches="tight", pad_inches=0.0)

title_str = 'delta =  ' + str(delta_for_varying_samples)
ax.set_title(title_str, fontsize=fontsize)

<ipython-input-10-f1d11af914b3>:9: MatplotlibDeprecationWarning: The seaborn styles shipped by Matplotlib are deprecated since 3.6, as they no longer correspond to the styles shipped by seaborn. However, they will remain available as 'seaborn-v0_8-<style>'. Alternatively, directly use the seaborn API instead.
  plt.style.use('seaborn-whitegrid')

Out[ ]:

Text(0.5, 1.0, 'delta =  10')

<Figure size 640x480 with 0 Axes>