Stochastic Gradient Descent

In machine learning , Gradient Descent is a an optimization algorithm used to minimize a function by iteratively moving towards the steepest descent as defined by the negative of the gradient. Here’s how it works, you start with a random point on the function you’re trying to minimize, for example a random starting point on the mountain. Then, you calculate the gradient (slope) of the function at that point. In the mountain analogy, this is like looking around you to find the steepest slope. Once you know the direction, you take a step downhill in that direction, and then you calculate the gradient again. Repeat this process until you reach the bottom. The size of each step is determined by the learning rate. However, if the learning rate is too small, it might take a long time to reach the bottom. If it’s too large, you might overshoot the lowest point. Finding the right balance is key.

The ‘Stochastic’ in Stochastic Gradient Descent

Stochastic Gradient Descent (SGD) adds a twist to the traditional gradient descent approach. The term ‘stochastic’ refers to a system or process that is linked with a random probability.

In traditional batch gradient descent, you calculate the gradient of the loss function with respect to the parameters for the entire training set. As you can imagine, for large datasets, this can be quite computationally intensive and time-consuming. This is where SGD comes into play. Instead of using the entire dataset to calculate the gradient, SGD randomly selects just one data point (or a few data points) to compute the gradient in each iteration.

Think of this process as if you were again descending a mountain, but this time in thick fog with limited visibility. Rather than viewing the entire landscape to decide your next step, you make your decision based on where your foot lands next. This step is small and random, but it’s repeated many times, each time adjusting your path slightly in response to the immediate terrain under your feet.

Steps in SGD

Initialization (Step 1)

First, you initialize the parameters (weights) of your model. This can be done randomly or by some other initialization technique. The starting point for SGD is crucial as it influences the path the algorithm will take.

Random Selection (Step 2)

In each iteration of the training process, SGD randomly selects a single data point (or a small batch of data points) from the entire dataset. This randomness is what makes it ‘stochastic’.

Compute the Gradient (Step 3)

Calculate the gradient of the loss function, but only for the randomly selected data point(s). The gradient is a vector that points in the direction of the steepest increase of the loss function. In the context of SGD, it tells you how to tweak the parameters to make the model more accurate for that particular data point.

\[ \nabla_{\theta} J(\theta) = \frac{\partial J(\theta)}{\partial \theta} \]

• \(\nabla_{\theta} J(\theta)\): This represents the gradient of the function \(J(\theta)\) with respect to the parameter \(\theta\).

• \(\frac{\partial J(\theta)}{\partial \theta}\): This is the partial derivative of the function \(J(\theta)\) with respect to \(\theta\).

In essence, the gradient \(\nabla_{\theta} J(\theta)\) is a vector of partial derivatives, which indicates the direction and rate of the fastest increase of the function \(J(\theta)\). This iteratively adjust \(\theta\) to minimize \(J(\theta)\).

Update the Parameters (Step 4)

Adjust the model parameters in the opposite direction of the gradient. Here’s where the learning rate η plays a crucial role. The formula for updating each parameter is:

\[ \theta_{\text{new}} = \theta_{\text{old}} - \eta \cdot \nabla_{\theta} J(\theta) \]

This equation represents one iteration of the gradient descent algorithm, which is used to minimize a function \(J(\theta)\).

• \(\theta_{\text{new}}\): The updated value of the parameter \(\theta\) after one iteration of gradient descent.

• \(\theta_{\text{old}}\): The current value of the parameter \(\theta\).

• \(\eta\): The learning rate, a positive scalar that controls the step size of each iteration. It determines how much \(\theta\) is adjusted in each step.

• \(\nabla_{\theta} J(\theta)\): The gradient of the function \(J(\theta)\) with respect to \(\theta\), which indicates the direction and rate of the steepest increase of the function.

In the context of minimizing \(J(\theta)\), the gradient descent algorithm updates the parameter \(\theta\) by moving it in the opposite direction of the gradient. This process is repeated iteratively to find the value of \(\theta\) that minimizes \(J(\theta)\).

The learning rate determines the size of the steps you take towards the minimum. If it’s too small, the algorithm will be slow; if it’s too large, you might overshoot the minimum.

Repeat until convergence (Step 5)

Repeat steps 2 to 4 for a set number of iterations or until the model performance stops improving. Each iteration provides a slightly updated model. Ideally, after many iterations, SGD converges to a set of parameters that minimize the loss function, although due to its stochastic nature, the path to convergence is not as smooth and may oscillate around the minimum.

Understanding Learning Rate

One of the most crucial hyperparameters in the Stochastic Gradient Descent (SGD) algorithm is the learning rate. This parameter can significantly impact the performance and convergence of the model. Understanding and choosing the right learning rate is a vital step in effectively employing SGD.

What is Learning Rate?

At this point you should have an idea of what learning rate is, but let’s better define it for clarity. The learning rate in SGD determines the size of the steps the algorithm takes towards the minimum of the loss function. It’s a scalar that scales the gradient, dictating how much the weights in the model should be adjusted during each update. If you visualize the loss function as a valley, the learning rate decides how big a step you take with each iteration as you walk down the valley.

Too High Learning Rate

If the learning rate is too high, the steps taken might be too large. This can lead to overshooting the minimum, causing the algorithm to diverge or oscillate wildly without finding a stable point. Think of it as taking leaps in the valley and possibly jumping over the lowest point back and forth.

Too Low Learning Rate

On the other hand, a very low learning rate leads to extremely small steps. While this might sound safe, it significantly slows down the convergence process. In a worst-case scenario, the algorithm might get stuck in a local minimum or even stop improving before reaching the minimum. Imagine moving so slowly down the valley that you either get stuck or it takes an impractically long time to reach the bottom.

Finding the Right Balance

The ideal learning rate is neither too high nor too low but strikes a balance, allowing the algorithm to converge efficiently to the global minimum. Typically, the learning rate is chosen through experimentation and is often set to decrease over time. This approach is called learning rate annealing or scheduling.

Learning Rate Scheduling

Learning rate scheduling involves adjusting the learning rate over time. Common strategies include:

• Time-Based Decay: The learning rate decreases over each update.

• Step Decay: Reduce the learning rate by some factor after a certain number of epochs.

• Exponential Decay: Decrease the learning rate exponentially.

• Adaptive Learning Rate: Methods like AdaGrad, RMSProp, and Adam adjust the learning rate automatically during training.

# Basic Libraries
import numpy as np
import pandas as pd

# Load Data
from sklearn.datasets import load_diabetes

# Data Visualization
import matplotlib.pyplot as plt
import seaborn as sns

# Model Fine Tuning
import optuna

# Filter Warnings
import warnings
warnings.filterwarnings('ignore')

SGD Regressor Class

class SGDRegressor:
    def __init__(self, learning_rate=0.01, epochs=100, batch_size=1, reg=None, reg_param=0.0):
        """
        Constructor for the SGDRegressor.

        Parameters:
        learning_rate (float): The step size used in each update.
        epochs (int): Number of passes over the training dataset.
        batch_size (int): Number of samples to be used in each batch.
        reg (str): Type of regularization ('l1' or 'l2'); None if no regularization.
        reg_param (float): Regularization parameter.

        The weights and bias are initialized as None and will be set during the fit method.
        """
        self.learning_rate = learning_rate
        self.epochs = epochs
        self.batch_size = batch_size
        self.reg = reg
        self.reg_param = reg_param
        self.weights = None
        self.bias = None

    def fit(self, X, y):
        """
        Fits the SGDRegressor to the training data.

        Parameters:
        X (numpy.ndarray): Training data, shape (m_samples, n_features).
        y (numpy.ndarray): Target values, shape (m_samples,).

        This method initializes the weights and bias, and then updates them over a number of epochs.
        """
        m, n = X.shape  # m is number of samples, n is number of features
        self.weights = np.zeros(n)
        self.bias = 0

        for _ in range(self.epochs):
            indices = np.random.permutation(m)
            X_shuffled = X[indices]
            y_shuffled = y[indices]

            for i in range(0, m, self.batch_size):
                X_batch = X_shuffled[i:i+self.batch_size]
                y_batch = y_shuffled[i:i+self.batch_size]

                gradient_w = -2 * np.dot(X_batch.T, (y_batch - np.dot(X_batch, self.weights) - self.bias)) / self.batch_size
                gradient_b = -2 * np.sum(y_batch - np.dot(X_batch, self.weights) - self.bias) / self.batch_size

                if self.reg == 'l1':
                    gradient_w += self.reg_param * np.sign(self.weights)
                elif self.reg == 'l2':
                    gradient_w += self.reg_param * self.weights

                self.weights -= self.learning_rate * gradient_w
                self.bias -= self.learning_rate * gradient_b

    def predict(self, X):
        """
        Predicts the target values using the linear model.

        Parameters:
        X (numpy.ndarray): Data for which to predict target values.

        Returns:
        numpy.ndarray: Predicted target values.
        """
        return np.dot(X, self.weights) + self.bias

    def compute_loss(self, X, y):
        """
        Computes the loss of the model.

        Parameters:
        X (numpy.ndarray): The input data.
        y (numpy.ndarray): The true target values.

        Returns:
        float: The computed loss value.
        """
        return (np.mean((y - self.predict(X)) ** 2) + self._get_regularization_loss()) ** 0.5

    def _get_regularization_loss(self):
        """
        Computes the regularization loss based on the regularization type.

        Returns:
        float: The regularization loss.
        """
        if self.reg == 'l1':
            return self.reg_param * np.sum(np.abs(self.weights))
        elif self.reg == 'l2':
            return self.reg_param * np.sum(self.weights ** 2)
        else:
            return 0

    def get_weights(self):
        """
        Returns the weights of the model.

        Returns:
        numpy.ndarray: The weights of the linear model.
        """
        return self.weights

The fit function performs the calculation of the gradient of the loss function [using Mean Squared Error (MSE)] with respect to the weights in a linear regression model.

gradient_w = -2 * np.dot(X_batch.T, (y_batch - np.dot(X_batch, self.weights) - self.bias)) / self.batch_size

1. Linear Regression Prediction: $\( \hat{y} = X \cdot w + b \)$

   • \( \hat{y} \) is the predicted output.

   • \( X \) is the input feature matrix (X_batch in the snippet).

   • \( w \) is the weight vector (self.weights in the snippet).

   • \( b \) is the bias term (self.bias in the snippet).

2. Error Calculation: $\( e = y - \hat{y} \)$

   • \( e \) is the error vector (difference between actual and predicted values).

   • \( y \) is the actual output (y_batch in the snippet).

   • \( \hat{y} \) is the predicted output.

3. Gradient of the Loss Function: The loss function for linear regression is typically the Mean Squared Error (MSE): $\( L(w, b) = \frac{1}{N} \sum_{i=1}^{N} (y_i - \hat{y}_i)^2 \)\( where \) N $ is the number of samples.

   The gradient of the MSE with respect to the weights \( w \) is given by: $\( \frac{\partial L}{\partial w} = -\frac{2}{N} X^T (y - \hat{y}) \)$

   • \( X^T \) is the transpose of the input feature matrix.

   • \( y - \hat{y} \) is the error vector.

4. Putting It All Together: $\( \text{gradient}_w = -2 \cdot \frac{1}{\text{batch\_size}} \cdot X_{\text{batch}}^T \cdot (y_{\text{batch}} - (X_{\text{batch}} \cdot \text{weights} + \text{bias})) \)$

   • The term \( X_{\text{batch}} \cdot \text{weights} + \text{bias} \) computes the predicted values \( \hat{y} \).

   • \( y_{\text{batch}} - \hat{y} \) computes the error vector.

   • Multiplying the error by \( X_{\text{batch}}^T \) and scaling by \( -2 / \text{batch\_size} \) gives the gradient of the loss function with respect to the weights.

This gradient indicates the direction and magnitude of the change needed in the weights to minimize the loss function. In gradient descent optimization, you would use this gradient to update the weights in the direction that reduces the loss.

Load Diabetes Data

# Load the diabetes dataset
diabetes = load_diabetes()
df = pd.DataFrame(data=diabetes.data, columns=diabetes.feature_names)
df['target'] = diabetes.target
df.head()

	age	sex	bmi	bp	s1	s2	s3	s4	s5	s6	target
0	0.038076	0.050680	0.061696	0.021872	-0.044223	-0.034821	-0.043401	-0.002592	0.019907	-0.017646	151.0
1	-0.001882	-0.044642	-0.051474	-0.026328	-0.008449	-0.019163	0.074412	-0.039493	-0.068332	-0.092204	75.0
2	0.085299	0.050680	0.044451	-0.005670	-0.045599	-0.034194	-0.032356	-0.002592	0.002861	-0.025930	141.0
3	-0.089063	-0.044642	-0.011595	-0.036656	0.012191	0.024991	-0.036038	0.034309	0.022688	-0.009362	206.0
4	0.005383	-0.044642	-0.036385	0.021872	0.003935	0.015596	0.008142	-0.002592	-0.031988	-0.046641	135.0

# Create histograms for each feature
df.hist(bins=10, figsize=(20, 15))
plt.tight_layout()
plt.show()

# Pairplot of the features
sns.pairplot(df)

<seaborn.axisgrid.PairGrid at 0x35174e410>

Split Data

# Get the input features (X) and target values (y)
X = diabetes.data
y = diabetes.target

# Split the dataset into training and test sets
def split_dataset(X, y, test_ratio=0.2):
    indices = np.random.permutation(len(X))
    test_size = int(len(X) * test_ratio)
    test_indices = indices[:test_size]
    train_indices = indices[test_size:]
    return X[train_indices], X[test_indices], y[train_indices], y[test_indices]

X_train, X_test, y_train, y_test = split_dataset(X, y)
X_train, X_val, y_train, y_val = split_dataset(X_train, y_train)

Fine-Tune Model with Optuna

def objective(trial):
    learning_rate = trial.suggest_loguniform('learning_rate', 1e-3, 1e-1)
    epochs = trial.suggest_int('epochs', 30, 250)
    batch_size = trial.suggest_categorical('batch_size', [1, 10, 50, 100])
    reg_param = trial.suggest_loguniform('reg_param', 1e-5, 1e-1)
    reg = trial.suggest_categorical('reg', ['l1', 'l2'])

    regressor = SGDRegressor(learning_rate=learning_rate, epochs=epochs, batch_size=batch_size, reg=reg, reg_param=reg_param)
    regressor.fit(X_train, y_train)

    # Compute the validation loss
    val_loss = regressor.compute_loss(X_val, y_val)

    return val_loss

optuna.logging.set_verbosity(optuna.logging.WARNING)
study = optuna.create_study(direction='minimize')
study.optimize(objective, n_trials=100)

best_params = study.best_params

for key, value in best_params.items():
    if key == 'reg':
        print(f'{key.capitalize()}: {value.capitalize()}')
    else:
        print(f'{key.capitalize()}: {value:.3f}')

Learning_rate: 0.053
Epochs: 156.000
Batch_size: 1.000
Reg_param: 0.001
Reg: L1

Predict Data

best_regressor = SGDRegressor(learning_rate=best_params['learning_rate'], epochs=best_params['epochs'], batch_size=best_params['batch_size'], reg='l1', reg_param=best_params['reg_param'])
best_regressor.fit(X_train, y_train)
predictions = best_regressor.predict(X_test)
loss = best_regressor.compute_loss(X_test, y_test)
print(f"Best model loss on test data: {loss:.2f}")

Best model loss on test data: 56.77

Solving using scikit(ML) or tensorflow(deep learning) libraries

from sklearn.linear_model import SGDRegressor
from sklearn.datasets import load_diabetes

# Load the diabetes dataset
diabetes = load_diabetes()

X = diabetes.data
y = diabetes.target
X_train, X_test, y_train, y_test = split_dataset(X, y)

# Create and fit the model
model = SGDRegressor(max_iter=1000, verbose=1)
model.fit(X_train, y_train)

# Making predictions
predictions = model.predict(X_test)
score = model.score(X_test, y_test)
print(f"Model Score: {loss:.2f}")

-- Epoch 1
Norm: 5.10, NNZs: 10, Bias: 100.522800, T: 354, Avg. loss: 6847.592915
Total training time: 0.00 seconds.
-- Epoch 2
Norm: 8.37, NNZs: 10, Bias: 127.264144, T: 708, Avg. loss: 3510.538724
Total training time: 0.00 seconds.
-- Epoch 3
Norm: 11.19, NNZs: 10, Bias: 138.999043, T: 1062, Avg. loss: 2983.511437
Total training time: 0.00 seconds.
-- Epoch 4
Norm: 13.74, NNZs: 10, Bias: 145.062305, T: 1416, Avg. loss: 2845.936159
Total training time: 0.00 seconds.
-- Epoch 5
Norm: 16.12, NNZs: 10, Bias: 148.137264, T: 1770, Avg. loss: 2801.797058
Total training time: 0.00 seconds.
-- Epoch 6
Norm: 18.38, NNZs: 10, Bias: 149.894353, T: 2124, Avg. loss: 2781.154892
Total training time: 0.00 seconds.
-- Epoch 7
Norm: 20.53, NNZs: 10, Bias: 150.800816, T: 2478, Avg. loss: 2768.680559
Total training time: 0.00 seconds.
-- Epoch 8
Norm: 22.60, NNZs: 10, Bias: 151.242143, T: 2832, Avg. loss: 2758.656448
Total training time: 0.00 seconds.
-- Epoch 9
Norm: 24.60, NNZs: 10, Bias: 151.771775, T: 3186, Avg. loss: 2749.442752
Total training time: 0.00 seconds.
-- Epoch 10
Norm: 26.54, NNZs: 10, Bias: 151.582218, T: 3540, Avg. loss: 2740.754809
Total training time: 0.00 seconds.
-- Epoch 11
Norm: 28.42, NNZs: 10, Bias: 151.747070, T: 3894, Avg. loss: 2733.306380
Total training time: 0.00 seconds.
-- Epoch 12
Norm: 30.25, NNZs: 10, Bias: 151.728445, T: 4248, Avg. loss: 2725.337360
Total training time: 0.00 seconds.
-- Epoch 13
Norm: 32.04, NNZs: 10, Bias: 151.894645, T: 4602, Avg. loss: 2717.771606
Total training time: 0.00 seconds.
-- Epoch 14
Norm: 33.79, NNZs: 10, Bias: 152.017881, T: 4956, Avg. loss: 2710.466485
Total training time: 0.00 seconds.
-- Epoch 15
Norm: 35.50, NNZs: 10, Bias: 151.995696, T: 5310, Avg. loss: 2703.286734
Total training time: 0.00 seconds.
-- Epoch 16
Norm: 37.18, NNZs: 10, Bias: 152.212952, T: 5664, Avg. loss: 2696.122004
Total training time: 0.00 seconds.
-- Epoch 17
Norm: 38.83, NNZs: 10, Bias: 152.163972, T: 6018, Avg. loss: 2689.480408
Total training time: 0.00 seconds.
-- Epoch 18
Norm: 40.45, NNZs: 10, Bias: 152.094615, T: 6372, Avg. loss: 2682.765072
Total training time: 0.00 seconds.
-- Epoch 19
Norm: 42.04, NNZs: 10, Bias: 152.197114, T: 6726, Avg. loss: 2676.141595
Total training time: 0.00 seconds.
-- Epoch 20
Norm: 43.61, NNZs: 10, Bias: 152.340600, T: 7080, Avg. loss: 2669.633727
Total training time: 0.00 seconds.
-- Epoch 21
Norm: 45.15, NNZs: 10, Bias: 152.168880, T: 7434, Avg. loss: 2663.476845
Total training time: 0.00 seconds.
-- Epoch 22
Norm: 46.67, NNZs: 10, Bias: 152.320332, T: 7788, Avg. loss: 2657.128712
Total training time: 0.00 seconds.
-- Epoch 23
Norm: 48.16, NNZs: 10, Bias: 152.151827, T: 8142, Avg. loss: 2651.034605
Total training time: 0.00 seconds.
-- Epoch 24
Norm: 49.64, NNZs: 10, Bias: 152.147000, T: 8496, Avg. loss: 2645.233606
Total training time: 0.00 seconds.
-- Epoch 25
Norm: 51.09, NNZs: 10, Bias: 151.978548, T: 8850, Avg. loss: 2639.213070
Total training time: 0.00 seconds.
-- Epoch 26
Norm: 52.53, NNZs: 10, Bias: 152.016123, T: 9204, Avg. loss: 2633.606540
Total training time: 0.00 seconds.
-- Epoch 27
Norm: 53.95, NNZs: 10, Bias: 151.953448, T: 9558, Avg. loss: 2627.711573
Total training time: 0.00 seconds.
-- Epoch 28
Norm: 55.36, NNZs: 10, Bias: 151.982892, T: 9912, Avg. loss: 2622.293902
Total training time: 0.00 seconds.
-- Epoch 29
Norm: 56.75, NNZs: 10, Bias: 152.095962, T: 10266, Avg. loss: 2616.644761
Total training time: 0.00 seconds.
-- Epoch 30
Norm: 58.12, NNZs: 10, Bias: 151.973540, T: 10620, Avg. loss: 2611.224341
Total training time: 0.00 seconds.
-- Epoch 31
Norm: 59.47, NNZs: 10, Bias: 151.889582, T: 10974, Avg. loss: 2605.816736
Total training time: 0.00 seconds.
-- Epoch 32
Norm: 60.82, NNZs: 10, Bias: 152.163368, T: 11328, Avg. loss: 2600.315900
Total training time: 0.00 seconds.
-- Epoch 33
Norm: 62.14, NNZs: 10, Bias: 152.010018, T: 11682, Avg. loss: 2595.325618
Total training time: 0.00 seconds.
-- Epoch 34
Norm: 63.46, NNZs: 10, Bias: 152.084830, T: 12036, Avg. loss: 2590.200594
Total training time: 0.00 seconds.
-- Epoch 35
Norm: 64.76, NNZs: 10, Bias: 152.133606, T: 12390, Avg. loss: 2585.106942
Total training time: 0.00 seconds.
-- Epoch 36
Norm: 66.05, NNZs: 10, Bias: 152.253280, T: 12744, Avg. loss: 2580.001488
Total training time: 0.00 seconds.
-- Epoch 37
Norm: 67.33, NNZs: 10, Bias: 152.338072, T: 13098, Avg. loss: 2574.975400
Total training time: 0.00 seconds.
-- Epoch 38
Norm: 68.59, NNZs: 10, Bias: 152.223540, T: 13452, Avg. loss: 2570.270727
Total training time: 0.00 seconds.
-- Epoch 39
Norm: 69.84, NNZs: 10, Bias: 152.206454, T: 13806, Avg. loss: 2565.371032
Total training time: 0.00 seconds.
-- Epoch 40
Norm: 71.09, NNZs: 10, Bias: 152.098278, T: 14160, Avg. loss: 2560.534019
Total training time: 0.00 seconds.
-- Epoch 41
Norm: 72.32, NNZs: 10, Bias: 152.224122, T: 14514, Avg. loss: 2555.629973
Total training time: 0.00 seconds.
-- Epoch 42
Norm: 73.54, NNZs: 10, Bias: 152.066227, T: 14868, Avg. loss: 2551.056167
Total training time: 0.00 seconds.
-- Epoch 43
Norm: 74.75, NNZs: 10, Bias: 152.161892, T: 15222, Avg. loss: 2546.431540
Total training time: 0.00 seconds.
-- Epoch 44
Norm: 75.95, NNZs: 10, Bias: 152.120903, T: 15576, Avg. loss: 2541.915219
Total training time: 0.00 seconds.
-- Epoch 45
Norm: 77.14, NNZs: 10, Bias: 152.142615, T: 15930, Avg. loss: 2537.380594
Total training time: 0.00 seconds.
-- Epoch 46
Norm: 78.32, NNZs: 10, Bias: 152.175895, T: 16284, Avg. loss: 2532.783632
Total training time: 0.00 seconds.
-- Epoch 47
Norm: 79.50, NNZs: 10, Bias: 152.135058, T: 16638, Avg. loss: 2528.448918
Total training time: 0.00 seconds.
-- Epoch 48
Norm: 80.66, NNZs: 10, Bias: 152.069777, T: 16992, Avg. loss: 2524.040417
Total training time: 0.00 seconds.
-- Epoch 49
Norm: 81.81, NNZs: 10, Bias: 152.192360, T: 17346, Avg. loss: 2519.527199
Total training time: 0.00 seconds.
-- Epoch 50
Norm: 82.96, NNZs: 10, Bias: 152.079148, T: 17700, Avg. loss: 2515.340414
Total training time: 0.00 seconds.
-- Epoch 51
Norm: 84.10, NNZs: 10, Bias: 151.993913, T: 18054, Avg. loss: 2511.063705
Total training time: 0.00 seconds.
-- Epoch 52
Norm: 85.22, NNZs: 10, Bias: 152.123636, T: 18408, Avg. loss: 2506.754218
Total training time: 0.00 seconds.
-- Epoch 53
Norm: 86.35, NNZs: 10, Bias: 152.053958, T: 18762, Avg. loss: 2502.610993
Total training time: 0.00 seconds.
-- Epoch 54
Norm: 87.46, NNZs: 10, Bias: 152.034670, T: 19116, Avg. loss: 2498.486367
Total training time: 0.00 seconds.
-- Epoch 55
Norm: 88.57, NNZs: 10, Bias: 151.955094, T: 19470, Avg. loss: 2494.311981
Total training time: 0.00 seconds.
-- Epoch 56
Norm: 89.66, NNZs: 10, Bias: 151.973684, T: 19824, Avg. loss: 2490.307278
Total training time: 0.00 seconds.
-- Epoch 57
Norm: 90.75, NNZs: 10, Bias: 152.036462, T: 20178, Avg. loss: 2486.252400
Total training time: 0.00 seconds.
-- Epoch 58
Norm: 91.84, NNZs: 10, Bias: 152.014623, T: 20532, Avg. loss: 2482.274852
Total training time: 0.00 seconds.
-- Epoch 59
Norm: 92.91, NNZs: 10, Bias: 151.950137, T: 20886, Avg. loss: 2478.261689
Total training time: 0.00 seconds.
-- Epoch 60
Norm: 93.98, NNZs: 10, Bias: 151.857641, T: 21240, Avg. loss: 2474.258519
Total training time: 0.00 seconds.
-- Epoch 61
Norm: 95.05, NNZs: 10, Bias: 151.824740, T: 21594, Avg. loss: 2470.453785
Total training time: 0.00 seconds.
-- Epoch 62
Norm: 96.10, NNZs: 10, Bias: 152.096498, T: 21948, Avg. loss: 2466.148073
Total training time: 0.00 seconds.
-- Epoch 63
Norm: 97.15, NNZs: 10, Bias: 152.017482, T: 22302, Avg. loss: 2462.749019
Total training time: 0.00 seconds.
-- Epoch 64
Norm: 98.19, NNZs: 10, Bias: 152.019798, T: 22656, Avg. loss: 2458.966399
Total training time: 0.00 seconds.
-- Epoch 65
Norm: 99.23, NNZs: 10, Bias: 152.098444, T: 23010, Avg. loss: 2455.133997
Total training time: 0.00 seconds.
-- Epoch 66
Norm: 100.26, NNZs: 10, Bias: 151.960577, T: 23364, Avg. loss: 2451.329072
Total training time: 0.00 seconds.
-- Epoch 67
Norm: 101.29, NNZs: 10, Bias: 152.060158, T: 23718, Avg. loss: 2447.667250
Total training time: 0.00 seconds.
-- Epoch 68
Norm: 102.30, NNZs: 10, Bias: 152.010467, T: 24072, Avg. loss: 2444.023875
Total training time: 0.00 seconds.
-- Epoch 69
Norm: 103.32, NNZs: 10, Bias: 152.051639, T: 24426, Avg. loss: 2440.402691
Total training time: 0.00 seconds.
-- Epoch 70
Norm: 104.32, NNZs: 10, Bias: 151.951170, T: 24780, Avg. loss: 2436.719735
Total training time: 0.00 seconds.
-- Epoch 71
Norm: 105.32, NNZs: 10, Bias: 151.837948, T: 25134, Avg. loss: 2433.073698
Total training time: 0.00 seconds.
-- Epoch 72
Norm: 106.32, NNZs: 10, Bias: 151.747552, T: 25488, Avg. loss: 2429.519226
Total training time: 0.00 seconds.
-- Epoch 73
Norm: 107.31, NNZs: 10, Bias: 151.956777, T: 25842, Avg. loss: 2425.843405
Total training time: 0.00 seconds.
-- Epoch 74
Norm: 108.29, NNZs: 10, Bias: 151.966820, T: 26196, Avg. loss: 2422.607534
Total training time: 0.00 seconds.
-- Epoch 75
Norm: 109.27, NNZs: 10, Bias: 152.039102, T: 26550, Avg. loss: 2419.049390
Total training time: 0.00 seconds.
-- Epoch 76
Norm: 110.24, NNZs: 10, Bias: 152.006543, T: 26904, Avg. loss: 2415.658165
Total training time: 0.00 seconds.
-- Epoch 77
Norm: 111.21, NNZs: 10, Bias: 151.959183, T: 27258, Avg. loss: 2412.211462
Total training time: 0.00 seconds.
-- Epoch 78
Norm: 112.17, NNZs: 10, Bias: 151.896312, T: 27612, Avg. loss: 2408.757224
Total training time: 0.00 seconds.
-- Epoch 79
Norm: 113.13, NNZs: 10, Bias: 151.918244, T: 27966, Avg. loss: 2405.426986
Total training time: 0.00 seconds.
-- Epoch 80
Norm: 114.08, NNZs: 10, Bias: 151.975757, T: 28320, Avg. loss: 2402.023996
Total training time: 0.00 seconds.
-- Epoch 81
Norm: 115.03, NNZs: 10, Bias: 151.912570, T: 28674, Avg. loss: 2398.704319
Total training time: 0.00 seconds.
-- Epoch 82
Norm: 115.98, NNZs: 10, Bias: 152.057560, T: 29028, Avg. loss: 2395.260191
Total training time: 0.00 seconds.
-- Epoch 83
Norm: 116.91, NNZs: 10, Bias: 151.955204, T: 29382, Avg. loss: 2392.051456
Total training time: 0.00 seconds.
-- Epoch 84
Norm: 117.85, NNZs: 10, Bias: 151.989871, T: 29736, Avg. loss: 2388.880291
Total training time: 0.00 seconds.
-- Epoch 85
Norm: 118.77, NNZs: 10, Bias: 151.999262, T: 30090, Avg. loss: 2385.634368
Total training time: 0.00 seconds.
-- Epoch 86
Norm: 119.70, NNZs: 10, Bias: 151.948684, T: 30444, Avg. loss: 2382.398325
Total training time: 0.00 seconds.
-- Epoch 87
Norm: 120.62, NNZs: 10, Bias: 151.941642, T: 30798, Avg. loss: 2379.228970
Total training time: 0.00 seconds.
-- Epoch 88
Norm: 121.53, NNZs: 10, Bias: 151.981575, T: 31152, Avg. loss: 2376.046877
Total training time: 0.00 seconds.
-- Epoch 89
Norm: 122.44, NNZs: 10, Bias: 151.948072, T: 31506, Avg. loss: 2372.903858
Total training time: 0.00 seconds.
-- Epoch 90
Norm: 123.35, NNZs: 10, Bias: 151.919086, T: 31860, Avg. loss: 2369.754329
Total training time: 0.00 seconds.
-- Epoch 91
Norm: 124.25, NNZs: 10, Bias: 152.054246, T: 32214, Avg. loss: 2366.502697
Total training time: 0.00 seconds.
-- Epoch 92
Norm: 125.15, NNZs: 10, Bias: 152.035976, T: 32568, Avg. loss: 2363.582713
Total training time: 0.00 seconds.
-- Epoch 93
Norm: 126.04, NNZs: 10, Bias: 151.933507, T: 32922, Avg. loss: 2360.418461
Total training time: 0.00 seconds.
-- Epoch 94
Norm: 126.93, NNZs: 10, Bias: 151.975000, T: 33276, Avg. loss: 2357.477548
Total training time: 0.00 seconds.
-- Epoch 95
Norm: 127.82, NNZs: 10, Bias: 151.974783, T: 33630, Avg. loss: 2354.383931
Total training time: 0.00 seconds.
-- Epoch 96
Norm: 128.70, NNZs: 10, Bias: 151.954445, T: 33984, Avg. loss: 2351.415100
Total training time: 0.00 seconds.
-- Epoch 97
Norm: 129.57, NNZs: 10, Bias: 151.985283, T: 34338, Avg. loss: 2348.428154
Total training time: 0.00 seconds.
-- Epoch 98
Norm: 130.45, NNZs: 10, Bias: 151.990429, T: 34692, Avg. loss: 2345.487501
Total training time: 0.00 seconds.
-- Epoch 99
Norm: 131.32, NNZs: 10, Bias: 151.979990, T: 35046, Avg. loss: 2342.525480
Total training time: 0.00 seconds.
-- Epoch 100
Norm: 132.18, NNZs: 10, Bias: 151.967602, T: 35400, Avg. loss: 2339.592622
Total training time: 0.00 seconds.
-- Epoch 101
Norm: 133.04, NNZs: 10, Bias: 151.894900, T: 35754, Avg. loss: 2336.657809
Total training time: 0.00 seconds.
-- Epoch 102
Norm: 133.90, NNZs: 10, Bias: 151.900300, T: 36108, Avg. loss: 2333.802809
Total training time: 0.00 seconds.
-- Epoch 103
Norm: 134.75, NNZs: 10, Bias: 151.886071, T: 36462, Avg. loss: 2330.922477
Total training time: 0.00 seconds.
-- Epoch 104
Norm: 135.60, NNZs: 10, Bias: 151.923963, T: 36816, Avg. loss: 2328.052986
Total training time: 0.00 seconds.
-- Epoch 105
Norm: 136.45, NNZs: 10, Bias: 152.044087, T: 37170, Avg. loss: 2325.064861
Total training time: 0.00 seconds.
-- Epoch 106
Norm: 137.29, NNZs: 10, Bias: 152.008408, T: 37524, Avg. loss: 2322.396867
Total training time: 0.00 seconds.
-- Epoch 107
Norm: 138.13, NNZs: 10, Bias: 152.007568, T: 37878, Avg. loss: 2319.600731
Total training time: 0.00 seconds.
-- Epoch 108
Norm: 138.97, NNZs: 10, Bias: 151.976091, T: 38232, Avg. loss: 2316.798743
Total training time: 0.00 seconds.
-- Epoch 109
Norm: 139.80, NNZs: 10, Bias: 152.006230, T: 38586, Avg. loss: 2314.021114
Total training time: 0.00 seconds.
-- Epoch 110
Norm: 140.63, NNZs: 10, Bias: 151.939883, T: 38940, Avg. loss: 2311.240450
Total training time: 0.00 seconds.
-- Epoch 111
Norm: 141.46, NNZs: 10, Bias: 152.006837, T: 39294, Avg. loss: 2308.505790
Total training time: 0.00 seconds.
-- Epoch 112
Norm: 142.28, NNZs: 10, Bias: 152.029079, T: 39648, Avg. loss: 2305.788982
Total training time: 0.00 seconds.
-- Epoch 113
Norm: 143.10, NNZs: 10, Bias: 152.054997, T: 40002, Avg. loss: 2303.104275
Total training time: 0.00 seconds.
-- Epoch 114
Norm: 143.91, NNZs: 10, Bias: 152.079981, T: 40356, Avg. loss: 2300.412534
Total training time: 0.00 seconds.
-- Epoch 115
Norm: 144.72, NNZs: 10, Bias: 152.170420, T: 40710, Avg. loss: 2297.619127
Total training time: 0.00 seconds.
-- Epoch 116
Norm: 145.53, NNZs: 10, Bias: 152.146938, T: 41064, Avg. loss: 2295.124440
Total training time: 0.00 seconds.
-- Epoch 117
Norm: 146.34, NNZs: 10, Bias: 152.119382, T: 41418, Avg. loss: 2292.466393
Total training time: 0.00 seconds.
-- Epoch 118
Norm: 147.14, NNZs: 10, Bias: 152.160462, T: 41772, Avg. loss: 2289.784262
Total training time: 0.00 seconds.
-- Epoch 119
Norm: 147.94, NNZs: 10, Bias: 151.996133, T: 42126, Avg. loss: 2287.015470
Total training time: 0.00 seconds.
-- Epoch 120
Norm: 148.73, NNZs: 10, Bias: 152.058363, T: 42480, Avg. loss: 2284.573128
Total training time: 0.00 seconds.
-- Epoch 121
Norm: 149.53, NNZs: 10, Bias: 152.042617, T: 42834, Avg. loss: 2282.040090
Total training time: 0.00 seconds.
-- Epoch 122
Norm: 150.32, NNZs: 10, Bias: 151.975273, T: 43188, Avg. loss: 2279.455581
Total training time: 0.00 seconds.
-- Epoch 123
Norm: 151.10, NNZs: 10, Bias: 151.914364, T: 43542, Avg. loss: 2276.874828
Total training time: 0.00 seconds.
-- Epoch 124
Norm: 151.89, NNZs: 10, Bias: 151.935666, T: 43896, Avg. loss: 2274.363782
Total training time: 0.00 seconds.
-- Epoch 125
Norm: 152.67, NNZs: 10, Bias: 152.011928, T: 44250, Avg. loss: 2271.775899
Total training time: 0.00 seconds.
-- Epoch 126
Norm: 153.44, NNZs: 10, Bias: 151.915485, T: 44604, Avg. loss: 2269.239387
Total training time: 0.00 seconds.
-- Epoch 127
Norm: 154.22, NNZs: 10, Bias: 151.851825, T: 44958, Avg. loss: 2266.778963
Total training time: 0.00 seconds.
-- Epoch 128
Norm: 154.99, NNZs: 10, Bias: 151.944901, T: 45312, Avg. loss: 2264.293744
Total training time: 0.00 seconds.
-- Epoch 129
Norm: 155.76, NNZs: 10, Bias: 151.892109, T: 45666, Avg. loss: 2261.858735
Total training time: 0.00 seconds.
-- Epoch 130
Norm: 156.53, NNZs: 10, Bias: 151.897697, T: 46020, Avg. loss: 2259.436750
Total training time: 0.00 seconds.
-- Epoch 131
Norm: 157.29, NNZs: 10, Bias: 151.916118, T: 46374, Avg. loss: 2256.929246
Total training time: 0.00 seconds.
-- Epoch 132
Norm: 158.05, NNZs: 10, Bias: 151.913399, T: 46728, Avg. loss: 2254.561938
Total training time: 0.00 seconds.
-- Epoch 133
Norm: 158.81, NNZs: 10, Bias: 151.955716, T: 47082, Avg. loss: 2252.125018
Total training time: 0.00 seconds.
-- Epoch 134
Norm: 159.56, NNZs: 10, Bias: 151.945257, T: 47436, Avg. loss: 2249.733810
Total training time: 0.00 seconds.
-- Epoch 135
Norm: 160.31, NNZs: 10, Bias: 151.912317, T: 47790, Avg. loss: 2247.337074
Total training time: 0.00 seconds.
-- Epoch 136
Norm: 161.06, NNZs: 10, Bias: 151.785152, T: 48144, Avg. loss: 2244.747095
Total training time: 0.00 seconds.
-- Epoch 137
Norm: 161.81, NNZs: 10, Bias: 151.806537, T: 48498, Avg. loss: 2242.610615
Total training time: 0.00 seconds.
-- Epoch 138
Norm: 162.55, NNZs: 10, Bias: 151.856726, T: 48852, Avg. loss: 2240.264359
Total training time: 0.00 seconds.
-- Epoch 139
Norm: 163.29, NNZs: 10, Bias: 151.841184, T: 49206, Avg. loss: 2237.873938
Total training time: 0.00 seconds.
-- Epoch 140
Norm: 164.03, NNZs: 10, Bias: 151.932983, T: 49560, Avg. loss: 2235.513532
Total training time: 0.00 seconds.
-- Epoch 141
Norm: 164.77, NNZs: 10, Bias: 151.947175, T: 49914, Avg. loss: 2233.276926
Total training time: 0.00 seconds.
-- Epoch 142
Norm: 165.50, NNZs: 10, Bias: 151.893026, T: 50268, Avg. loss: 2230.944132
Total training time: 0.00 seconds.
-- Epoch 143
Norm: 166.23, NNZs: 10, Bias: 151.885910, T: 50622, Avg. loss: 2228.678801
Total training time: 0.00 seconds.
-- Epoch 144
Norm: 166.96, NNZs: 10, Bias: 152.002423, T: 50976, Avg. loss: 2226.256668
Total training time: 0.00 seconds.
-- Epoch 145
Norm: 167.69, NNZs: 10, Bias: 151.987111, T: 51330, Avg. loss: 2224.122309
Total training time: 0.00 seconds.
-- Epoch 146
Norm: 168.41, NNZs: 10, Bias: 151.958092, T: 51684, Avg. loss: 2221.850361
Total training time: 0.00 seconds.
-- Epoch 147
Norm: 169.13, NNZs: 10, Bias: 151.912605, T: 52038, Avg. loss: 2219.576026
Total training time: 0.00 seconds.
-- Epoch 148
Norm: 169.85, NNZs: 10, Bias: 151.914231, T: 52392, Avg. loss: 2217.382681
Total training time: 0.00 seconds.
-- Epoch 149
Norm: 170.56, NNZs: 10, Bias: 151.936525, T: 52746, Avg. loss: 2215.163995
Total training time: 0.00 seconds.
-- Epoch 150
Norm: 171.28, NNZs: 10, Bias: 152.008985, T: 53100, Avg. loss: 2212.902890
Total training time: 0.00 seconds.
-- Epoch 151
Norm: 171.99, NNZs: 10, Bias: 152.030317, T: 53454, Avg. loss: 2210.742093
Total training time: 0.00 seconds.
-- Epoch 152
Norm: 172.70, NNZs: 10, Bias: 151.963544, T: 53808, Avg. loss: 2208.547265
Total training time: 0.00 seconds.
-- Epoch 153
Norm: 173.40, NNZs: 10, Bias: 151.957131, T: 54162, Avg. loss: 2206.369295
Total training time: 0.00 seconds.
-- Epoch 154
Norm: 174.11, NNZs: 10, Bias: 151.903959, T: 54516, Avg. loss: 2204.198101
Total training time: 0.00 seconds.
-- Epoch 155
Norm: 174.81, NNZs: 10, Bias: 151.854475, T: 54870, Avg. loss: 2202.025852
Total training time: 0.00 seconds.
-- Epoch 156
Norm: 175.51, NNZs: 10, Bias: 151.874466, T: 55224, Avg. loss: 2199.906158
Total training time: 0.00 seconds.
-- Epoch 157
Norm: 176.20, NNZs: 10, Bias: 151.805530, T: 55578, Avg. loss: 2197.706325
Total training time: 0.00 seconds.
-- Epoch 158
Norm: 176.90, NNZs: 10, Bias: 151.906914, T: 55932, Avg. loss: 2195.592132
Total training time: 0.00 seconds.
-- Epoch 159
Norm: 177.59, NNZs: 10, Bias: 151.870643, T: 56286, Avg. loss: 2193.505806
Total training time: 0.00 seconds.
-- Epoch 160
Norm: 178.28, NNZs: 10, Bias: 151.915705, T: 56640, Avg. loss: 2191.432855
Total training time: 0.00 seconds.
-- Epoch 161
Norm: 178.97, NNZs: 10, Bias: 151.994275, T: 56994, Avg. loss: 2189.243881
Total training time: 0.00 seconds.
-- Epoch 162
Norm: 179.65, NNZs: 10, Bias: 151.924962, T: 57348, Avg. loss: 2187.223926
Total training time: 0.00 seconds.
-- Epoch 163
Norm: 180.34, NNZs: 10, Bias: 151.939724, T: 57702, Avg. loss: 2185.198211
Total training time: 0.00 seconds.
-- Epoch 164
Norm: 181.02, NNZs: 10, Bias: 151.915334, T: 58056, Avg. loss: 2183.099872
Total training time: 0.00 seconds.
-- Epoch 165
Norm: 181.70, NNZs: 10, Bias: 151.898474, T: 58410, Avg. loss: 2181.071399
Total training time: 0.00 seconds.
-- Epoch 166
Norm: 182.37, NNZs: 10, Bias: 151.861766, T: 58764, Avg. loss: 2179.008018
Total training time: 0.00 seconds.
-- Epoch 167
Norm: 183.05, NNZs: 10, Bias: 151.932679, T: 59118, Avg. loss: 2176.939773
Total training time: 0.00 seconds.
-- Epoch 168
Norm: 183.72, NNZs: 10, Bias: 151.896503, T: 59472, Avg. loss: 2174.957989
Total training time: 0.00 seconds.
-- Epoch 169
Norm: 184.39, NNZs: 10, Bias: 151.935275, T: 59826, Avg. loss: 2172.935490
Total training time: 0.00 seconds.
-- Epoch 170
Norm: 185.06, NNZs: 10, Bias: 151.929470, T: 60180, Avg. loss: 2170.947708
Total training time: 0.00 seconds.
-- Epoch 171
Norm: 185.72, NNZs: 10, Bias: 152.048965, T: 60534, Avg. loss: 2168.744901
Total training time: 0.00 seconds.
-- Epoch 172
Norm: 186.39, NNZs: 10, Bias: 151.946970, T: 60888, Avg. loss: 2166.898624
Total training time: 0.00 seconds.
-- Epoch 173
Norm: 187.05, NNZs: 10, Bias: 152.010734, T: 61242, Avg. loss: 2164.922827
Total training time: 0.00 seconds.
-- Epoch 174
Norm: 187.71, NNZs: 10, Bias: 151.972937, T: 61596, Avg. loss: 2163.023195
Total training time: 0.00 seconds.
-- Epoch 175
Norm: 188.37, NNZs: 10, Bias: 151.950863, T: 61950, Avg. loss: 2161.052720
Total training time: 0.00 seconds.
-- Epoch 176
Norm: 189.02, NNZs: 10, Bias: 151.993136, T: 62304, Avg. loss: 2159.076923
Total training time: 0.00 seconds.
-- Epoch 177
Norm: 189.68, NNZs: 10, Bias: 151.959928, T: 62658, Avg. loss: 2157.166214
Total training time: 0.00 seconds.
-- Epoch 178
Norm: 190.33, NNZs: 10, Bias: 151.946966, T: 63012, Avg. loss: 2155.233835
Total training time: 0.00 seconds.
-- Epoch 179
Norm: 190.98, NNZs: 10, Bias: 151.849274, T: 63366, Avg. loss: 2153.198705
Total training time: 0.00 seconds.
-- Epoch 180
Norm: 191.63, NNZs: 10, Bias: 151.889269, T: 63720, Avg. loss: 2151.378021
Total training time: 0.00 seconds.
-- Epoch 181
Norm: 192.28, NNZs: 10, Bias: 151.850525, T: 64074, Avg. loss: 2149.418195
Total training time: 0.00 seconds.
-- Epoch 182
Norm: 192.92, NNZs: 10, Bias: 151.884058, T: 64428, Avg. loss: 2147.573920
Total training time: 0.00 seconds.
-- Epoch 183
Norm: 193.56, NNZs: 10, Bias: 151.902655, T: 64782, Avg. loss: 2145.690213
Total training time: 0.00 seconds.
-- Epoch 184
Norm: 194.20, NNZs: 10, Bias: 151.917317, T: 65136, Avg. loss: 2143.824040
Total training time: 0.00 seconds.
-- Epoch 185
Norm: 194.84, NNZs: 10, Bias: 151.911920, T: 65490, Avg. loss: 2141.956766
Total training time: 0.00 seconds.
-- Epoch 186
Norm: 195.48, NNZs: 10, Bias: 151.964738, T: 65844, Avg. loss: 2140.033076
Total training time: 0.00 seconds.
-- Epoch 187
Norm: 196.11, NNZs: 10, Bias: 152.004397, T: 66198, Avg. loss: 2138.170401
Total training time: 0.00 seconds.
-- Epoch 188
Norm: 196.75, NNZs: 10, Bias: 151.998132, T: 66552, Avg. loss: 2136.393114
Total training time: 0.00 seconds.
-- Epoch 189
Norm: 197.38, NNZs: 10, Bias: 152.015950, T: 66906, Avg. loss: 2134.509269
Total training time: 0.00 seconds.
-- Epoch 190
Norm: 198.01, NNZs: 10, Bias: 151.966013, T: 67260, Avg. loss: 2132.711456
Total training time: 0.00 seconds.
-- Epoch 191
Norm: 198.63, NNZs: 10, Bias: 151.976259, T: 67614, Avg. loss: 2130.884577
Total training time: 0.00 seconds.
-- Epoch 192
Norm: 199.26, NNZs: 10, Bias: 151.900668, T: 67968, Avg. loss: 2128.991274
Total training time: 0.00 seconds.
-- Epoch 193
Norm: 199.88, NNZs: 10, Bias: 151.918837, T: 68322, Avg. loss: 2127.258303
Total training time: 0.00 seconds.
-- Epoch 194
Norm: 200.50, NNZs: 10, Bias: 151.924402, T: 68676, Avg. loss: 2125.472438
Total training time: 0.00 seconds.
-- Epoch 195
Norm: 201.12, NNZs: 10, Bias: 151.868214, T: 69030, Avg. loss: 2123.643142
Total training time: 0.00 seconds.
-- Epoch 196
Norm: 201.74, NNZs: 10, Bias: 151.847406, T: 69384, Avg. loss: 2121.890945
Total training time: 0.00 seconds.
-- Epoch 197
Norm: 202.36, NNZs: 10, Bias: 151.961574, T: 69738, Avg. loss: 2119.932272
Total training time: 0.00 seconds.
-- Epoch 198
Norm: 202.97, NNZs: 10, Bias: 151.926874, T: 70092, Avg. loss: 2118.346943
Total training time: 0.00 seconds.
-- Epoch 199
Norm: 203.59, NNZs: 10, Bias: 151.896234, T: 70446, Avg. loss: 2116.584389
Total training time: 0.00 seconds.
-- Epoch 200
Norm: 204.20, NNZs: 10, Bias: 151.920957, T: 70800, Avg. loss: 2114.824708
Total training time: 0.00 seconds.
-- Epoch 201
Norm: 204.81, NNZs: 10, Bias: 151.922375, T: 71154, Avg. loss: 2113.079849
Total training time: 0.00 seconds.
-- Epoch 202
Norm: 205.42, NNZs: 10, Bias: 151.830875, T: 71508, Avg. loss: 2111.235443
Total training time: 0.00 seconds.
-- Epoch 203
Norm: 206.02, NNZs: 10, Bias: 151.832829, T: 71862, Avg. loss: 2109.627648
Total training time: 0.00 seconds.
-- Epoch 204
Norm: 206.63, NNZs: 10, Bias: 151.856147, T: 72216, Avg. loss: 2107.890473
Total training time: 0.00 seconds.
-- Epoch 205
Norm: 207.23, NNZs: 10, Bias: 151.802245, T: 72570, Avg. loss: 2106.098259
Total training time: 0.00 seconds.
-- Epoch 206
Norm: 207.83, NNZs: 10, Bias: 151.840534, T: 72924, Avg. loss: 2104.475787
Total training time: 0.00 seconds.
-- Epoch 207
Norm: 208.43, NNZs: 10, Bias: 151.847280, T: 73278, Avg. loss: 2102.772120
Total training time: 0.00 seconds.
-- Epoch 208
Norm: 209.03, NNZs: 10, Bias: 151.892118, T: 73632, Avg. loss: 2101.058921
Total training time: 0.00 seconds.
-- Epoch 209
Norm: 209.62, NNZs: 10, Bias: 151.907959, T: 73986, Avg. loss: 2099.390543
Total training time: 0.00 seconds.
-- Epoch 210
Norm: 210.22, NNZs: 10, Bias: 151.925537, T: 74340, Avg. loss: 2097.713980
Total training time: 0.00 seconds.
-- Epoch 211
Norm: 210.81, NNZs: 10, Bias: 151.902503, T: 74694, Avg. loss: 2096.013896
Total training time: 0.00 seconds.
-- Epoch 212
Norm: 211.40, NNZs: 10, Bias: 151.897928, T: 75048, Avg. loss: 2094.375719
Total training time: 0.00 seconds.
-- Epoch 213
Norm: 211.99, NNZs: 10, Bias: 151.825759, T: 75402, Avg. loss: 2092.656124
Total training time: 0.00 seconds.
-- Epoch 214
Norm: 212.58, NNZs: 10, Bias: 151.892758, T: 75756, Avg. loss: 2091.026495
Total training time: 0.00 seconds.
-- Epoch 215
Norm: 213.17, NNZs: 10, Bias: 151.861081, T: 76110, Avg. loss: 2089.395604
Total training time: 0.00 seconds.
-- Epoch 216
Norm: 213.75, NNZs: 10, Bias: 151.927230, T: 76464, Avg. loss: 2087.719879
Total training time: 0.00 seconds.
-- Epoch 217
Norm: 214.34, NNZs: 10, Bias: 151.880697, T: 76818, Avg. loss: 2086.130931
Total training time: 0.00 seconds.
-- Epoch 218
Norm: 214.92, NNZs: 10, Bias: 151.863655, T: 77172, Avg. loss: 2084.522100
Total training time: 0.00 seconds.
-- Epoch 219
Norm: 215.50, NNZs: 10, Bias: 151.865943, T: 77526, Avg. loss: 2082.889188
Total training time: 0.00 seconds.
-- Epoch 220
Norm: 216.08, NNZs: 10, Bias: 151.890592, T: 77880, Avg. loss: 2081.293953
Total training time: 0.00 seconds.
-- Epoch 221
Norm: 216.65, NNZs: 10, Bias: 151.970661, T: 78234, Avg. loss: 2079.587550
Total training time: 0.00 seconds.
-- Epoch 222
Norm: 217.23, NNZs: 10, Bias: 151.905199, T: 78588, Avg. loss: 2078.047186
Total training time: 0.00 seconds.
-- Epoch 223
Norm: 217.80, NNZs: 10, Bias: 151.872707, T: 78942, Avg. loss: 2076.497212
Total training time: 0.00 seconds.
-- Epoch 224
Norm: 218.38, NNZs: 10, Bias: 151.894774, T: 79296, Avg. loss: 2074.891562
Total training time: 0.00 seconds.
-- Epoch 225
Norm: 218.95, NNZs: 10, Bias: 151.836823, T: 79650, Avg. loss: 2073.293306
Total training time: 0.00 seconds.
-- Epoch 226
Norm: 219.52, NNZs: 10, Bias: 151.877287, T: 80004, Avg. loss: 2071.748933
Total training time: 0.00 seconds.
-- Epoch 227
Norm: 220.09, NNZs: 10, Bias: 151.907750, T: 80358, Avg. loss: 2070.182959
Total training time: 0.00 seconds.
-- Epoch 228
Norm: 220.65, NNZs: 10, Bias: 151.844627, T: 80712, Avg. loss: 2068.596123
Total training time: 0.00 seconds.
-- Epoch 229
Norm: 221.22, NNZs: 10, Bias: 151.892950, T: 81066, Avg. loss: 2067.047766
Total training time: 0.00 seconds.
-- Epoch 230
Norm: 221.78, NNZs: 10, Bias: 151.940438, T: 81420, Avg. loss: 2065.481941
Total training time: 0.00 seconds.
-- Epoch 231
Norm: 222.34, NNZs: 10, Bias: 151.892010, T: 81774, Avg. loss: 2063.977548
Total training time: 0.00 seconds.
-- Epoch 232
Norm: 222.91, NNZs: 10, Bias: 151.845757, T: 82128, Avg. loss: 2062.416144
Total training time: 0.00 seconds.
-- Epoch 233
Norm: 223.46, NNZs: 10, Bias: 151.871541, T: 82482, Avg. loss: 2060.919311
Total training time: 0.00 seconds.
-- Epoch 234
Norm: 224.02, NNZs: 10, Bias: 151.796132, T: 82836, Avg. loss: 2059.301826
Total training time: 0.00 seconds.
-- Epoch 235
Norm: 224.58, NNZs: 10, Bias: 151.791389, T: 83190, Avg. loss: 2057.879004
Total training time: 0.00 seconds.
-- Epoch 236
Norm: 225.13, NNZs: 10, Bias: 151.783550, T: 83544, Avg. loss: 2056.352163
Total training time: 0.00 seconds.
-- Epoch 237
Norm: 225.69, NNZs: 10, Bias: 151.826141, T: 83898, Avg. loss: 2054.859568
Total training time: 0.00 seconds.
-- Epoch 238
Norm: 226.24, NNZs: 10, Bias: 151.828251, T: 84252, Avg. loss: 2053.379547
Total training time: 0.00 seconds.
-- Epoch 239
Norm: 226.79, NNZs: 10, Bias: 151.869806, T: 84606, Avg. loss: 2051.863523
Total training time: 0.00 seconds.
-- Epoch 240
Norm: 227.34, NNZs: 10, Bias: 151.907541, T: 84960, Avg. loss: 2050.373406
Total training time: 0.00 seconds.
-- Epoch 241
Norm: 227.89, NNZs: 10, Bias: 151.905410, T: 85314, Avg. loss: 2048.910893
Total training time: 0.00 seconds.
-- Epoch 242
Norm: 228.44, NNZs: 10, Bias: 151.888679, T: 85668, Avg. loss: 2047.426381
Total training time: 0.00 seconds.
-- Epoch 243
Norm: 228.98, NNZs: 10, Bias: 151.854616, T: 86022, Avg. loss: 2045.939197
Total training time: 0.00 seconds.
-- Epoch 244
Norm: 229.53, NNZs: 10, Bias: 151.906201, T: 86376, Avg. loss: 2044.454176
Total training time: 0.00 seconds.
-- Epoch 245
Norm: 230.07, NNZs: 10, Bias: 151.914160, T: 86730, Avg. loss: 2043.037919
Total training time: 0.00 seconds.
-- Epoch 246
Norm: 230.61, NNZs: 10, Bias: 151.912255, T: 87084, Avg. loss: 2041.589001
Total training time: 0.00 seconds.
-- Epoch 247
Norm: 231.15, NNZs: 10, Bias: 151.932220, T: 87438, Avg. loss: 2040.119011
Total training time: 0.00 seconds.
-- Epoch 248
Norm: 231.69, NNZs: 10, Bias: 151.930865, T: 87792, Avg. loss: 2038.688347
Total training time: 0.00 seconds.
-- Epoch 249
Norm: 232.23, NNZs: 10, Bias: 151.935513, T: 88146, Avg. loss: 2037.250874
Total training time: 0.00 seconds.
-- Epoch 250
Norm: 232.76, NNZs: 10, Bias: 151.883288, T: 88500, Avg. loss: 2035.805759
Total training time: 0.00 seconds.
-- Epoch 251
Norm: 233.30, NNZs: 10, Bias: 151.792379, T: 88854, Avg. loss: 2034.238995
Total training time: 0.00 seconds.
-- Epoch 252
Norm: 233.83, NNZs: 10, Bias: 151.850368, T: 89208, Avg. loss: 2032.934743
Total training time: 0.00 seconds.
-- Epoch 253
Norm: 234.36, NNZs: 10, Bias: 151.836450, T: 89562, Avg. loss: 2031.561950
Total training time: 0.00 seconds.
-- Epoch 254
Norm: 234.89, NNZs: 10, Bias: 151.869553, T: 89916, Avg. loss: 2030.135674
Total training time: 0.00 seconds.
-- Epoch 255
Norm: 235.42, NNZs: 10, Bias: 151.874708, T: 90270, Avg. loss: 2028.745501
Total training time: 0.00 seconds.
-- Epoch 256
Norm: 235.95, NNZs: 10, Bias: 151.870960, T: 90624, Avg. loss: 2027.334700
Total training time: 0.00 seconds.
-- Epoch 257
Norm: 236.48, NNZs: 10, Bias: 151.888405, T: 90978, Avg. loss: 2025.945494
Total training time: 0.00 seconds.
-- Epoch 258
Norm: 237.00, NNZs: 10, Bias: 151.838265, T: 91332, Avg. loss: 2024.531512
Total training time: 0.00 seconds.
-- Epoch 259
Norm: 237.53, NNZs: 10, Bias: 151.858040, T: 91686, Avg. loss: 2023.160673
Total training time: 0.00 seconds.
-- Epoch 260
Norm: 238.05, NNZs: 10, Bias: 151.841846, T: 92040, Avg. loss: 2021.794334
Total training time: 0.00 seconds.
-- Epoch 261
Norm: 238.57, NNZs: 10, Bias: 151.847695, T: 92394, Avg. loss: 2020.425703
Total training time: 0.00 seconds.
-- Epoch 262
Norm: 239.09, NNZs: 10, Bias: 151.881983, T: 92748, Avg. loss: 2019.039588
Total training time: 0.00 seconds.
-- Epoch 263
Norm: 239.61, NNZs: 10, Bias: 151.904628, T: 93102, Avg. loss: 2017.665167
Total training time: 0.00 seconds.
-- Epoch 264
Norm: 240.13, NNZs: 10, Bias: 151.904261, T: 93456, Avg. loss: 2016.340739
Total training time: 0.00 seconds.
-- Epoch 265
Norm: 240.65, NNZs: 10, Bias: 151.965769, T: 93810, Avg. loss: 2014.874812
Total training time: 0.00 seconds.
-- Epoch 266
Norm: 241.17, NNZs: 10, Bias: 151.909656, T: 94164, Avg. loss: 2013.606849
Total training time: 0.00 seconds.
-- Epoch 267
Norm: 241.68, NNZs: 10, Bias: 151.925029, T: 94518, Avg. loss: 2012.275181
Total training time: 0.00 seconds.
-- Epoch 268
Norm: 242.19, NNZs: 10, Bias: 151.928670, T: 94872, Avg. loss: 2010.955695
Total training time: 0.00 seconds.
-- Epoch 269
Norm: 242.71, NNZs: 10, Bias: 151.956999, T: 95226, Avg. loss: 2009.595626
Total training time: 0.00 seconds.
-- Epoch 270
Norm: 243.22, NNZs: 10, Bias: 151.957702, T: 95580, Avg. loss: 2008.292917
Total training time: 0.00 seconds.
-- Epoch 271
Norm: 243.73, NNZs: 10, Bias: 151.945419, T: 95934, Avg. loss: 2006.968807
Total training time: 0.00 seconds.
-- Epoch 272
Norm: 244.23, NNZs: 10, Bias: 151.926857, T: 96288, Avg. loss: 2005.648184
Total training time: 0.00 seconds.
-- Epoch 273
Norm: 244.74, NNZs: 10, Bias: 151.934656, T: 96642, Avg. loss: 2004.331106
Total training time: 0.00 seconds.
-- Epoch 274
Norm: 245.25, NNZs: 10, Bias: 151.948277, T: 96996, Avg. loss: 2003.010359
Total training time: 0.00 seconds.
-- Epoch 275
Norm: 245.75, NNZs: 10, Bias: 151.899421, T: 97350, Avg. loss: 2001.706409
Total training time: 0.00 seconds.
-- Epoch 276
Norm: 246.26, NNZs: 10, Bias: 151.888319, T: 97704, Avg. loss: 2000.403534
Total training time: 0.00 seconds.
-- Epoch 277
Norm: 246.76, NNZs: 10, Bias: 151.862975, T: 98058, Avg. loss: 1999.101752
Total training time: 0.00 seconds.
-- Epoch 278
Norm: 247.26, NNZs: 10, Bias: 151.843703, T: 98412, Avg. loss: 1997.811431
Total training time: 0.00 seconds.
-- Epoch 279
Norm: 247.76, NNZs: 10, Bias: 151.870026, T: 98766, Avg. loss: 1996.522130
Total training time: 0.00 seconds.
-- Epoch 280
Norm: 248.26, NNZs: 10, Bias: 151.871073, T: 99120, Avg. loss: 1995.266258
Total training time: 0.00 seconds.
-- Epoch 281
Norm: 248.76, NNZs: 10, Bias: 151.809395, T: 99474, Avg. loss: 1993.908397
Total training time: 0.01 seconds.
-- Epoch 282
Norm: 249.26, NNZs: 10, Bias: 151.823379, T: 99828, Avg. loss: 1992.712312
Total training time: 0.01 seconds.
-- Epoch 283
Norm: 249.75, NNZs: 10, Bias: 151.839586, T: 100182, Avg. loss: 1991.434701
Total training time: 0.01 seconds.
-- Epoch 284
Norm: 250.25, NNZs: 10, Bias: 151.869171, T: 100536, Avg. loss: 1990.173090
Total training time: 0.01 seconds.
-- Epoch 285
Norm: 250.74, NNZs: 10, Bias: 151.885272, T: 100890, Avg. loss: 1988.858779
Total training time: 0.01 seconds.
-- Epoch 286
Norm: 251.23, NNZs: 10, Bias: 151.894772, T: 101244, Avg. loss: 1987.663095
Total training time: 0.01 seconds.
-- Epoch 287
Norm: 251.72, NNZs: 10, Bias: 151.841267, T: 101598, Avg. loss: 1986.381116
Total training time: 0.01 seconds.
-- Epoch 288
Norm: 252.21, NNZs: 10, Bias: 151.809339, T: 101952, Avg. loss: 1985.154307
Total training time: 0.01 seconds.
-- Epoch 289
Norm: 252.70, NNZs: 10, Bias: 151.760077, T: 102306, Avg. loss: 1983.875721
Total training time: 0.01 seconds.
-- Epoch 290
Norm: 253.19, NNZs: 10, Bias: 151.784865, T: 102660, Avg. loss: 1982.692881
Total training time: 0.01 seconds.
-- Epoch 291
Norm: 253.68, NNZs: 10, Bias: 151.820827, T: 103014, Avg. loss: 1981.454847
Total training time: 0.01 seconds.
-- Epoch 292
Norm: 254.16, NNZs: 10, Bias: 151.835305, T: 103368, Avg. loss: 1980.216921
Total training time: 0.01 seconds.
-- Epoch 293
Norm: 254.65, NNZs: 10, Bias: 151.847368, T: 103722, Avg. loss: 1979.009247
Total training time: 0.01 seconds.
-- Epoch 294
Norm: 255.13, NNZs: 10, Bias: 151.836495, T: 104076, Avg. loss: 1977.783335
Total training time: 0.01 seconds.
-- Epoch 295
Norm: 255.61, NNZs: 10, Bias: 151.890892, T: 104430, Avg. loss: 1976.519497
Total training time: 0.01 seconds.
-- Epoch 296
Norm: 256.10, NNZs: 10, Bias: 151.863321, T: 104784, Avg. loss: 1975.352784
Total training time: 0.01 seconds.
-- Epoch 297
Norm: 256.58, NNZs: 10, Bias: 151.925347, T: 105138, Avg. loss: 1974.078062
Total training time: 0.01 seconds.
-- Epoch 298
Norm: 257.06, NNZs: 10, Bias: 151.853483, T: 105492, Avg. loss: 1972.892257
Total training time: 0.01 seconds.
-- Epoch 299
Norm: 257.53, NNZs: 10, Bias: 151.857010, T: 105846, Avg. loss: 1971.758584
Total training time: 0.01 seconds.
-- Epoch 300
Norm: 258.01, NNZs: 10, Bias: 151.840926, T: 106200, Avg. loss: 1970.540952
Total training time: 0.01 seconds.
-- Epoch 301
Norm: 258.49, NNZs: 10, Bias: 151.849551, T: 106554, Avg. loss: 1969.347129
Total training time: 0.01 seconds.
-- Epoch 302
Norm: 258.96, NNZs: 10, Bias: 151.855584, T: 106908, Avg. loss: 1968.180907
Total training time: 0.01 seconds.
-- Epoch 303
Norm: 259.44, NNZs: 10, Bias: 151.830851, T: 107262, Avg. loss: 1966.989525
Total training time: 0.01 seconds.
-- Epoch 304
Norm: 259.91, NNZs: 10, Bias: 151.822350, T: 107616, Avg. loss: 1965.804244
Total training time: 0.01 seconds.
-- Epoch 305
Norm: 260.38, NNZs: 10, Bias: 151.800098, T: 107970, Avg. loss: 1964.638247
Total training time: 0.01 seconds.
-- Epoch 306
Norm: 260.85, NNZs: 10, Bias: 151.843183, T: 108324, Avg. loss: 1963.453398
Total training time: 0.01 seconds.
-- Epoch 307
Norm: 261.32, NNZs: 10, Bias: 151.849388, T: 108678, Avg. loss: 1962.306198
Total training time: 0.01 seconds.
-- Epoch 308
Norm: 261.79, NNZs: 10, Bias: 151.857021, T: 109032, Avg. loss: 1961.150579
Total training time: 0.01 seconds.
-- Epoch 309
Norm: 262.26, NNZs: 10, Bias: 151.800960, T: 109386, Avg. loss: 1959.935109
Total training time: 0.01 seconds.
-- Epoch 310
Norm: 262.73, NNZs: 10, Bias: 151.875862, T: 109740, Avg. loss: 1958.752587
Total training time: 0.01 seconds.
-- Epoch 311
Norm: 263.19, NNZs: 10, Bias: 151.790014, T: 110094, Avg. loss: 1957.513459
Total training time: 0.01 seconds.
-- Epoch 312
Norm: 263.66, NNZs: 10, Bias: 151.815671, T: 110448, Avg. loss: 1956.538711
Total training time: 0.01 seconds.
-- Epoch 313
Norm: 264.12, NNZs: 10, Bias: 151.847008, T: 110802, Avg. loss: 1955.374483
Total training time: 0.01 seconds.
-- Epoch 314
Norm: 264.59, NNZs: 10, Bias: 151.774522, T: 111156, Avg. loss: 1954.159757
Total training time: 0.01 seconds.
-- Epoch 315
Norm: 265.05, NNZs: 10, Bias: 151.807517, T: 111510, Avg. loss: 1953.101223
Total training time: 0.01 seconds.
-- Epoch 316
Norm: 265.51, NNZs: 10, Bias: 151.791879, T: 111864, Avg. loss: 1951.980565
Total training time: 0.01 seconds.
-- Epoch 317
Norm: 265.97, NNZs: 10, Bias: 151.831519, T: 112218, Avg. loss: 1950.838165
Total training time: 0.01 seconds.
-- Epoch 318
Norm: 266.43, NNZs: 10, Bias: 151.811982, T: 112572, Avg. loss: 1949.727590
Total training time: 0.01 seconds.
-- Epoch 319
Norm: 266.89, NNZs: 10, Bias: 151.788747, T: 112926, Avg. loss: 1948.607592
Total training time: 0.01 seconds.
-- Epoch 320
Norm: 267.34, NNZs: 10, Bias: 151.793328, T: 113280, Avg. loss: 1947.506424
Total training time: 0.01 seconds.
-- Epoch 321
Norm: 267.80, NNZs: 10, Bias: 151.830975, T: 113634, Avg. loss: 1946.368960
Total training time: 0.01 seconds.
-- Epoch 322
Norm: 268.25, NNZs: 10, Bias: 151.814733, T: 113988, Avg. loss: 1945.276772
Total training time: 0.01 seconds.
-- Epoch 323
Norm: 268.71, NNZs: 10, Bias: 151.836013, T: 114342, Avg. loss: 1944.164553
Total training time: 0.01 seconds.
-- Epoch 324
Norm: 269.16, NNZs: 10, Bias: 151.871691, T: 114696, Avg. loss: 1943.037416
Total training time: 0.01 seconds.
-- Epoch 325
Norm: 269.61, NNZs: 10, Bias: 151.853862, T: 115050, Avg. loss: 1941.972861
Total training time: 0.01 seconds.
-- Epoch 326
Norm: 270.07, NNZs: 10, Bias: 151.854440, T: 115404, Avg. loss: 1940.878910
Total training time: 0.01 seconds.
-- Epoch 327
Norm: 270.52, NNZs: 10, Bias: 151.797679, T: 115758, Avg. loss: 1939.716802
Total training time: 0.01 seconds.
-- Epoch 328
Norm: 270.97, NNZs: 10, Bias: 151.813845, T: 116112, Avg. loss: 1938.695423
Total training time: 0.01 seconds.
-- Epoch 329
Norm: 271.41, NNZs: 10, Bias: 151.786378, T: 116466, Avg. loss: 1937.604581
Total training time: 0.01 seconds.
-- Epoch 330
Norm: 271.86, NNZs: 10, Bias: 151.775234, T: 116820, Avg. loss: 1936.521583
Total training time: 0.01 seconds.
-- Epoch 331
Norm: 272.31, NNZs: 10, Bias: 151.781302, T: 117174, Avg. loss: 1935.458277
Total training time: 0.01 seconds.
-- Epoch 332
Norm: 272.76, NNZs: 10, Bias: 151.826461, T: 117528, Avg. loss: 1934.343741
Total training time: 0.01 seconds.
-- Epoch 333
Norm: 273.20, NNZs: 10, Bias: 151.858923, T: 117882, Avg. loss: 1933.291959
Total training time: 0.01 seconds.
-- Epoch 334
Norm: 273.64, NNZs: 10, Bias: 151.849628, T: 118236, Avg. loss: 1932.248567
Total training time: 0.01 seconds.
-- Epoch 335
Norm: 274.09, NNZs: 10, Bias: 151.837540, T: 118590, Avg. loss: 1931.186991
Total training time: 0.01 seconds.
-- Epoch 336
Norm: 274.53, NNZs: 10, Bias: 151.876660, T: 118944, Avg. loss: 1930.091205
Total training time: 0.01 seconds.
-- Epoch 337
Norm: 274.97, NNZs: 10, Bias: 151.871067, T: 119298, Avg. loss: 1929.072482
Total training time: 0.01 seconds.
-- Epoch 338
Norm: 275.41, NNZs: 10, Bias: 151.858856, T: 119652, Avg. loss: 1928.021247
Total training time: 0.01 seconds.
-- Epoch 339
Norm: 275.85, NNZs: 10, Bias: 151.871975, T: 120006, Avg. loss: 1926.965509
Total training time: 0.01 seconds.
-- Epoch 340
Norm: 276.29, NNZs: 10, Bias: 151.908408, T: 120360, Avg. loss: 1925.885064
Total training time: 0.01 seconds.
-- Epoch 341
Norm: 276.73, NNZs: 10, Bias: 151.929034, T: 120714, Avg. loss: 1924.871818
Total training time: 0.01 seconds.
-- Epoch 342
Norm: 277.16, NNZs: 10, Bias: 151.940664, T: 121068, Avg. loss: 1923.829158
Total training time: 0.01 seconds.
-- Epoch 343
Norm: 277.60, NNZs: 10, Bias: 151.940635, T: 121422, Avg. loss: 1922.773926
Total training time: 0.01 seconds.
-- Epoch 344
Norm: 278.03, NNZs: 10, Bias: 151.922979, T: 121776, Avg. loss: 1921.776987
Total training time: 0.01 seconds.
-- Epoch 345
Norm: 278.47, NNZs: 10, Bias: 151.949367, T: 122130, Avg. loss: 1920.709816
Total training time: 0.01 seconds.
-- Epoch 346
Norm: 278.90, NNZs: 10, Bias: 151.912099, T: 122484, Avg. loss: 1919.714192
Total training time: 0.01 seconds.
-- Epoch 347
Norm: 279.33, NNZs: 10, Bias: 151.858664, T: 122838, Avg. loss: 1918.655879
Total training time: 0.01 seconds.
-- Epoch 348
Norm: 279.77, NNZs: 10, Bias: 151.876150, T: 123192, Avg. loss: 1917.666604
Total training time: 0.01 seconds.
-- Epoch 349
Norm: 280.20, NNZs: 10, Bias: 151.845175, T: 123546, Avg. loss: 1916.657465
Total training time: 0.01 seconds.
-- Epoch 350
Norm: 280.63, NNZs: 10, Bias: 151.873288, T: 123900, Avg. loss: 1915.634142
Total training time: 0.01 seconds.
-- Epoch 351
Norm: 281.05, NNZs: 10, Bias: 151.882618, T: 124254, Avg. loss: 1914.634351
Total training time: 0.01 seconds.
-- Epoch 352
Norm: 281.48, NNZs: 10, Bias: 151.850762, T: 124608, Avg. loss: 1913.627719
Total training time: 0.01 seconds.
-- Epoch 353
Norm: 281.91, NNZs: 10, Bias: 151.891092, T: 124962, Avg. loss: 1912.588862
Total training time: 0.01 seconds.
-- Epoch 354
Norm: 282.34, NNZs: 10, Bias: 151.888719, T: 125316, Avg. loss: 1911.621884
Total training time: 0.01 seconds.
-- Epoch 355
Norm: 282.76, NNZs: 10, Bias: 151.884977, T: 125670, Avg. loss: 1910.623489
Total training time: 0.01 seconds.
-- Epoch 356
Norm: 283.19, NNZs: 10, Bias: 151.843097, T: 126024, Avg. loss: 1909.601944
Total training time: 0.01 seconds.
-- Epoch 357
Norm: 283.61, NNZs: 10, Bias: 151.835150, T: 126378, Avg. loss: 1908.629853
Total training time: 0.01 seconds.
-- Epoch 358
Norm: 284.03, NNZs: 10, Bias: 151.782847, T: 126732, Avg. loss: 1907.606836
Total training time: 0.01 seconds.
-- Epoch 359
Norm: 284.46, NNZs: 10, Bias: 151.776379, T: 127086, Avg. loss: 1906.673425
Total training time: 0.01 seconds.
-- Epoch 360
Norm: 284.88, NNZs: 10, Bias: 151.767956, T: 127440, Avg. loss: 1905.680828
Total training time: 0.01 seconds.
-- Epoch 361
Norm: 285.30, NNZs: 10, Bias: 151.820348, T: 127794, Avg. loss: 1904.681709
Total training time: 0.01 seconds.
-- Epoch 362
Norm: 285.72, NNZs: 10, Bias: 151.842241, T: 128148, Avg. loss: 1903.730618
Total training time: 0.01 seconds.
-- Epoch 363
Norm: 286.14, NNZs: 10, Bias: 151.884490, T: 128502, Avg. loss: 1902.715279
Total training time: 0.01 seconds.
-- Epoch 364
Norm: 286.55, NNZs: 10, Bias: 151.830336, T: 128856, Avg. loss: 1901.733941
Total training time: 0.01 seconds.
-- Epoch 365
Norm: 286.97, NNZs: 10, Bias: 151.830913, T: 129210, Avg. loss: 1900.827720
Total training time: 0.01 seconds.
-- Epoch 366
Norm: 287.39, NNZs: 10, Bias: 151.841200, T: 129564, Avg. loss: 1899.860671
Total training time: 0.01 seconds.
-- Epoch 367
Norm: 287.80, NNZs: 10, Bias: 151.815968, T: 129918, Avg. loss: 1898.877141
Total training time: 0.01 seconds.
-- Epoch 368
Norm: 288.22, NNZs: 10, Bias: 151.871526, T: 130272, Avg. loss: 1897.862627
Total training time: 0.01 seconds.
-- Epoch 369
Norm: 288.63, NNZs: 10, Bias: 151.881507, T: 130626, Avg. loss: 1896.983225
Total training time: 0.01 seconds.
-- Epoch 370
Norm: 289.05, NNZs: 10, Bias: 151.872414, T: 130980, Avg. loss: 1896.031126
Total training time: 0.01 seconds.
-- Epoch 371
Norm: 289.46, NNZs: 10, Bias: 151.855044, T: 131334, Avg. loss: 1895.086829
Total training time: 0.01 seconds.
-- Epoch 372
Norm: 289.87, NNZs: 10, Bias: 151.841038, T: 131688, Avg. loss: 1894.134816
Total training time: 0.01 seconds.
-- Epoch 373
Norm: 290.28, NNZs: 10, Bias: 151.822225, T: 132042, Avg. loss: 1893.183290
Total training time: 0.01 seconds.
-- Epoch 374
Norm: 290.69, NNZs: 10, Bias: 151.799626, T: 132396, Avg. loss: 1892.234626
Total training time: 0.01 seconds.
-- Epoch 375
Norm: 291.10, NNZs: 10, Bias: 151.770084, T: 132750, Avg. loss: 1891.289612
Total training time: 0.01 seconds.
-- Epoch 376
Norm: 291.51, NNZs: 10, Bias: 151.798903, T: 133104, Avg. loss: 1890.375627
Total training time: 0.01 seconds.
-- Epoch 377
Norm: 291.92, NNZs: 10, Bias: 151.825406, T: 133458, Avg. loss: 1889.435946
Total training time: 0.01 seconds.
-- Epoch 378
Norm: 292.32, NNZs: 10, Bias: 151.841162, T: 133812, Avg. loss: 1888.504488
Total training time: 0.01 seconds.
-- Epoch 379
Norm: 292.73, NNZs: 10, Bias: 151.887839, T: 134166, Avg. loss: 1887.534365
Total training time: 0.01 seconds.
-- Epoch 380
Norm: 293.13, NNZs: 10, Bias: 151.824170, T: 134520, Avg. loss: 1886.620650
Total training time: 0.01 seconds.
-- Epoch 381
Norm: 293.54, NNZs: 10, Bias: 151.782517, T: 134874, Avg. loss: 1885.720865
Total training time: 0.01 seconds.
-- Epoch 382
Norm: 293.94, NNZs: 10, Bias: 151.773598, T: 135228, Avg. loss: 1884.831066
Total training time: 0.01 seconds.
-- Epoch 383
Norm: 294.35, NNZs: 10, Bias: 151.752653, T: 135582, Avg. loss: 1883.909736
Total training time: 0.01 seconds.
-- Epoch 384
Norm: 294.75, NNZs: 10, Bias: 151.744749, T: 135936, Avg. loss: 1883.005565
Total training time: 0.01 seconds.
-- Epoch 385
Norm: 295.15, NNZs: 10, Bias: 151.708995, T: 136290, Avg. loss: 1882.046467
Total training time: 0.01 seconds.
-- Epoch 386
Norm: 295.55, NNZs: 10, Bias: 151.746835, T: 136644, Avg. loss: 1881.183347
Total training time: 0.01 seconds.
-- Epoch 387
Norm: 295.95, NNZs: 10, Bias: 151.744922, T: 136998, Avg. loss: 1880.287999
Total training time: 0.01 seconds.
-- Epoch 388
Norm: 296.35, NNZs: 10, Bias: 151.750822, T: 137352, Avg. loss: 1879.386353
Total training time: 0.01 seconds.
-- Epoch 389
Norm: 296.75, NNZs: 10, Bias: 151.790166, T: 137706, Avg. loss: 1878.467427
Total training time: 0.01 seconds.
-- Epoch 390
Norm: 297.15, NNZs: 10, Bias: 151.744761, T: 138060, Avg. loss: 1877.533452
Total training time: 0.01 seconds.
-- Epoch 391
Norm: 297.54, NNZs: 10, Bias: 151.739591, T: 138414, Avg. loss: 1876.683469
Total training time: 0.01 seconds.
-- Epoch 392
Norm: 297.94, NNZs: 10, Bias: 151.783603, T: 138768, Avg. loss: 1875.790555
Total training time: 0.01 seconds.
-- Epoch 393
Norm: 298.33, NNZs: 10, Bias: 151.792167, T: 139122, Avg. loss: 1874.916939
Total training time: 0.01 seconds.
-- Epoch 394
Norm: 298.73, NNZs: 10, Bias: 151.778661, T: 139476, Avg. loss: 1874.021123
Total training time: 0.01 seconds.
-- Epoch 395
Norm: 299.12, NNZs: 10, Bias: 151.807864, T: 139830, Avg. loss: 1873.133345
Total training time: 0.01 seconds.
-- Epoch 396
Norm: 299.52, NNZs: 10, Bias: 151.790795, T: 140184, Avg. loss: 1872.256075
Total training time: 0.01 seconds.
-- Epoch 397
Norm: 299.91, NNZs: 10, Bias: 151.792630, T: 140538, Avg. loss: 1871.387573
Total training time: 0.01 seconds.
-- Epoch 398
Norm: 300.30, NNZs: 10, Bias: 151.804573, T: 140892, Avg. loss: 1870.512556
Total training time: 0.01 seconds.
-- Epoch 399
Norm: 300.69, NNZs: 10, Bias: 151.833679, T: 141246, Avg. loss: 1869.623324
Total training time: 0.01 seconds.
-- Epoch 400
Norm: 301.08, NNZs: 10, Bias: 151.843734, T: 141600, Avg. loss: 1868.765177
Total training time: 0.01 seconds.
-- Epoch 401
Norm: 301.47, NNZs: 10, Bias: 151.830018, T: 141954, Avg. loss: 1867.894988
Total training time: 0.01 seconds.
-- Epoch 402
Norm: 301.86, NNZs: 10, Bias: 151.845513, T: 142308, Avg. loss: 1867.029438
Total training time: 0.01 seconds.
-- Epoch 403
Norm: 302.25, NNZs: 10, Bias: 151.846403, T: 142662, Avg. loss: 1866.169523
Total training time: 0.01 seconds.
-- Epoch 404
Norm: 302.64, NNZs: 10, Bias: 151.878304, T: 143016, Avg. loss: 1865.283535
Total training time: 0.01 seconds.
-- Epoch 405
Norm: 303.02, NNZs: 10, Bias: 151.824278, T: 143370, Avg. loss: 1864.419448
Total training time: 0.01 seconds.
-- Epoch 406
Norm: 303.41, NNZs: 10, Bias: 151.849851, T: 143724, Avg. loss: 1863.572769
Total training time: 0.01 seconds.
-- Epoch 407
Norm: 303.80, NNZs: 10, Bias: 151.803377, T: 144078, Avg. loss: 1862.711245
Total training time: 0.01 seconds.
-- Epoch 408
Norm: 304.18, NNZs: 10, Bias: 151.837679, T: 144432, Avg. loss: 1861.874686
Total training time: 0.01 seconds.
-- Epoch 409
Norm: 304.57, NNZs: 10, Bias: 151.860433, T: 144786, Avg. loss: 1861.024784
Total training time: 0.01 seconds.
-- Epoch 410
Norm: 304.95, NNZs: 10, Bias: 151.831386, T: 145140, Avg. loss: 1860.194553
Total training time: 0.01 seconds.
-- Epoch 411
Norm: 305.33, NNZs: 10, Bias: 151.797841, T: 145494, Avg. loss: 1859.333971
Total training time: 0.01 seconds.
-- Epoch 412
Norm: 305.71, NNZs: 10, Bias: 151.836628, T: 145848, Avg. loss: 1858.491586
Total training time: 0.01 seconds.
-- Epoch 413
Norm: 306.09, NNZs: 10, Bias: 151.869685, T: 146202, Avg. loss: 1857.652005
Total training time: 0.01 seconds.
-- Epoch 414
Norm: 306.48, NNZs: 10, Bias: 151.839405, T: 146556, Avg. loss: 1856.833343
Total training time: 0.01 seconds.
-- Epoch 415
Norm: 306.85, NNZs: 10, Bias: 151.872772, T: 146910, Avg. loss: 1855.968703
Total training time: 0.01 seconds.
-- Epoch 416
Norm: 307.23, NNZs: 10, Bias: 151.879927, T: 147264, Avg. loss: 1855.152912
Total training time: 0.01 seconds.
-- Epoch 417
Norm: 307.61, NNZs: 10, Bias: 151.853764, T: 147618, Avg. loss: 1854.346226
Total training time: 0.01 seconds.
-- Epoch 418
Norm: 307.99, NNZs: 10, Bias: 151.846779, T: 147972, Avg. loss: 1853.516738
Total training time: 0.01 seconds.
-- Epoch 419
Norm: 308.37, NNZs: 10, Bias: 151.885015, T: 148326, Avg. loss: 1852.627490
Total training time: 0.01 seconds.
-- Epoch 420
Norm: 308.74, NNZs: 10, Bias: 151.853495, T: 148680, Avg. loss: 1851.872254
Total training time: 0.01 seconds.
-- Epoch 421
Norm: 309.12, NNZs: 10, Bias: 151.898840, T: 149034, Avg. loss: 1850.976984
Total training time: 0.01 seconds.
-- Epoch 422
Norm: 309.49, NNZs: 10, Bias: 151.897111, T: 149388, Avg. loss: 1850.242445
Total training time: 0.01 seconds.
-- Epoch 423
Norm: 309.87, NNZs: 10, Bias: 151.873426, T: 149742, Avg. loss: 1849.427544
Total training time: 0.01 seconds.
-- Epoch 424
Norm: 310.24, NNZs: 10, Bias: 151.870216, T: 150096, Avg. loss: 1848.611936
Total training time: 0.01 seconds.
-- Epoch 425
Norm: 310.62, NNZs: 10, Bias: 151.876016, T: 150450, Avg. loss: 1847.801872
Total training time: 0.01 seconds.
-- Epoch 426
Norm: 310.99, NNZs: 10, Bias: 151.828023, T: 150804, Avg. loss: 1846.963070
Total training time: 0.01 seconds.
-- Epoch 427
Norm: 311.36, NNZs: 10, Bias: 151.790111, T: 151158, Avg. loss: 1846.156386
Total training time: 0.01 seconds.
-- Epoch 428
Norm: 311.73, NNZs: 10, Bias: 151.797134, T: 151512, Avg. loss: 1845.387187
Total training time: 0.01 seconds.
-- Epoch 429
Norm: 312.10, NNZs: 10, Bias: 151.805149, T: 151866, Avg. loss: 1844.583691
Total training time: 0.01 seconds.
-- Epoch 430
Norm: 312.47, NNZs: 10, Bias: 151.806762, T: 152220, Avg. loss: 1843.781917
Total training time: 0.01 seconds.
-- Epoch 431
Norm: 312.84, NNZs: 10, Bias: 151.806589, T: 152574, Avg. loss: 1842.994480
Total training time: 0.01 seconds.
-- Epoch 432
Norm: 313.21, NNZs: 10, Bias: 151.827450, T: 152928, Avg. loss: 1842.187937
Total training time: 0.01 seconds.
-- Epoch 433
Norm: 313.58, NNZs: 10, Bias: 151.799551, T: 153282, Avg. loss: 1841.379875
Total training time: 0.01 seconds.
-- Epoch 434
Norm: 313.94, NNZs: 10, Bias: 151.819622, T: 153636, Avg. loss: 1840.605410
Total training time: 0.01 seconds.
-- Epoch 435
Norm: 314.31, NNZs: 10, Bias: 151.772212, T: 153990, Avg. loss: 1839.762942
Total training time: 0.01 seconds.
-- Epoch 436
Norm: 314.68, NNZs: 10, Bias: 151.797740, T: 154344, Avg. loss: 1839.018980
Total training time: 0.01 seconds.
-- Epoch 437
Norm: 315.04, NNZs: 10, Bias: 151.849069, T: 154698, Avg. loss: 1838.203368
Total training time: 0.01 seconds.
-- Epoch 438
Norm: 315.41, NNZs: 10, Bias: 151.841834, T: 155052, Avg. loss: 1837.477325
Total training time: 0.01 seconds.
-- Epoch 439
Norm: 315.77, NNZs: 10, Bias: 151.857622, T: 155406, Avg. loss: 1836.680591
Total training time: 0.01 seconds.
-- Epoch 440
Norm: 316.14, NNZs: 10, Bias: 151.812507, T: 155760, Avg. loss: 1835.891556
Total training time: 0.01 seconds.
-- Epoch 441
Norm: 316.50, NNZs: 10, Bias: 151.809462, T: 156114, Avg. loss: 1835.142941
Total training time: 0.01 seconds.
-- Epoch 442
Norm: 316.86, NNZs: 10, Bias: 151.814887, T: 156468, Avg. loss: 1834.361330
Total training time: 0.01 seconds.
-- Epoch 443
Norm: 317.22, NNZs: 10, Bias: 151.761839, T: 156822, Avg. loss: 1833.535290
Total training time: 0.01 seconds.
-- Epoch 444
Norm: 317.58, NNZs: 10, Bias: 151.753258, T: 157176, Avg. loss: 1832.821192
Total training time: 0.01 seconds.
-- Epoch 445
Norm: 317.94, NNZs: 10, Bias: 151.811247, T: 157530, Avg. loss: 1832.004525
Total training time: 0.01 seconds.
-- Epoch 446
Norm: 318.30, NNZs: 10, Bias: 151.813460, T: 157884, Avg. loss: 1831.296201
Total training time: 0.01 seconds.
-- Epoch 447
Norm: 318.66, NNZs: 10, Bias: 151.807328, T: 158238, Avg. loss: 1830.539629
Total training time: 0.01 seconds.
-- Epoch 448
Norm: 319.02, NNZs: 10, Bias: 151.837602, T: 158592, Avg. loss: 1829.751681
Total training time: 0.01 seconds.
-- Epoch 449
Norm: 319.38, NNZs: 10, Bias: 151.832731, T: 158946, Avg. loss: 1829.018703
Total training time: 0.01 seconds.
-- Epoch 450
Norm: 319.74, NNZs: 10, Bias: 151.791564, T: 159300, Avg. loss: 1828.228821
Total training time: 0.01 seconds.
-- Epoch 451
Norm: 320.09, NNZs: 10, Bias: 151.776192, T: 159654, Avg. loss: 1827.505483
Total training time: 0.01 seconds.
-- Epoch 452
Norm: 320.45, NNZs: 10, Bias: 151.769686, T: 160008, Avg. loss: 1826.752203
Total training time: 0.01 seconds.
-- Epoch 453
Norm: 320.80, NNZs: 10, Bias: 151.742934, T: 160362, Avg. loss: 1825.970786
Total training time: 0.01 seconds.
-- Epoch 454
Norm: 321.16, NNZs: 10, Bias: 151.772084, T: 160716, Avg. loss: 1825.241765
Total training time: 0.01 seconds.
-- Epoch 455
Norm: 321.51, NNZs: 10, Bias: 151.792175, T: 161070, Avg. loss: 1824.497435
Total training time: 0.01 seconds.
-- Epoch 456
Norm: 321.87, NNZs: 10, Bias: 151.777265, T: 161424, Avg. loss: 1823.768367
Total training time: 0.01 seconds.
-- Epoch 457
Norm: 322.22, NNZs: 10, Bias: 151.802434, T: 161778, Avg. loss: 1823.016004
Total training time: 0.01 seconds.
-- Epoch 458
Norm: 322.57, NNZs: 10, Bias: 151.803854, T: 162132, Avg. loss: 1822.289300
Total training time: 0.01 seconds.
-- Epoch 459
Norm: 322.92, NNZs: 10, Bias: 151.829934, T: 162486, Avg. loss: 1821.487047
Total training time: 0.01 seconds.
-- Epoch 460
Norm: 323.27, NNZs: 10, Bias: 151.813813, T: 162840, Avg. loss: 1820.815579
Total training time: 0.01 seconds.
-- Epoch 461
Norm: 323.62, NNZs: 10, Bias: 151.789195, T: 163194, Avg. loss: 1820.070873
Total training time: 0.01 seconds.
-- Epoch 462
Norm: 323.97, NNZs: 10, Bias: 151.773100, T: 163548, Avg. loss: 1819.338710
Total training time: 0.01 seconds.
-- Epoch 463
Norm: 324.32, NNZs: 10, Bias: 151.769463, T: 163902, Avg. loss: 1818.616404
Total training time: 0.01 seconds.
-- Epoch 464
Norm: 324.67, NNZs: 10, Bias: 151.773390, T: 164256, Avg. loss: 1817.891217
Total training time: 0.01 seconds.
-- Epoch 465
Norm: 325.02, NNZs: 10, Bias: 151.793736, T: 164610, Avg. loss: 1817.163867
Total training time: 0.01 seconds.
-- Epoch 466
Norm: 325.37, NNZs: 10, Bias: 151.799382, T: 164964, Avg. loss: 1816.435516
Total training time: 0.01 seconds.
-- Epoch 467
Norm: 325.72, NNZs: 10, Bias: 151.774839, T: 165318, Avg. loss: 1815.700731
Total training time: 0.01 seconds.
-- Epoch 468
Norm: 326.06, NNZs: 10, Bias: 151.789175, T: 165672, Avg. loss: 1814.995327
Total training time: 0.01 seconds.
-- Epoch 469
Norm: 326.41, NNZs: 10, Bias: 151.765037, T: 166026, Avg. loss: 1814.251792
Total training time: 0.01 seconds.
-- Epoch 470
Norm: 326.75, NNZs: 10, Bias: 151.767522, T: 166380, Avg. loss: 1813.568334
Total training time: 0.01 seconds.
-- Epoch 471
Norm: 327.10, NNZs: 10, Bias: 151.795912, T: 166734, Avg. loss: 1812.842722
Total training time: 0.01 seconds.
-- Epoch 472
Norm: 327.44, NNZs: 10, Bias: 151.804199, T: 167088, Avg. loss: 1812.138050
Total training time: 0.01 seconds.
-- Epoch 473
Norm: 327.79, NNZs: 10, Bias: 151.774157, T: 167442, Avg. loss: 1811.409449
Total training time: 0.01 seconds.
-- Epoch 474
Norm: 328.13, NNZs: 10, Bias: 151.787081, T: 167796, Avg. loss: 1810.717721
Total training time: 0.01 seconds.
-- Epoch 475
Norm: 328.47, NNZs: 10, Bias: 151.853654, T: 168150, Avg. loss: 1809.909467
Total training time: 0.01 seconds.
-- Epoch 476
Norm: 328.81, NNZs: 10, Bias: 151.836005, T: 168504, Avg. loss: 1809.302768
Total training time: 0.01 seconds.
-- Epoch 477
Norm: 329.15, NNZs: 10, Bias: 151.842293, T: 168858, Avg. loss: 1808.590074
Total training time: 0.01 seconds.
-- Epoch 478
Norm: 329.50, NNZs: 10, Bias: 151.882249, T: 169212, Avg. loss: 1807.843908
Total training time: 0.01 seconds.
-- Epoch 479
Norm: 329.84, NNZs: 10, Bias: 151.837707, T: 169566, Avg. loss: 1807.177733
Total training time: 0.01 seconds.
-- Epoch 480
Norm: 330.17, NNZs: 10, Bias: 151.801300, T: 169920, Avg. loss: 1806.471595
Total training time: 0.01 seconds.
-- Epoch 481
Norm: 330.51, NNZs: 10, Bias: 151.796158, T: 170274, Avg. loss: 1805.805653
Total training time: 0.01 seconds.
-- Epoch 482
Norm: 330.85, NNZs: 10, Bias: 151.749016, T: 170628, Avg. loss: 1805.061712
Total training time: 0.01 seconds.
-- Epoch 483
Norm: 331.19, NNZs: 10, Bias: 151.727932, T: 170982, Avg. loss: 1804.396931
Total training time: 0.01 seconds.
-- Epoch 484
Norm: 331.53, NNZs: 10, Bias: 151.735555, T: 171336, Avg. loss: 1803.722340
Total training time: 0.01 seconds.
-- Epoch 485
Norm: 331.86, NNZs: 10, Bias: 151.780159, T: 171690, Avg. loss: 1803.014799
Total training time: 0.01 seconds.
-- Epoch 486
Norm: 332.20, NNZs: 10, Bias: 151.777862, T: 172044, Avg. loss: 1802.351707
Total training time: 0.01 seconds.
-- Epoch 487
Norm: 332.54, NNZs: 10, Bias: 151.801707, T: 172398, Avg. loss: 1801.657387
Total training time: 0.01 seconds.
-- Epoch 488
Norm: 332.87, NNZs: 10, Bias: 151.810654, T: 172752, Avg. loss: 1800.981908
Total training time: 0.01 seconds.
-- Epoch 489
Norm: 333.21, NNZs: 10, Bias: 151.829644, T: 173106, Avg. loss: 1800.294557
Total training time: 0.01 seconds.
-- Epoch 490
Norm: 333.54, NNZs: 10, Bias: 151.804191, T: 173460, Avg. loss: 1799.614428
Total training time: 0.01 seconds.
-- Epoch 491
Norm: 333.87, NNZs: 10, Bias: 151.812676, T: 173814, Avg. loss: 1798.925647
Total training time: 0.01 seconds.
-- Epoch 492
Norm: 334.21, NNZs: 10, Bias: 151.841773, T: 174168, Avg. loss: 1798.245530
Total training time: 0.01 seconds.
-- Epoch 493
Norm: 334.54, NNZs: 10, Bias: 151.853552, T: 174522, Avg. loss: 1797.587929
Total training time: 0.01 seconds.
-- Epoch 494
Norm: 334.87, NNZs: 10, Bias: 151.839185, T: 174876, Avg. loss: 1796.921773
Total training time: 0.01 seconds.
-- Epoch 495
Norm: 335.20, NNZs: 10, Bias: 151.857257, T: 175230, Avg. loss: 1796.232814
Total training time: 0.01 seconds.
-- Epoch 496
Norm: 335.53, NNZs: 10, Bias: 151.845527, T: 175584, Avg. loss: 1795.581804
Total training time: 0.01 seconds.
-- Epoch 497
Norm: 335.86, NNZs: 10, Bias: 151.836592, T: 175938, Avg. loss: 1794.908485
Total training time: 0.01 seconds.
-- Epoch 498
Norm: 336.19, NNZs: 10, Bias: 151.851769, T: 176292, Avg. loss: 1794.234074
Total training time: 0.01 seconds.
-- Epoch 499
Norm: 336.52, NNZs: 10, Bias: 151.819260, T: 176646, Avg. loss: 1793.570466
Total training time: 0.01 seconds.
-- Epoch 500
Norm: 336.85, NNZs: 10, Bias: 151.867657, T: 177000, Avg. loss: 1792.853582
Total training time: 0.01 seconds.
-- Epoch 501
Norm: 337.18, NNZs: 10, Bias: 151.873639, T: 177354, Avg. loss: 1792.254937
Total training time: 0.01 seconds.
-- Epoch 502
Norm: 337.51, NNZs: 10, Bias: 151.875755, T: 177708, Avg. loss: 1791.597460
Total training time: 0.01 seconds.
-- Epoch 503
Norm: 337.83, NNZs: 10, Bias: 151.839054, T: 178062, Avg. loss: 1790.932081
Total training time: 0.01 seconds.
-- Epoch 504
Norm: 338.16, NNZs: 10, Bias: 151.835625, T: 178416, Avg. loss: 1790.285457
Total training time: 0.01 seconds.
-- Epoch 505
Norm: 338.49, NNZs: 10, Bias: 151.815225, T: 178770, Avg. loss: 1789.630332
Total training time: 0.01 seconds.
-- Epoch 506
Norm: 338.81, NNZs: 10, Bias: 151.766255, T: 179124, Avg. loss: 1788.926076
Total training time: 0.01 seconds.
-- Epoch 507
Norm: 339.14, NNZs: 10, Bias: 151.761736, T: 179478, Avg. loss: 1788.329428
Total training time: 0.01 seconds.
-- Epoch 508
Norm: 339.46, NNZs: 10, Bias: 151.762594, T: 179832, Avg. loss: 1787.673973
Total training time: 0.01 seconds.
-- Epoch 509
Norm: 339.79, NNZs: 10, Bias: 151.767220, T: 180186, Avg. loss: 1787.028524
Total training time: 0.01 seconds.
-- Epoch 510
Norm: 340.11, NNZs: 10, Bias: 151.769997, T: 180540, Avg. loss: 1786.382632
Total training time: 0.01 seconds.
-- Epoch 511
Norm: 340.43, NNZs: 10, Bias: 151.764516, T: 180894, Avg. loss: 1785.721043
Total training time: 0.01 seconds.
-- Epoch 512
Norm: 340.76, NNZs: 10, Bias: 151.793496, T: 181248, Avg. loss: 1785.073573
Total training time: 0.01 seconds.
-- Epoch 513
Norm: 341.08, NNZs: 10, Bias: 151.801079, T: 181602, Avg. loss: 1784.452096
Total training time: 0.01 seconds.
-- Epoch 514
Norm: 341.40, NNZs: 10, Bias: 151.792275, T: 181956, Avg. loss: 1783.812153
Total training time: 0.01 seconds.
-- Epoch 515
Norm: 341.72, NNZs: 10, Bias: 151.775715, T: 182310, Avg. loss: 1783.170044
Total training time: 0.01 seconds.
-- Epoch 516
Norm: 342.04, NNZs: 10, Bias: 151.766994, T: 182664, Avg. loss: 1782.537581
Total training time: 0.01 seconds.
-- Epoch 517
Norm: 342.36, NNZs: 10, Bias: 151.768695, T: 183018, Avg. loss: 1781.903908
Total training time: 0.01 seconds.
-- Epoch 518
Norm: 342.68, NNZs: 10, Bias: 151.811649, T: 183372, Avg. loss: 1781.234952
Total training time: 0.01 seconds.
-- Epoch 519
Norm: 343.00, NNZs: 10, Bias: 151.772663, T: 183726, Avg. loss: 1780.592979
Total training time: 0.01 seconds.
-- Epoch 520
Norm: 343.32, NNZs: 10, Bias: 151.773287, T: 184080, Avg. loss: 1780.007148
Total training time: 0.01 seconds.
-- Epoch 521
Norm: 343.63, NNZs: 10, Bias: 151.799435, T: 184434, Avg. loss: 1779.368650
Total training time: 0.01 seconds.
-- Epoch 522
Norm: 343.95, NNZs: 10, Bias: 151.773148, T: 184788, Avg. loss: 1778.731109
Total training time: 0.01 seconds.
-- Epoch 523
Norm: 344.27, NNZs: 10, Bias: 151.782094, T: 185142, Avg. loss: 1778.131846
Total training time: 0.01 seconds.
-- Epoch 524
Norm: 344.59, NNZs: 10, Bias: 151.786174, T: 185496, Avg. loss: 1777.504175
Total training time: 0.01 seconds.
-- Epoch 525
Norm: 344.90, NNZs: 10, Bias: 151.786411, T: 185850, Avg. loss: 1776.881900
Total training time: 0.01 seconds.
-- Epoch 526
Norm: 345.22, NNZs: 10, Bias: 151.763856, T: 186204, Avg. loss: 1776.249090
Total training time: 0.01 seconds.
-- Epoch 527
Norm: 345.53, NNZs: 10, Bias: 151.768907, T: 186558, Avg. loss: 1775.643790
Total training time: 0.01 seconds.
-- Epoch 528
Norm: 345.85, NNZs: 10, Bias: 151.796871, T: 186912, Avg. loss: 1775.013920
Total training time: 0.01 seconds.
-- Epoch 529
Norm: 346.16, NNZs: 10, Bias: 151.790716, T: 187266, Avg. loss: 1774.397659
Total training time: 0.01 seconds.
-- Epoch 530
Norm: 346.47, NNZs: 10, Bias: 151.802682, T: 187620, Avg. loss: 1773.788012
Total training time: 0.01 seconds.
-- Epoch 531
Norm: 346.79, NNZs: 10, Bias: 151.810468, T: 187974, Avg. loss: 1773.176548
Total training time: 0.01 seconds.
-- Epoch 532
Norm: 347.10, NNZs: 10, Bias: 151.803358, T: 188328, Avg. loss: 1772.570695
Total training time: 0.01 seconds.
-- Epoch 533
Norm: 347.41, NNZs: 10, Bias: 151.816254, T: 188682, Avg. loss: 1771.941705
Total training time: 0.01 seconds.
-- Epoch 534
Norm: 347.72, NNZs: 10, Bias: 151.835285, T: 189036, Avg. loss: 1771.338663
Total training time: 0.01 seconds.
-- Epoch 535
Norm: 348.03, NNZs: 10, Bias: 151.815415, T: 189390, Avg. loss: 1770.737049
Total training time: 0.01 seconds.
-- Epoch 536
Norm: 348.34, NNZs: 10, Bias: 151.835177, T: 189744, Avg. loss: 1770.127198
Total training time: 0.01 seconds.
-- Epoch 537
Norm: 348.66, NNZs: 10, Bias: 151.809475, T: 190098, Avg. loss: 1769.529839
Total training time: 0.01 seconds.
-- Epoch 538
Norm: 348.96, NNZs: 10, Bias: 151.778247, T: 190452, Avg. loss: 1768.911100
Total training time: 0.01 seconds.
-- Epoch 539
Norm: 349.27, NNZs: 10, Bias: 151.752045, T: 190806, Avg. loss: 1768.305499
Total training time: 0.01 seconds.
-- Epoch 540
Norm: 349.58, NNZs: 10, Bias: 151.750839, T: 191160, Avg. loss: 1767.706348
Total training time: 0.01 seconds.
-- Epoch 541
Norm: 349.89, NNZs: 10, Bias: 151.743093, T: 191514, Avg. loss: 1767.129930
Total training time: 0.01 seconds.
-- Epoch 542
Norm: 350.20, NNZs: 10, Bias: 151.759688, T: 191868, Avg. loss: 1766.531357
Total training time: 0.01 seconds.
-- Epoch 543
Norm: 350.51, NNZs: 10, Bias: 151.788028, T: 192222, Avg. loss: 1765.921020
Total training time: 0.01 seconds.
-- Epoch 544
Norm: 350.81, NNZs: 10, Bias: 151.772269, T: 192576, Avg. loss: 1765.322519
Total training time: 0.01 seconds.
-- Epoch 545
Norm: 351.12, NNZs: 10, Bias: 151.787755, T: 192930, Avg. loss: 1764.740131
Total training time: 0.01 seconds.
-- Epoch 546
Norm: 351.42, NNZs: 10, Bias: 151.767947, T: 193284, Avg. loss: 1764.148351
Total training time: 0.01 seconds.
-- Epoch 547
Norm: 351.73, NNZs: 10, Bias: 151.809541, T: 193638, Avg. loss: 1763.513856
Total training time: 0.01 seconds.
-- Epoch 548
Norm: 352.04, NNZs: 10, Bias: 151.813025, T: 193992, Avg. loss: 1762.980808
Total training time: 0.01 seconds.
-- Epoch 549
Norm: 352.34, NNZs: 10, Bias: 151.802857, T: 194346, Avg. loss: 1762.380716
Total training time: 0.01 seconds.
-- Epoch 550
Norm: 352.64, NNZs: 10, Bias: 151.791651, T: 194700, Avg. loss: 1761.809088
Total training time: 0.01 seconds.
-- Epoch 551
Norm: 352.95, NNZs: 10, Bias: 151.798591, T: 195054, Avg. loss: 1761.219095
Total training time: 0.01 seconds.
-- Epoch 552
Norm: 353.25, NNZs: 10, Bias: 151.776775, T: 195408, Avg. loss: 1760.620465
Total training time: 0.01 seconds.
-- Epoch 553
Norm: 353.55, NNZs: 10, Bias: 151.802342, T: 195762, Avg. loss: 1760.033987
Total training time: 0.01 seconds.
-- Epoch 554
Norm: 353.86, NNZs: 10, Bias: 151.819952, T: 196116, Avg. loss: 1759.465212
Total training time: 0.01 seconds.
-- Epoch 555
Norm: 354.16, NNZs: 10, Bias: 151.814817, T: 196470, Avg. loss: 1758.894201
Total training time: 0.01 seconds.
-- Epoch 556
Norm: 354.46, NNZs: 10, Bias: 151.804489, T: 196824, Avg. loss: 1758.325844
Total training time: 0.01 seconds.
-- Epoch 557
Norm: 354.76, NNZs: 10, Bias: 151.807749, T: 197178, Avg. loss: 1757.748394
Total training time: 0.01 seconds.
-- Epoch 558
Norm: 355.06, NNZs: 10, Bias: 151.803888, T: 197532, Avg. loss: 1757.176306
Total training time: 0.01 seconds.
-- Epoch 559
Norm: 355.36, NNZs: 10, Bias: 151.764063, T: 197886, Avg. loss: 1756.557191
Total training time: 0.01 seconds.
-- Epoch 560
Norm: 355.66, NNZs: 10, Bias: 151.768991, T: 198240, Avg. loss: 1756.026241
Total training time: 0.01 seconds.
-- Epoch 561
Norm: 355.96, NNZs: 10, Bias: 151.808025, T: 198594, Avg. loss: 1755.427854
Total training time: 0.01 seconds.
-- Epoch 562
Norm: 356.26, NNZs: 10, Bias: 151.780777, T: 198948, Avg. loss: 1754.873217
Total training time: 0.01 seconds.
-- Epoch 563
Norm: 356.55, NNZs: 10, Bias: 151.777968, T: 199302, Avg. loss: 1754.315300
Total training time: 0.01 seconds.
-- Epoch 564
Norm: 356.85, NNZs: 10, Bias: 151.781820, T: 199656, Avg. loss: 1753.749407
Total training time: 0.01 seconds.
-- Epoch 565
Norm: 357.15, NNZs: 10, Bias: 151.774515, T: 200010, Avg. loss: 1753.189855
Total training time: 0.01 seconds.
-- Epoch 566
Norm: 357.45, NNZs: 10, Bias: 151.790249, T: 200364, Avg. loss: 1752.614015
Total training time: 0.01 seconds.
-- Epoch 567
Norm: 357.74, NNZs: 10, Bias: 151.818631, T: 200718, Avg. loss: 1752.031014
Total training time: 0.01 seconds.
-- Epoch 568
Norm: 358.04, NNZs: 10, Bias: 151.797812, T: 201072, Avg. loss: 1751.496942
Total training time: 0.01 seconds.
-- Epoch 569
Norm: 358.33, NNZs: 10, Bias: 151.842931, T: 201426, Avg. loss: 1750.886366
Total training time: 0.01 seconds.
-- Epoch 570
Norm: 358.63, NNZs: 10, Bias: 151.832044, T: 201780, Avg. loss: 1750.384479
Total training time: 0.01 seconds.
-- Epoch 571
Norm: 358.92, NNZs: 10, Bias: 151.844800, T: 202134, Avg. loss: 1749.816155
Total training time: 0.01 seconds.
-- Epoch 572
Norm: 359.22, NNZs: 10, Bias: 151.853549, T: 202488, Avg. loss: 1749.258357
Total training time: 0.01 seconds.
-- Epoch 573
Norm: 359.51, NNZs: 10, Bias: 151.824815, T: 202842, Avg. loss: 1748.685535
Total training time: 0.01 seconds.
-- Epoch 574
Norm: 359.81, NNZs: 10, Bias: 151.829701, T: 203196, Avg. loss: 1748.155099
Total training time: 0.01 seconds.
-- Epoch 575
Norm: 360.10, NNZs: 10, Bias: 151.780975, T: 203550, Avg. loss: 1747.560068
Total training time: 0.01 seconds.
-- Epoch 576
Norm: 360.39, NNZs: 10, Bias: 151.802466, T: 203904, Avg. loss: 1747.047019
Total training time: 0.01 seconds.
-- Epoch 577
Norm: 360.68, NNZs: 10, Bias: 151.761142, T: 204258, Avg. loss: 1746.474570
Total training time: 0.01 seconds.
-- Epoch 578
Norm: 360.97, NNZs: 10, Bias: 151.764754, T: 204612, Avg. loss: 1745.938220
Total training time: 0.01 seconds.
-- Epoch 579
Norm: 361.27, NNZs: 10, Bias: 151.760254, T: 204966, Avg. loss: 1745.410677
Total training time: 0.01 seconds.
-- Epoch 580
Norm: 361.56, NNZs: 10, Bias: 151.773713, T: 205320, Avg. loss: 1744.861327
Total training time: 0.01 seconds.
-- Epoch 581
Norm: 361.85, NNZs: 10, Bias: 151.798139, T: 205674, Avg. loss: 1744.311432
Total training time: 0.01 seconds.
-- Epoch 582
Norm: 362.14, NNZs: 10, Bias: 151.779857, T: 206028, Avg. loss: 1743.766373
Total training time: 0.01 seconds.
-- Epoch 583
Norm: 362.43, NNZs: 10, Bias: 151.796174, T: 206382, Avg. loss: 1743.204749
Total training time: 0.01 seconds.
-- Epoch 584
Norm: 362.72, NNZs: 10, Bias: 151.800841, T: 206736, Avg. loss: 1742.695851
Total training time: 0.01 seconds.
-- Epoch 585
Norm: 363.00, NNZs: 10, Bias: 151.783110, T: 207090, Avg. loss: 1742.151785
Total training time: 0.01 seconds.
-- Epoch 586
Norm: 363.29, NNZs: 10, Bias: 151.750543, T: 207444, Avg. loss: 1741.596021
Total training time: 0.01 seconds.
-- Epoch 587
Norm: 363.58, NNZs: 10, Bias: 151.771722, T: 207798, Avg. loss: 1741.065502
Total training time: 0.01 seconds.
-- Epoch 588
Norm: 363.87, NNZs: 10, Bias: 151.747668, T: 208152, Avg. loss: 1740.530860
Total training time: 0.01 seconds.
-- Epoch 589
Norm: 364.15, NNZs: 10, Bias: 151.779521, T: 208506, Avg. loss: 1739.990470
Total training time: 0.01 seconds.
-- Epoch 590
Norm: 364.44, NNZs: 10, Bias: 151.802510, T: 208860, Avg. loss: 1739.467263
Total training time: 0.01 seconds.
-- Epoch 591
Norm: 364.73, NNZs: 10, Bias: 151.777209, T: 209214, Avg. loss: 1738.938711
Total training time: 0.01 seconds.
-- Epoch 592
Norm: 365.01, NNZs: 10, Bias: 151.775810, T: 209568, Avg. loss: 1738.415197
Total training time: 0.01 seconds.
-- Epoch 593
Norm: 365.30, NNZs: 10, Bias: 151.771414, T: 209922, Avg. loss: 1737.881877
Total training time: 0.01 seconds.
-- Epoch 594
Norm: 365.58, NNZs: 10, Bias: 151.797428, T: 210276, Avg. loss: 1737.343607
Total training time: 0.01 seconds.
-- Epoch 595
Norm: 365.87, NNZs: 10, Bias: 151.787129, T: 210630, Avg. loss: 1736.827545
Total training time: 0.01 seconds.
-- Epoch 596
Norm: 366.15, NNZs: 10, Bias: 151.817690, T: 210984, Avg. loss: 1736.276172
Total training time: 0.01 seconds.
-- Epoch 597
Norm: 366.43, NNZs: 10, Bias: 151.786069, T: 211338, Avg. loss: 1735.759475
Total training time: 0.01 seconds.
-- Epoch 598
Norm: 366.72, NNZs: 10, Bias: 151.800581, T: 211692, Avg. loss: 1735.252734
Total training time: 0.01 seconds.
-- Epoch 599
Norm: 367.00, NNZs: 10, Bias: 151.793037, T: 212046, Avg. loss: 1734.718275
Total training time: 0.01 seconds.
-- Epoch 600
Norm: 367.28, NNZs: 10, Bias: 151.807929, T: 212400, Avg. loss: 1734.206928
Total training time: 0.01 seconds.
-- Epoch 601
Norm: 367.57, NNZs: 10, Bias: 151.817656, T: 212754, Avg. loss: 1733.689013
Total training time: 0.01 seconds.
-- Epoch 602
Norm: 367.85, NNZs: 10, Bias: 151.811469, T: 213108, Avg. loss: 1733.172612
Total training time: 0.01 seconds.
-- Epoch 603
Norm: 368.13, NNZs: 10, Bias: 151.804590, T: 213462, Avg. loss: 1732.655155
Total training time: 0.01 seconds.
-- Epoch 604
Norm: 368.41, NNZs: 10, Bias: 151.775523, T: 213816, Avg. loss: 1732.118844
Total training time: 0.01 seconds.
-- Epoch 605
Norm: 368.69, NNZs: 10, Bias: 151.818435, T: 214170, Avg. loss: 1731.584032
Total training time: 0.01 seconds.
-- Epoch 606
Norm: 368.97, NNZs: 10, Bias: 151.781326, T: 214524, Avg. loss: 1731.077104
Total training time: 0.01 seconds.
-- Epoch 607
Norm: 369.25, NNZs: 10, Bias: 151.774435, T: 214878, Avg. loss: 1730.592201
Total training time: 0.01 seconds.
-- Epoch 608
Norm: 369.53, NNZs: 10, Bias: 151.754512, T: 215232, Avg. loss: 1730.070682
Total training time: 0.01 seconds.
-- Epoch 609
Norm: 369.81, NNZs: 10, Bias: 151.737920, T: 215586, Avg. loss: 1729.561992
Total training time: 0.01 seconds.
-- Epoch 610
Norm: 370.09, NNZs: 10, Bias: 151.755074, T: 215940, Avg. loss: 1729.061331
Total training time: 0.01 seconds.
-- Epoch 611
Norm: 370.36, NNZs: 10, Bias: 151.776518, T: 216294, Avg. loss: 1728.540350
Total training time: 0.01 seconds.
-- Epoch 612
Norm: 370.64, NNZs: 10, Bias: 151.785922, T: 216648, Avg. loss: 1728.046200
Total training time: 0.01 seconds.
-- Epoch 613
Norm: 370.92, NNZs: 10, Bias: 151.776434, T: 217002, Avg. loss: 1727.540022
Total training time: 0.01 seconds.
-- Epoch 614
Norm: 371.19, NNZs: 10, Bias: 151.783742, T: 217356, Avg. loss: 1727.033222
Total training time: 0.01 seconds.
-- Epoch 615
Norm: 371.47, NNZs: 10, Bias: 151.724812, T: 217710, Avg. loss: 1726.409421
Total training time: 0.01 seconds.
-- Epoch 616
Norm: 371.75, NNZs: 10, Bias: 151.743775, T: 218064, Avg. loss: 1726.025650
Total training time: 0.01 seconds.
-- Epoch 617
Norm: 372.02, NNZs: 10, Bias: 151.758215, T: 218418, Avg. loss: 1725.527005
Total training time: 0.01 seconds.
-- Epoch 618
Norm: 372.30, NNZs: 10, Bias: 151.722743, T: 218772, Avg. loss: 1724.990499
Total training time: 0.01 seconds.
-- Epoch 619
Norm: 372.57, NNZs: 10, Bias: 151.740830, T: 219126, Avg. loss: 1724.529163
Total training time: 0.01 seconds.
-- Epoch 620
Norm: 372.85, NNZs: 10, Bias: 151.749545, T: 219480, Avg. loss: 1724.029381
Total training time: 0.01 seconds.
-- Epoch 621
Norm: 373.12, NNZs: 10, Bias: 151.739886, T: 219834, Avg. loss: 1723.526651
Total training time: 0.01 seconds.
-- Epoch 622
Norm: 373.39, NNZs: 10, Bias: 151.739877, T: 220188, Avg. loss: 1723.040116
Total training time: 0.01 seconds.
-- Epoch 623
Norm: 373.67, NNZs: 10, Bias: 151.751224, T: 220542, Avg. loss: 1722.541655
Total training time: 0.01 seconds.
-- Epoch 624
Norm: 373.94, NNZs: 10, Bias: 151.781784, T: 220896, Avg. loss: 1722.027384
Total training time: 0.01 seconds.
-- Epoch 625
Norm: 374.21, NNZs: 10, Bias: 151.812420, T: 221250, Avg. loss: 1721.524715
Total training time: 0.01 seconds.
-- Epoch 626
Norm: 374.49, NNZs: 10, Bias: 151.797133, T: 221604, Avg. loss: 1721.061572
Total training time: 0.01 seconds.
-- Epoch 627
Norm: 374.76, NNZs: 10, Bias: 151.805375, T: 221958, Avg. loss: 1720.571833
Total training time: 0.01 seconds.
-- Epoch 628
Norm: 375.03, NNZs: 10, Bias: 151.804920, T: 222312, Avg. loss: 1720.085658
Total training time: 0.01 seconds.
-- Epoch 629
Norm: 375.30, NNZs: 10, Bias: 151.815234, T: 222666, Avg. loss: 1719.589621
Total training time: 0.01 seconds.
-- Epoch 630
Norm: 375.57, NNZs: 10, Bias: 151.810458, T: 223020, Avg. loss: 1719.108792
Total training time: 0.01 seconds.
-- Epoch 631
Norm: 375.84, NNZs: 10, Bias: 151.787938, T: 223374, Avg. loss: 1718.611428
Total training time: 0.01 seconds.
-- Epoch 632
Norm: 376.11, NNZs: 10, Bias: 151.788710, T: 223728, Avg. loss: 1718.127574
Total training time: 0.01 seconds.
-- Epoch 633
Norm: 376.38, NNZs: 10, Bias: 151.768703, T: 224082, Avg. loss: 1717.644152
Total training time: 0.01 seconds.
-- Epoch 634
Norm: 376.65, NNZs: 10, Bias: 151.755807, T: 224436, Avg. loss: 1717.161193
Total training time: 0.01 seconds.
-- Epoch 635
Norm: 376.92, NNZs: 10, Bias: 151.717542, T: 224790, Avg. loss: 1716.627795
Total training time: 0.01 seconds.
-- Epoch 636
Norm: 377.19, NNZs: 10, Bias: 151.743553, T: 225144, Avg. loss: 1716.191756
Total training time: 0.01 seconds.
-- Epoch 637
Norm: 377.46, NNZs: 10, Bias: 151.766578, T: 225498, Avg. loss: 1715.710778
Total training time: 0.01 seconds.
-- Epoch 638
Norm: 377.72, NNZs: 10, Bias: 151.765453, T: 225852, Avg. loss: 1715.240550
Total training time: 0.01 seconds.
-- Epoch 639
Norm: 377.99, NNZs: 10, Bias: 151.765573, T: 226206, Avg. loss: 1714.756910
Total training time: 0.01 seconds.
-- Epoch 640
Norm: 378.26, NNZs: 10, Bias: 151.792222, T: 226560, Avg. loss: 1714.271571
Total training time: 0.01 seconds.
-- Epoch 641
Norm: 378.52, NNZs: 10, Bias: 151.798808, T: 226914, Avg. loss: 1713.808698
Total training time: 0.01 seconds.
-- Epoch 642
Norm: 378.79, NNZs: 10, Bias: 151.806811, T: 227268, Avg. loss: 1713.335930
Total training time: 0.01 seconds.
-- Epoch 643
Norm: 379.06, NNZs: 10, Bias: 151.826668, T: 227622, Avg. loss: 1712.841073
Total training time: 0.01 seconds.
-- Epoch 644
Norm: 379.32, NNZs: 10, Bias: 151.813048, T: 227976, Avg. loss: 1712.380368
Total training time: 0.01 seconds.
-- Epoch 645
Norm: 379.59, NNZs: 10, Bias: 151.819982, T: 228330, Avg. loss: 1711.914492
Total training time: 0.01 seconds.
-- Epoch 646
Norm: 379.85, NNZs: 10, Bias: 151.817332, T: 228684, Avg. loss: 1711.448487
Total training time: 0.01 seconds.
-- Epoch 647
Norm: 380.12, NNZs: 10, Bias: 151.824261, T: 229038, Avg. loss: 1710.974583
Total training time: 0.01 seconds.
-- Epoch 648
Norm: 380.38, NNZs: 10, Bias: 151.811356, T: 229392, Avg. loss: 1710.509775
Total training time: 0.01 seconds.
-- Epoch 649
Norm: 380.64, NNZs: 10, Bias: 151.821747, T: 229746, Avg. loss: 1710.030585
Total training time: 0.01 seconds.
-- Epoch 650
Norm: 380.91, NNZs: 10, Bias: 151.823717, T: 230100, Avg. loss: 1709.570194
Total training time: 0.01 seconds.
-- Epoch 651
Norm: 381.17, NNZs: 10, Bias: 151.838206, T: 230454, Avg. loss: 1709.096785
Total training time: 0.01 seconds.
-- Epoch 652
Norm: 381.43, NNZs: 10, Bias: 151.839901, T: 230808, Avg. loss: 1708.631664
Total training time: 0.01 seconds.
-- Epoch 653
Norm: 381.70, NNZs: 10, Bias: 151.837195, T: 231162, Avg. loss: 1708.181846
Total training time: 0.01 seconds.
-- Epoch 654
Norm: 381.96, NNZs: 10, Bias: 151.812053, T: 231516, Avg. loss: 1707.710484
Total training time: 0.01 seconds.
-- Epoch 655
Norm: 382.22, NNZs: 10, Bias: 151.812807, T: 231870, Avg. loss: 1707.252710
Total training time: 0.01 seconds.
-- Epoch 656
Norm: 382.48, NNZs: 10, Bias: 151.818947, T: 232224, Avg. loss: 1706.777376
Total training time: 0.01 seconds.
-- Epoch 657
Norm: 382.74, NNZs: 10, Bias: 151.840905, T: 232578, Avg. loss: 1706.305533
Total training time: 0.01 seconds.
-- Epoch 658
Norm: 383.00, NNZs: 10, Bias: 151.838242, T: 232932, Avg. loss: 1705.872933
Total training time: 0.01 seconds.
-- Epoch 659
Norm: 383.26, NNZs: 10, Bias: 151.814256, T: 233286, Avg. loss: 1705.415723
Total training time: 0.01 seconds.
-- Epoch 660
Norm: 383.52, NNZs: 10, Bias: 151.810313, T: 233640, Avg. loss: 1704.954944
Total training time: 0.01 seconds.
-- Epoch 661
Norm: 383.78, NNZs: 10, Bias: 151.774686, T: 233994, Avg. loss: 1704.460540
Total training time: 0.01 seconds.
-- Epoch 662
Norm: 384.04, NNZs: 10, Bias: 151.763858, T: 234348, Avg. loss: 1704.038840
Total training time: 0.01 seconds.
-- Epoch 663
Norm: 384.30, NNZs: 10, Bias: 151.778486, T: 234702, Avg. loss: 1703.586767
Total training time: 0.01 seconds.
-- Epoch 664
Norm: 384.56, NNZs: 10, Bias: 151.774848, T: 235056, Avg. loss: 1703.141598
Total training time: 0.01 seconds.
-- Epoch 665
Norm: 384.81, NNZs: 10, Bias: 151.781632, T: 235410, Avg. loss: 1702.687468
Total training time: 0.01 seconds.
-- Epoch 666
Norm: 385.07, NNZs: 10, Bias: 151.804026, T: 235764, Avg. loss: 1702.224747
Total training time: 0.01 seconds.
-- Epoch 667
Norm: 385.33, NNZs: 10, Bias: 151.816399, T: 236118, Avg. loss: 1701.776713
Total training time: 0.01 seconds.
-- Epoch 668
Norm: 385.59, NNZs: 10, Bias: 151.770242, T: 236472, Avg. loss: 1701.290266
Total training time: 0.01 seconds.
-- Epoch 669
Norm: 385.84, NNZs: 10, Bias: 151.756172, T: 236826, Avg. loss: 1700.884147
Total training time: 0.01 seconds.
-- Epoch 670
Norm: 386.10, NNZs: 10, Bias: 151.770881, T: 237180, Avg. loss: 1700.433503
Total training time: 0.01 seconds.
-- Epoch 671
Norm: 386.36, NNZs: 10, Bias: 151.753423, T: 237534, Avg. loss: 1699.970004
Total training time: 0.01 seconds.
-- Epoch 672
Norm: 386.61, NNZs: 10, Bias: 151.784001, T: 237888, Avg. loss: 1699.517011
Total training time: 0.01 seconds.
-- Epoch 673
Norm: 386.87, NNZs: 10, Bias: 151.771917, T: 238242, Avg. loss: 1699.096063
Total training time: 0.01 seconds.
-- Epoch 674
Norm: 387.12, NNZs: 10, Bias: 151.755913, T: 238596, Avg. loss: 1698.642349
Total training time: 0.01 seconds.
-- Epoch 675
Norm: 387.38, NNZs: 10, Bias: 151.756592, T: 238950, Avg. loss: 1698.219692
Total training time: 0.01 seconds.
-- Epoch 676
Norm: 387.63, NNZs: 10, Bias: 151.750834, T: 239304, Avg. loss: 1697.773665
Total training time: 0.01 seconds.
-- Epoch 677
Norm: 387.88, NNZs: 10, Bias: 151.762079, T: 239658, Avg. loss: 1697.334791
Total training time: 0.01 seconds.
-- Epoch 678
Norm: 388.14, NNZs: 10, Bias: 151.755010, T: 240012, Avg. loss: 1696.886591
Total training time: 0.01 seconds.
-- Epoch 679
Norm: 388.39, NNZs: 10, Bias: 151.757666, T: 240366, Avg. loss: 1696.454653
Total training time: 0.01 seconds.
-- Epoch 680
Norm: 388.64, NNZs: 10, Bias: 151.748373, T: 240720, Avg. loss: 1696.014767
Total training time: 0.01 seconds.
-- Epoch 681
Norm: 388.90, NNZs: 10, Bias: 151.766206, T: 241074, Avg. loss: 1695.572080
Total training time: 0.01 seconds.
-- Epoch 682
Norm: 389.15, NNZs: 10, Bias: 151.773966, T: 241428, Avg. loss: 1695.140474
Total training time: 0.01 seconds.
-- Epoch 683
Norm: 389.40, NNZs: 10, Bias: 151.769926, T: 241782, Avg. loss: 1694.706005
Total training time: 0.01 seconds.
-- Epoch 684
Norm: 389.65, NNZs: 10, Bias: 151.736047, T: 242136, Avg. loss: 1694.243759
Total training time: 0.01 seconds.
-- Epoch 685
Norm: 389.90, NNZs: 10, Bias: 151.747538, T: 242490, Avg. loss: 1693.838549
Total training time: 0.01 seconds.
-- Epoch 686
Norm: 390.16, NNZs: 10, Bias: 151.735696, T: 242844, Avg. loss: 1693.398561
Total training time: 0.01 seconds.
-- Epoch 687
Norm: 390.41, NNZs: 10, Bias: 151.752147, T: 243198, Avg. loss: 1692.952448
Total training time: 0.01 seconds.
-- Epoch 688
Norm: 390.66, NNZs: 10, Bias: 151.758854, T: 243552, Avg. loss: 1692.540909
Total training time: 0.01 seconds.
-- Epoch 689
Norm: 390.91, NNZs: 10, Bias: 151.742186, T: 243906, Avg. loss: 1692.105630
Total training time: 0.01 seconds.
-- Epoch 690
Norm: 391.16, NNZs: 10, Bias: 151.763363, T: 244260, Avg. loss: 1691.671970
Total training time: 0.01 seconds.
-- Epoch 691
Norm: 391.41, NNZs: 10, Bias: 151.750378, T: 244614, Avg. loss: 1691.247506
Total training time: 0.01 seconds.
-- Epoch 692
Norm: 391.65, NNZs: 10, Bias: 151.747696, T: 244968, Avg. loss: 1690.823876
Total training time: 0.01 seconds.
-- Epoch 693
Norm: 391.90, NNZs: 10, Bias: 151.784894, T: 245322, Avg. loss: 1690.363270
Total training time: 0.01 seconds.
-- Epoch 694
Norm: 392.15, NNZs: 10, Bias: 151.782501, T: 245676, Avg. loss: 1689.972249
Total training time: 0.01 seconds.
-- Epoch 695
Norm: 392.40, NNZs: 10, Bias: 151.756153, T: 246030, Avg. loss: 1689.523298
Total training time: 0.01 seconds.
-- Epoch 696
Norm: 392.65, NNZs: 10, Bias: 151.781363, T: 246384, Avg. loss: 1689.102258
Total training time: 0.01 seconds.
-- Epoch 697
Norm: 392.89, NNZs: 10, Bias: 151.750240, T: 246738, Avg. loss: 1688.678174
Total training time: 0.01 seconds.
-- Epoch 698
Norm: 393.14, NNZs: 10, Bias: 151.732123, T: 247092, Avg. loss: 1688.268006
Total training time: 0.01 seconds.
-- Epoch 699
Norm: 393.39, NNZs: 10, Bias: 151.724029, T: 247446, Avg. loss: 1687.852988
Total training time: 0.01 seconds.
-- Epoch 700
Norm: 393.63, NNZs: 10, Bias: 151.749747, T: 247800, Avg. loss: 1687.427972
Total training time: 0.01 seconds.
-- Epoch 701
Norm: 393.88, NNZs: 10, Bias: 151.756470, T: 248154, Avg. loss: 1687.016617
Total training time: 0.01 seconds.
-- Epoch 702
Norm: 394.13, NNZs: 10, Bias: 151.750592, T: 248508, Avg. loss: 1686.587997
Total training time: 0.01 seconds.
-- Epoch 703
Norm: 394.37, NNZs: 10, Bias: 151.762047, T: 248862, Avg. loss: 1686.170417
Total training time: 0.01 seconds.
-- Epoch 704
Norm: 394.62, NNZs: 10, Bias: 151.781173, T: 249216, Avg. loss: 1685.754768
Total training time: 0.01 seconds.
-- Epoch 705
Norm: 394.86, NNZs: 10, Bias: 151.770406, T: 249570, Avg. loss: 1685.344393
Total training time: 0.01 seconds.
-- Epoch 706
Norm: 395.11, NNZs: 10, Bias: 151.775912, T: 249924, Avg. loss: 1684.927702
Total training time: 0.01 seconds.
-- Epoch 707
Norm: 395.35, NNZs: 10, Bias: 151.767083, T: 250278, Avg. loss: 1684.514179
Total training time: 0.01 seconds.
-- Epoch 708
Norm: 395.59, NNZs: 10, Bias: 151.786639, T: 250632, Avg. loss: 1684.094356
Total training time: 0.01 seconds.
-- Epoch 709
Norm: 395.84, NNZs: 10, Bias: 151.775283, T: 250986, Avg. loss: 1683.685469
Total training time: 0.01 seconds.
-- Epoch 710
Norm: 396.08, NNZs: 10, Bias: 151.762000, T: 251340, Avg. loss: 1683.267966
Total training time: 0.01 seconds.
-- Epoch 711
Norm: 396.32, NNZs: 10, Bias: 151.748015, T: 251694, Avg. loss: 1682.853711
Total training time: 0.01 seconds.
-- Epoch 712
Norm: 396.57, NNZs: 10, Bias: 151.731444, T: 252048, Avg. loss: 1682.438300
Total training time: 0.01 seconds.
-- Epoch 713
Norm: 396.81, NNZs: 10, Bias: 151.761365, T: 252402, Avg. loss: 1682.029320
Total training time: 0.01 seconds.
-- Epoch 714
Norm: 397.05, NNZs: 10, Bias: 151.751743, T: 252756, Avg. loss: 1681.634696
Total training time: 0.01 seconds.
-- Epoch 715
Norm: 397.29, NNZs: 10, Bias: 151.744453, T: 253110, Avg. loss: 1681.215496
Total training time: 0.01 seconds.
-- Epoch 716
Norm: 397.53, NNZs: 10, Bias: 151.772944, T: 253464, Avg. loss: 1680.800481
Total training time: 0.01 seconds.
-- Epoch 717
Norm: 397.78, NNZs: 10, Bias: 151.790060, T: 253818, Avg. loss: 1680.405442
Total training time: 0.01 seconds.
-- Epoch 718
Norm: 398.02, NNZs: 10, Bias: 151.784701, T: 254172, Avg. loss: 1680.000751
Total training time: 0.01 seconds.
-- Epoch 719
Norm: 398.26, NNZs: 10, Bias: 151.804583, T: 254526, Avg. loss: 1679.591850
Total training time: 0.01 seconds.
-- Epoch 720
Norm: 398.50, NNZs: 10, Bias: 151.833907, T: 254880, Avg. loss: 1679.167788
Total training time: 0.01 seconds.
-- Epoch 721
Norm: 398.74, NNZs: 10, Bias: 151.818494, T: 255234, Avg. loss: 1678.794957
Total training time: 0.01 seconds.
-- Epoch 722
Norm: 398.98, NNZs: 10, Bias: 151.815214, T: 255588, Avg. loss: 1678.391331
Total training time: 0.01 seconds.
-- Epoch 723
Norm: 399.22, NNZs: 10, Bias: 151.822894, T: 255942, Avg. loss: 1677.980235
Total training time: 0.01 seconds.
-- Epoch 724
Norm: 399.46, NNZs: 10, Bias: 151.823452, T: 256296, Avg. loss: 1677.577441
Total training time: 0.01 seconds.
-- Epoch 725
Norm: 399.69, NNZs: 10, Bias: 151.844868, T: 256650, Avg. loss: 1677.162498
Total training time: 0.01 seconds.
-- Epoch 726
Norm: 399.93, NNZs: 10, Bias: 151.882857, T: 257004, Avg. loss: 1676.712770
Total training time: 0.01 seconds.
-- Epoch 727
Norm: 400.17, NNZs: 10, Bias: 151.874830, T: 257358, Avg. loss: 1676.395270
Total training time: 0.01 seconds.
-- Epoch 728
Norm: 400.41, NNZs: 10, Bias: 151.867522, T: 257712, Avg. loss: 1675.996596
Total training time: 0.01 seconds.
-- Epoch 729
Norm: 400.65, NNZs: 10, Bias: 151.880273, T: 258066, Avg. loss: 1675.573556
Total training time: 0.01 seconds.
-- Epoch 730
Norm: 400.88, NNZs: 10, Bias: 151.818215, T: 258420, Avg. loss: 1675.152186
Total training time: 0.01 seconds.
-- Epoch 731
Norm: 401.12, NNZs: 10, Bias: 151.776884, T: 258774, Avg. loss: 1674.773677
Total training time: 0.01 seconds.
-- Epoch 732
Norm: 401.36, NNZs: 10, Bias: 151.756006, T: 259128, Avg. loss: 1674.387658
Total training time: 0.01 seconds.
-- Epoch 733
Norm: 401.59, NNZs: 10, Bias: 151.772019, T: 259482, Avg. loss: 1674.008272
Total training time: 0.01 seconds.
-- Epoch 734
Norm: 401.83, NNZs: 10, Bias: 151.788758, T: 259836, Avg. loss: 1673.614444
Total training time: 0.01 seconds.
-- Epoch 735
Norm: 402.06, NNZs: 10, Bias: 151.804141, T: 260190, Avg. loss: 1673.223347
Total training time: 0.01 seconds.
-- Epoch 736
Norm: 402.30, NNZs: 10, Bias: 151.802891, T: 260544, Avg. loss: 1672.839732
Total training time: 0.01 seconds.
-- Epoch 737
Norm: 402.53, NNZs: 10, Bias: 151.802492, T: 260898, Avg. loss: 1672.446917
Total training time: 0.01 seconds.
-- Epoch 738
Norm: 402.77, NNZs: 10, Bias: 151.783872, T: 261252, Avg. loss: 1672.047895
Total training time: 0.01 seconds.
-- Epoch 739
Norm: 403.00, NNZs: 10, Bias: 151.758739, T: 261606, Avg. loss: 1671.636232
Total training time: 0.01 seconds.
-- Epoch 740
Norm: 403.24, NNZs: 10, Bias: 151.770289, T: 261960, Avg. loss: 1671.277158
Total training time: 0.01 seconds.
-- Epoch 741
Norm: 403.47, NNZs: 10, Bias: 151.771910, T: 262314, Avg. loss: 1670.887287
Total training time: 0.01 seconds.
-- Epoch 742
Norm: 403.71, NNZs: 10, Bias: 151.777333, T: 262668, Avg. loss: 1670.504857
Total training time: 0.01 seconds.
-- Epoch 743
Norm: 403.94, NNZs: 10, Bias: 151.766587, T: 263022, Avg. loss: 1670.117558
Total training time: 0.01 seconds.
-- Epoch 744
Norm: 404.17, NNZs: 10, Bias: 151.743919, T: 263376, Avg. loss: 1669.721149
Total training time: 0.01 seconds.
-- Epoch 745
Norm: 404.40, NNZs: 10, Bias: 151.743888, T: 263730, Avg. loss: 1669.347962
Total training time: 0.01 seconds.
-- Epoch 746
Norm: 404.64, NNZs: 10, Bias: 151.763085, T: 264084, Avg. loss: 1668.957500
Total training time: 0.01 seconds.
-- Epoch 747
Norm: 404.87, NNZs: 10, Bias: 151.774241, T: 264438, Avg. loss: 1668.575195
Total training time: 0.01 seconds.
-- Epoch 748
Norm: 405.10, NNZs: 10, Bias: 151.788299, T: 264792, Avg. loss: 1668.193267
Total training time: 0.01 seconds.
-- Epoch 749
Norm: 405.33, NNZs: 10, Bias: 151.823394, T: 265146, Avg. loss: 1667.780518
Total training time: 0.01 seconds.
-- Epoch 750
Norm: 405.56, NNZs: 10, Bias: 151.791615, T: 265500, Avg. loss: 1667.417166
Total training time: 0.01 seconds.
-- Epoch 751
Norm: 405.79, NNZs: 10, Bias: 151.801681, T: 265854, Avg. loss: 1667.055254
Total training time: 0.01 seconds.
-- Epoch 752
Norm: 406.03, NNZs: 10, Bias: 151.790394, T: 266208, Avg. loss: 1666.677946
Total training time: 0.01 seconds.
-- Epoch 753
Norm: 406.26, NNZs: 10, Bias: 151.794932, T: 266562, Avg. loss: 1666.299017
Total training time: 0.01 seconds.
-- Epoch 754
Norm: 406.49, NNZs: 10, Bias: 151.765302, T: 266916, Avg. loss: 1665.880120
Total training time: 0.01 seconds.
-- Epoch 755
Norm: 406.72, NNZs: 10, Bias: 151.745556, T: 267270, Avg. loss: 1665.531932
Total training time: 0.01 seconds.
-- Epoch 756
Norm: 406.95, NNZs: 10, Bias: 151.739146, T: 267624, Avg. loss: 1665.162869
Total training time: 0.01 seconds.
-- Epoch 757
Norm: 407.17, NNZs: 10, Bias: 151.713282, T: 267978, Avg. loss: 1664.741358
Total training time: 0.01 seconds.
-- Epoch 758
Norm: 407.40, NNZs: 10, Bias: 151.746285, T: 268332, Avg. loss: 1664.399746
Total training time: 0.01 seconds.
-- Epoch 759
Norm: 407.63, NNZs: 10, Bias: 151.767276, T: 268686, Avg. loss: 1664.027832
Total training time: 0.01 seconds.
-- Epoch 760
Norm: 407.86, NNZs: 10, Bias: 151.752066, T: 269040, Avg. loss: 1663.660416
Total training time: 0.01 seconds.
-- Epoch 761
Norm: 408.09, NNZs: 10, Bias: 151.746907, T: 269394, Avg. loss: 1663.292206
Total training time: 0.01 seconds.
-- Epoch 762
Norm: 408.32, NNZs: 10, Bias: 151.753253, T: 269748, Avg. loss: 1662.921439
Total training time: 0.01 seconds.
-- Epoch 763
Norm: 408.54, NNZs: 10, Bias: 151.763539, T: 270102, Avg. loss: 1662.526913
Total training time: 0.01 seconds.
-- Epoch 764
Norm: 408.77, NNZs: 10, Bias: 151.787837, T: 270456, Avg. loss: 1662.161384
Total training time: 0.01 seconds.
-- Epoch 765
Norm: 409.00, NNZs: 10, Bias: 151.759508, T: 270810, Avg. loss: 1661.786256
Total training time: 0.01 seconds.
-- Epoch 766
Norm: 409.23, NNZs: 10, Bias: 151.780548, T: 271164, Avg. loss: 1661.429409
Total training time: 0.01 seconds.
-- Epoch 767
Norm: 409.45, NNZs: 10, Bias: 151.745921, T: 271518, Avg. loss: 1661.021870
Total training time: 0.01 seconds.
-- Epoch 768
Norm: 409.68, NNZs: 10, Bias: 151.721551, T: 271872, Avg. loss: 1660.678644
Total training time: 0.01 seconds.
-- Epoch 769
Norm: 409.90, NNZs: 10, Bias: 151.728614, T: 272226, Avg. loss: 1660.329321
Total training time: 0.01 seconds.
-- Epoch 770
Norm: 410.13, NNZs: 10, Bias: 151.720957, T: 272580, Avg. loss: 1659.951539
Total training time: 0.01 seconds.
-- Epoch 771
Norm: 410.36, NNZs: 10, Bias: 151.717769, T: 272934, Avg. loss: 1659.595726
Total training time: 0.01 seconds.
-- Epoch 772
Norm: 410.58, NNZs: 10, Bias: 151.725660, T: 273288, Avg. loss: 1659.237392
Total training time: 0.01 seconds.
-- Epoch 773
Norm: 410.81, NNZs: 10, Bias: 151.727590, T: 273642, Avg. loss: 1658.870918
Total training time: 0.01 seconds.
-- Epoch 774
Norm: 411.03, NNZs: 10, Bias: 151.717382, T: 273996, Avg. loss: 1658.500365
Total training time: 0.01 seconds.
-- Epoch 775
Norm: 411.25, NNZs: 10, Bias: 151.747920, T: 274350, Avg. loss: 1658.123994
Total training time: 0.01 seconds.
-- Epoch 776
Norm: 411.48, NNZs: 10, Bias: 151.750568, T: 274704, Avg. loss: 1657.781395
Total training time: 0.01 seconds.
-- Epoch 777
Norm: 411.70, NNZs: 10, Bias: 151.750302, T: 275058, Avg. loss: 1657.418727
Total training time: 0.01 seconds.
-- Epoch 778
Norm: 411.93, NNZs: 10, Bias: 151.727844, T: 275412, Avg. loss: 1657.041911
Total training time: 0.01 seconds.
-- Epoch 779
Norm: 412.15, NNZs: 10, Bias: 151.762481, T: 275766, Avg. loss: 1656.671447
Total training time: 0.01 seconds.
-- Epoch 780
Norm: 412.37, NNZs: 10, Bias: 151.764500, T: 276120, Avg. loss: 1656.328560
Total training time: 0.01 seconds.
-- Epoch 781
Norm: 412.59, NNZs: 10, Bias: 151.784850, T: 276474, Avg. loss: 1655.969541
Total training time: 0.01 seconds.
-- Epoch 782
Norm: 412.82, NNZs: 10, Bias: 151.798875, T: 276828, Avg. loss: 1655.610325
Total training time: 0.01 seconds.
-- Epoch 783
Norm: 413.04, NNZs: 10, Bias: 151.785079, T: 277182, Avg. loss: 1655.245707
Total training time: 0.01 seconds.
-- Epoch 784
Norm: 413.26, NNZs: 10, Bias: 151.805931, T: 277536, Avg. loss: 1654.890287
Total training time: 0.01 seconds.
-- Epoch 785
Norm: 413.48, NNZs: 10, Bias: 151.785028, T: 277890, Avg. loss: 1654.545320
Total training time: 0.01 seconds.
-- Epoch 786
Norm: 413.70, NNZs: 10, Bias: 151.774517, T: 278244, Avg. loss: 1654.190947
Total training time: 0.01 seconds.
-- Epoch 787
Norm: 413.93, NNZs: 10, Bias: 151.772874, T: 278598, Avg. loss: 1653.837106
Total training time: 0.01 seconds.
-- Epoch 788
Norm: 414.15, NNZs: 10, Bias: 151.783572, T: 278952, Avg. loss: 1653.472999
Total training time: 0.01 seconds.
-- Epoch 789
Norm: 414.37, NNZs: 10, Bias: 151.781726, T: 279306, Avg. loss: 1653.130078
Total training time: 0.01 seconds.
-- Epoch 790
Norm: 414.59, NNZs: 10, Bias: 151.798704, T: 279660, Avg. loss: 1652.767090
Total training time: 0.01 seconds.
-- Epoch 791
Norm: 414.81, NNZs: 10, Bias: 151.782433, T: 280014, Avg. loss: 1652.417294
Total training time: 0.01 seconds.
-- Epoch 792
Norm: 415.03, NNZs: 10, Bias: 151.797942, T: 280368, Avg. loss: 1652.066181
Total training time: 0.01 seconds.
-- Epoch 793
Norm: 415.25, NNZs: 10, Bias: 151.793940, T: 280722, Avg. loss: 1651.719398
Total training time: 0.01 seconds.
-- Epoch 794
Norm: 415.47, NNZs: 10, Bias: 151.776912, T: 281076, Avg. loss: 1651.363034
Total training time: 0.01 seconds.
-- Epoch 795
Norm: 415.68, NNZs: 10, Bias: 151.798647, T: 281430, Avg. loss: 1651.003630
Total training time: 0.01 seconds.
-- Epoch 796
Norm: 415.90, NNZs: 10, Bias: 151.823345, T: 281784, Avg. loss: 1650.652016
Total training time: 0.01 seconds.
-- Epoch 797
Norm: 416.12, NNZs: 10, Bias: 151.837407, T: 282138, Avg. loss: 1650.308878
Total training time: 0.01 seconds.
-- Epoch 798
Norm: 416.34, NNZs: 10, Bias: 151.816093, T: 282492, Avg. loss: 1649.973807
Total training time: 0.01 seconds.
-- Epoch 799
Norm: 416.56, NNZs: 10, Bias: 151.811996, T: 282846, Avg. loss: 1649.621308
Total training time: 0.01 seconds.
-- Epoch 800
Norm: 416.78, NNZs: 10, Bias: 151.813338, T: 283200, Avg. loss: 1649.273924
Total training time: 0.01 seconds.
-- Epoch 801
Norm: 416.99, NNZs: 10, Bias: 151.796596, T: 283554, Avg. loss: 1648.924889
Total training time: 0.01 seconds.
-- Epoch 802
Norm: 417.21, NNZs: 10, Bias: 151.813627, T: 283908, Avg. loss: 1648.567875
Total training time: 0.01 seconds.
-- Epoch 803
Norm: 417.43, NNZs: 10, Bias: 151.812925, T: 284262, Avg. loss: 1648.240449
Total training time: 0.01 seconds.
-- Epoch 804
Norm: 417.64, NNZs: 10, Bias: 151.804504, T: 284616, Avg. loss: 1647.897363
Total training time: 0.01 seconds.
-- Epoch 805
Norm: 417.86, NNZs: 10, Bias: 151.826054, T: 284970, Avg. loss: 1647.532628
Total training time: 0.01 seconds.
-- Epoch 806
Norm: 418.08, NNZs: 10, Bias: 151.829311, T: 285324, Avg. loss: 1647.205199
Total training time: 0.01 seconds.
-- Epoch 807
Norm: 418.29, NNZs: 10, Bias: 151.766726, T: 285678, Avg. loss: 1646.780092
Total training time: 0.01 seconds.
-- Epoch 808
Norm: 418.51, NNZs: 10, Bias: 151.754349, T: 286032, Avg. loss: 1646.524255
Total training time: 0.01 seconds.
-- Epoch 809
Norm: 418.72, NNZs: 10, Bias: 151.790998, T: 286386, Avg. loss: 1646.148533
Total training time: 0.01 seconds.
-- Epoch 810
Norm: 418.94, NNZs: 10, Bias: 151.744215, T: 286740, Avg. loss: 1645.791525
Total training time: 0.01 seconds.
-- Epoch 811
Norm: 419.15, NNZs: 10, Bias: 151.759651, T: 287094, Avg. loss: 1645.501709
Total training time: 0.01 seconds.
-- Epoch 812
Norm: 419.37, NNZs: 10, Bias: 151.750390, T: 287448, Avg. loss: 1645.161121
Total training time: 0.01 seconds.
-- Epoch 813
Norm: 419.58, NNZs: 10, Bias: 151.745123, T: 287802, Avg. loss: 1644.826933
Total training time: 0.01 seconds.
-- Epoch 814
Norm: 419.80, NNZs: 10, Bias: 151.742344, T: 288156, Avg. loss: 1644.481493
Total training time: 0.01 seconds.
-- Epoch 815
Norm: 420.01, NNZs: 10, Bias: 151.738726, T: 288510, Avg. loss: 1644.133458
Total training time: 0.01 seconds.
-- Epoch 816
Norm: 420.22, NNZs: 10, Bias: 151.746461, T: 288864, Avg. loss: 1643.817707
Total training time: 0.01 seconds.
-- Epoch 817
Norm: 420.44, NNZs: 10, Bias: 151.802442, T: 289218, Avg. loss: 1643.395518
Total training time: 0.01 seconds.
-- Epoch 818
Norm: 420.65, NNZs: 10, Bias: 151.801330, T: 289572, Avg. loss: 1643.140563
Total training time: 0.01 seconds.
-- Epoch 819
Norm: 420.86, NNZs: 10, Bias: 151.777641, T: 289926, Avg. loss: 1642.786710
Total training time: 0.01 seconds.
-- Epoch 820
Norm: 421.08, NNZs: 10, Bias: 151.779445, T: 290280, Avg. loss: 1642.473615
Total training time: 0.01 seconds.
-- Epoch 821
Norm: 421.29, NNZs: 10, Bias: 151.757706, T: 290634, Avg. loss: 1642.130323
Total training time: 0.01 seconds.
-- Epoch 822
Norm: 421.50, NNZs: 10, Bias: 151.782190, T: 290988, Avg. loss: 1641.790786
Total training time: 0.01 seconds.
-- Epoch 823
Norm: 421.71, NNZs: 10, Bias: 151.760265, T: 291342, Avg. loss: 1641.450678
Total training time: 0.01 seconds.
-- Epoch 824
Norm: 421.92, NNZs: 10, Bias: 151.795538, T: 291696, Avg. loss: 1641.103650
Total training time: 0.01 seconds.
-- Epoch 825
Norm: 422.14, NNZs: 10, Bias: 151.778341, T: 292050, Avg. loss: 1640.807002
Total training time: 0.01 seconds.
-- Epoch 826
Norm: 422.35, NNZs: 10, Bias: 151.760450, T: 292404, Avg. loss: 1640.469310
Total training time: 0.01 seconds.
-- Epoch 827
Norm: 422.56, NNZs: 10, Bias: 151.741832, T: 292758, Avg. loss: 1640.130184
Total training time: 0.01 seconds.
-- Epoch 828
Norm: 422.77, NNZs: 10, Bias: 151.764577, T: 293112, Avg. loss: 1639.800522
Total training time: 0.01 seconds.
-- Epoch 829
Norm: 422.98, NNZs: 10, Bias: 151.762378, T: 293466, Avg. loss: 1639.493456
Total training time: 0.01 seconds.
-- Epoch 830
Norm: 423.19, NNZs: 10, Bias: 151.769276, T: 293820, Avg. loss: 1639.159307
Total training time: 0.01 seconds.
-- Epoch 831
Norm: 423.40, NNZs: 10, Bias: 151.775238, T: 294174, Avg. loss: 1638.834256
Total training time: 0.01 seconds.
-- Epoch 832
Norm: 423.61, NNZs: 10, Bias: 151.775489, T: 294528, Avg. loss: 1638.505224
Total training time: 0.01 seconds.
-- Epoch 833
Norm: 423.82, NNZs: 10, Bias: 151.768460, T: 294882, Avg. loss: 1638.175347
Total training time: 0.01 seconds.
-- Epoch 834
Norm: 424.03, NNZs: 10, Bias: 151.760631, T: 295236, Avg. loss: 1637.852104
Total training time: 0.01 seconds.
-- Epoch 835
Norm: 424.24, NNZs: 10, Bias: 151.794782, T: 295590, Avg. loss: 1637.483559
Total training time: 0.01 seconds.
-- Epoch 836
Norm: 424.45, NNZs: 10, Bias: 151.753148, T: 295944, Avg. loss: 1637.159009
Total training time: 0.01 seconds.
-- Epoch 837
Norm: 424.65, NNZs: 10, Bias: 151.749777, T: 296298, Avg. loss: 1636.880101
Total training time: 0.01 seconds.
-- Epoch 838
Norm: 424.86, NNZs: 10, Bias: 151.773006, T: 296652, Avg. loss: 1636.543957
Total training time: 0.01 seconds.
-- Epoch 839
Norm: 425.07, NNZs: 10, Bias: 151.768590, T: 297006, Avg. loss: 1636.233315
Total training time: 0.01 seconds.
-- Epoch 840
Norm: 425.28, NNZs: 10, Bias: 151.765001, T: 297360, Avg. loss: 1635.909340
Total training time: 0.01 seconds.
-- Epoch 841
Norm: 425.49, NNZs: 10, Bias: 151.781746, T: 297714, Avg. loss: 1635.561387
Total training time: 0.01 seconds.
-- Epoch 842
Norm: 425.69, NNZs: 10, Bias: 151.773719, T: 298068, Avg. loss: 1635.263523
Total training time: 0.01 seconds.
-- Epoch 843
Norm: 425.90, NNZs: 10, Bias: 151.765750, T: 298422, Avg. loss: 1634.944408
Total training time: 0.01 seconds.
-- Epoch 844
Norm: 426.11, NNZs: 10, Bias: 151.762136, T: 298776, Avg. loss: 1634.614749
Total training time: 0.01 seconds.
-- Epoch 845
Norm: 426.31, NNZs: 10, Bias: 151.768040, T: 299130, Avg. loss: 1634.305821
Total training time: 0.01 seconds.
-- Epoch 846
Norm: 426.52, NNZs: 10, Bias: 151.746808, T: 299484, Avg. loss: 1633.973257
Total training time: 0.01 seconds.
-- Epoch 847
Norm: 426.73, NNZs: 10, Bias: 151.755826, T: 299838, Avg. loss: 1633.670418
Total training time: 0.01 seconds.
-- Epoch 848
Norm: 426.93, NNZs: 10, Bias: 151.749783, T: 300192, Avg. loss: 1633.350658
Total training time: 0.01 seconds.
-- Epoch 849
Norm: 427.14, NNZs: 10, Bias: 151.753633, T: 300546, Avg. loss: 1633.033690
Total training time: 0.01 seconds.
-- Epoch 850
Norm: 427.34, NNZs: 10, Bias: 151.776006, T: 300900, Avg. loss: 1632.705042
Total training time: 0.01 seconds.
-- Epoch 851
Norm: 427.55, NNZs: 10, Bias: 151.779096, T: 301254, Avg. loss: 1632.396402
Total training time: 0.01 seconds.
-- Epoch 852
Norm: 427.75, NNZs: 10, Bias: 151.769311, T: 301608, Avg. loss: 1632.074500
Total training time: 0.01 seconds.
-- Epoch 853
Norm: 427.96, NNZs: 10, Bias: 151.770928, T: 301962, Avg. loss: 1631.760610
Total training time: 0.02 seconds.
-- Epoch 854
Norm: 428.16, NNZs: 10, Bias: 151.747002, T: 302316, Avg. loss: 1631.436344
Total training time: 0.02 seconds.
-- Epoch 855
Norm: 428.37, NNZs: 10, Bias: 151.776199, T: 302670, Avg. loss: 1631.118336
Total training time: 0.02 seconds.
-- Epoch 856
Norm: 428.57, NNZs: 10, Bias: 151.756499, T: 303024, Avg. loss: 1630.810099
Total training time: 0.02 seconds.
-- Epoch 857
Norm: 428.77, NNZs: 10, Bias: 151.760582, T: 303378, Avg. loss: 1630.512002
Total training time: 0.02 seconds.
-- Epoch 858
Norm: 428.98, NNZs: 10, Bias: 151.788748, T: 303732, Avg. loss: 1630.178575
Total training time: 0.02 seconds.
-- Epoch 859
Norm: 429.18, NNZs: 10, Bias: 151.797128, T: 304086, Avg. loss: 1629.880060
Total training time: 0.02 seconds.
-- Epoch 860
Norm: 429.38, NNZs: 10, Bias: 151.804278, T: 304440, Avg. loss: 1629.572192
Total training time: 0.02 seconds.
-- Epoch 861
Norm: 429.59, NNZs: 10, Bias: 151.790057, T: 304794, Avg. loss: 1629.258610
Total training time: 0.02 seconds.
-- Epoch 862
Norm: 429.79, NNZs: 10, Bias: 151.778967, T: 305148, Avg. loss: 1628.953727
Total training time: 0.02 seconds.
-- Epoch 863
Norm: 429.99, NNZs: 10, Bias: 151.723487, T: 305502, Avg. loss: 1628.547136
Total training time: 0.02 seconds.
-- Epoch 864
Norm: 430.19, NNZs: 10, Bias: 151.750649, T: 305856, Avg. loss: 1628.321664
Total training time: 0.02 seconds.
-- Epoch 865
Norm: 430.40, NNZs: 10, Bias: 151.738437, T: 306210, Avg. loss: 1628.021229
Total training time: 0.02 seconds.
-- Epoch 866
Norm: 430.60, NNZs: 10, Bias: 151.733535, T: 306564, Avg. loss: 1627.716516
Total training time: 0.02 seconds.
-- Epoch 867
Norm: 430.80, NNZs: 10, Bias: 151.733073, T: 306918, Avg. loss: 1627.409881
Total training time: 0.02 seconds.
-- Epoch 868
Norm: 431.00, NNZs: 10, Bias: 151.748244, T: 307272, Avg. loss: 1627.102889
Total training time: 0.02 seconds.
-- Epoch 869
Norm: 431.20, NNZs: 10, Bias: 151.750703, T: 307626, Avg. loss: 1626.800557
Total training time: 0.02 seconds.
-- Epoch 870
Norm: 431.40, NNZs: 10, Bias: 151.750838, T: 307980, Avg. loss: 1626.481219
Total training time: 0.02 seconds.
-- Epoch 871
Norm: 431.60, NNZs: 10, Bias: 151.784824, T: 308334, Avg. loss: 1626.146415
Total training time: 0.02 seconds.
-- Epoch 872
Norm: 431.80, NNZs: 10, Bias: 151.804073, T: 308688, Avg. loss: 1625.869682
Total training time: 0.02 seconds.
-- Epoch 873
Norm: 432.00, NNZs: 10, Bias: 151.836649, T: 309042, Avg. loss: 1625.534053
Total training time: 0.02 seconds.
-- Epoch 874
Norm: 432.20, NNZs: 10, Bias: 151.806859, T: 309396, Avg. loss: 1625.265660
Total training time: 0.02 seconds.
-- Epoch 875
Norm: 432.40, NNZs: 10, Bias: 151.834766, T: 309750, Avg. loss: 1624.930103
Total training time: 0.02 seconds.
-- Epoch 876
Norm: 432.60, NNZs: 10, Bias: 151.792290, T: 310104, Avg. loss: 1624.631074
Total training time: 0.02 seconds.
-- Epoch 877
Norm: 432.80, NNZs: 10, Bias: 151.804757, T: 310458, Avg. loss: 1624.352893
Total training time: 0.02 seconds.
-- Epoch 878
Norm: 433.00, NNZs: 10, Bias: 151.790398, T: 310812, Avg. loss: 1624.060095
Total training time: 0.02 seconds.
-- Epoch 879
Norm: 433.20, NNZs: 10, Bias: 151.773204, T: 311166, Avg. loss: 1623.752668
Total training time: 0.02 seconds.
-- Epoch 880
Norm: 433.40, NNZs: 10, Bias: 151.792605, T: 311520, Avg. loss: 1623.447232
Total training time: 0.02 seconds.
-- Epoch 881
Norm: 433.60, NNZs: 10, Bias: 151.790454, T: 311874, Avg. loss: 1623.161377
Total training time: 0.02 seconds.
-- Epoch 882
Norm: 433.80, NNZs: 10, Bias: 151.800101, T: 312228, Avg. loss: 1622.854584
Total training time: 0.02 seconds.
-- Epoch 883
Norm: 433.99, NNZs: 10, Bias: 151.780818, T: 312582, Avg. loss: 1622.551964
Total training time: 0.02 seconds.
-- Epoch 884
Norm: 434.19, NNZs: 10, Bias: 151.788824, T: 312936, Avg. loss: 1622.259349
Total training time: 0.02 seconds.
-- Epoch 885
Norm: 434.39, NNZs: 10, Bias: 151.791667, T: 313290, Avg. loss: 1621.962102
Total training time: 0.02 seconds.
-- Epoch 886
Norm: 434.59, NNZs: 10, Bias: 151.789101, T: 313644, Avg. loss: 1621.664092
Total training time: 0.02 seconds.
-- Epoch 887
Norm: 434.78, NNZs: 10, Bias: 151.761040, T: 313998, Avg. loss: 1621.351696
Total training time: 0.02 seconds.
-- Epoch 888
Norm: 434.98, NNZs: 10, Bias: 151.797757, T: 314352, Avg. loss: 1621.032640
Total training time: 0.02 seconds.
-- Epoch 889
Norm: 435.18, NNZs: 10, Bias: 151.776031, T: 314706, Avg. loss: 1620.766419
Total training time: 0.02 seconds.
-- Epoch 890
Norm: 435.37, NNZs: 10, Bias: 151.772376, T: 315060, Avg. loss: 1620.479687
Total training time: 0.02 seconds.
-- Epoch 891
Norm: 435.57, NNZs: 10, Bias: 151.767988, T: 315414, Avg. loss: 1620.180158
Total training time: 0.02 seconds.
-- Epoch 892
Norm: 435.77, NNZs: 10, Bias: 151.754262, T: 315768, Avg. loss: 1619.880483
Total training time: 0.02 seconds.
-- Epoch 893
Norm: 435.96, NNZs: 10, Bias: 151.789741, T: 316122, Avg. loss: 1619.560850
Total training time: 0.02 seconds.
-- Epoch 894
Norm: 436.16, NNZs: 10, Bias: 151.793304, T: 316476, Avg. loss: 1619.296963
Total training time: 0.02 seconds.
-- Epoch 895
Norm: 436.35, NNZs: 10, Bias: 151.768324, T: 316830, Avg. loss: 1618.988477
Total training time: 0.02 seconds.
-- Epoch 896
Norm: 436.55, NNZs: 10, Bias: 151.761772, T: 317184, Avg. loss: 1618.713118
Total training time: 0.02 seconds.
-- Epoch 897
Norm: 436.74, NNZs: 10, Bias: 151.730953, T: 317538, Avg. loss: 1618.391572
Total training time: 0.02 seconds.
-- Epoch 898
Norm: 436.94, NNZs: 10, Bias: 151.743123, T: 317892, Avg. loss: 1618.125385
Total training time: 0.02 seconds.
-- Epoch 899
Norm: 437.13, NNZs: 10, Bias: 151.751791, T: 318246, Avg. loss: 1617.835389
Total training time: 0.02 seconds.
-- Epoch 900
Norm: 437.33, NNZs: 10, Bias: 151.775272, T: 318600, Avg. loss: 1617.532636
Total training time: 0.02 seconds.
-- Epoch 901
Norm: 437.52, NNZs: 10, Bias: 151.774676, T: 318954, Avg. loss: 1617.257778
Total training time: 0.02 seconds.
-- Epoch 902
Norm: 437.71, NNZs: 10, Bias: 151.767228, T: 319308, Avg. loss: 1616.963489
Total training time: 0.02 seconds.
-- Epoch 903
Norm: 437.91, NNZs: 10, Bias: 151.757855, T: 319662, Avg. loss: 1616.662282
Total training time: 0.02 seconds.
-- Epoch 904
Norm: 438.10, NNZs: 10, Bias: 151.764283, T: 320016, Avg. loss: 1616.384150
Total training time: 0.02 seconds.
-- Epoch 905
Norm: 438.29, NNZs: 10, Bias: 151.779806, T: 320370, Avg. loss: 1616.087552
Total training time: 0.02 seconds.
-- Epoch 906
Norm: 438.49, NNZs: 10, Bias: 151.760578, T: 320724, Avg. loss: 1615.793973
Total training time: 0.02 seconds.
-- Epoch 907
Norm: 438.68, NNZs: 10, Bias: 151.768580, T: 321078, Avg. loss: 1615.522570
Total training time: 0.02 seconds.
-- Epoch 908
Norm: 438.87, NNZs: 10, Bias: 151.780563, T: 321432, Avg. loss: 1615.229423
Total training time: 0.02 seconds.
-- Epoch 909
Norm: 439.06, NNZs: 10, Bias: 151.780398, T: 321786, Avg. loss: 1614.950148
Total training time: 0.02 seconds.
-- Epoch 910
Norm: 439.26, NNZs: 10, Bias: 151.773473, T: 322140, Avg. loss: 1614.663001
Total training time: 0.02 seconds.
-- Epoch 911
Norm: 439.45, NNZs: 10, Bias: 151.804215, T: 322494, Avg. loss: 1614.345346
Total training time: 0.02 seconds.
-- Epoch 912
Norm: 439.64, NNZs: 10, Bias: 151.775435, T: 322848, Avg. loss: 1614.060736
Total training time: 0.02 seconds.
-- Epoch 913
Norm: 439.83, NNZs: 10, Bias: 151.775234, T: 323202, Avg. loss: 1613.804876
Total training time: 0.02 seconds.
-- Epoch 914
Norm: 440.02, NNZs: 10, Bias: 151.785315, T: 323556, Avg. loss: 1613.519522
Total training time: 0.02 seconds.
-- Epoch 915
Norm: 440.21, NNZs: 10, Bias: 151.771999, T: 323910, Avg. loss: 1613.235538
Total training time: 0.02 seconds.
-- Epoch 916
Norm: 440.41, NNZs: 10, Bias: 151.780838, T: 324264, Avg. loss: 1612.948662
Total training time: 0.02 seconds.
-- Epoch 917
Norm: 440.60, NNZs: 10, Bias: 151.786348, T: 324618, Avg. loss: 1612.669956
Total training time: 0.02 seconds.
-- Epoch 918
Norm: 440.79, NNZs: 10, Bias: 151.786752, T: 324972, Avg. loss: 1612.389109
Total training time: 0.02 seconds.
-- Epoch 919
Norm: 440.98, NNZs: 10, Bias: 151.770794, T: 325326, Avg. loss: 1612.099722
Total training time: 0.02 seconds.
-- Epoch 920
Norm: 441.17, NNZs: 10, Bias: 151.785504, T: 325680, Avg. loss: 1611.819354
Total training time: 0.02 seconds.
-- Epoch 921
Norm: 441.36, NNZs: 10, Bias: 151.777350, T: 326034, Avg. loss: 1611.543387
Total training time: 0.02 seconds.
-- Epoch 922
Norm: 441.55, NNZs: 10, Bias: 151.747523, T: 326388, Avg. loss: 1611.240914
Total training time: 0.02 seconds.
-- Epoch 923
Norm: 441.74, NNZs: 10, Bias: 151.716345, T: 326742, Avg. loss: 1610.937910
Total training time: 0.02 seconds.
-- Epoch 924
Norm: 441.93, NNZs: 10, Bias: 151.727871, T: 327096, Avg. loss: 1610.706773
Total training time: 0.02 seconds.
-- Epoch 925
Norm: 442.11, NNZs: 10, Bias: 151.716066, T: 327450, Avg. loss: 1610.421907
Total training time: 0.02 seconds.
-- Epoch 926
Norm: 442.30, NNZs: 10, Bias: 151.743557, T: 327804, Avg. loss: 1610.140521
Total training time: 0.02 seconds.
-- Epoch 927
Norm: 442.49, NNZs: 10, Bias: 151.776603, T: 328158, Avg. loss: 1609.845694
Total training time: 0.02 seconds.
-- Epoch 928
Norm: 442.68, NNZs: 10, Bias: 151.759397, T: 328512, Avg. loss: 1609.585188
Total training time: 0.02 seconds.
-- Epoch 929
Norm: 442.87, NNZs: 10, Bias: 151.774337, T: 328866, Avg. loss: 1609.308998
Total training time: 0.02 seconds.
-- Epoch 930
Norm: 443.06, NNZs: 10, Bias: 151.774776, T: 329220, Avg. loss: 1609.041978
Total training time: 0.02 seconds.
-- Epoch 931
Norm: 443.24, NNZs: 10, Bias: 151.771525, T: 329574, Avg. loss: 1608.761600
Total training time: 0.02 seconds.
-- Epoch 932
Norm: 443.43, NNZs: 10, Bias: 151.758040, T: 329928, Avg. loss: 1608.471542
Total training time: 0.02 seconds.
-- Epoch 933
Norm: 443.62, NNZs: 10, Bias: 151.773917, T: 330282, Avg. loss: 1608.206508
Total training time: 0.02 seconds.
-- Epoch 934
Norm: 443.81, NNZs: 10, Bias: 151.764650, T: 330636, Avg. loss: 1607.932457
Total training time: 0.02 seconds.
-- Epoch 935
Norm: 443.99, NNZs: 10, Bias: 151.762132, T: 330990, Avg. loss: 1607.653576
Total training time: 0.02 seconds.
-- Epoch 936
Norm: 444.18, NNZs: 10, Bias: 151.759207, T: 331344, Avg. loss: 1607.391927
Total training time: 0.02 seconds.
-- Epoch 937
Norm: 444.37, NNZs: 10, Bias: 151.763339, T: 331698, Avg. loss: 1607.116193
Total training time: 0.02 seconds.
-- Epoch 938
Norm: 444.55, NNZs: 10, Bias: 151.756924, T: 332052, Avg. loss: 1606.835307
Total training time: 0.02 seconds.
-- Epoch 939
Norm: 444.74, NNZs: 10, Bias: 151.756853, T: 332406, Avg. loss: 1606.566257
Total training time: 0.02 seconds.
-- Epoch 940
Norm: 444.93, NNZs: 10, Bias: 151.743124, T: 332760, Avg. loss: 1606.267400
Total training time: 0.02 seconds.
-- Epoch 941
Norm: 445.11, NNZs: 10, Bias: 151.795261, T: 333114, Avg. loss: 1605.947107
Total training time: 0.02 seconds.
-- Epoch 942
Norm: 445.30, NNZs: 10, Bias: 151.771786, T: 333468, Avg. loss: 1605.731697
Total training time: 0.02 seconds.
-- Epoch 943
Norm: 445.48, NNZs: 10, Bias: 151.773235, T: 333822, Avg. loss: 1605.476413
Total training time: 0.02 seconds.
-- Epoch 944
Norm: 445.67, NNZs: 10, Bias: 151.800216, T: 334176, Avg. loss: 1605.170692
Total training time: 0.02 seconds.
-- Epoch 945
Norm: 445.85, NNZs: 10, Bias: 151.789327, T: 334530, Avg. loss: 1604.941152
Total training time: 0.02 seconds.
-- Epoch 946
Norm: 446.04, NNZs: 10, Bias: 151.792920, T: 334884, Avg. loss: 1604.670223
Total training time: 0.02 seconds.
-- Epoch 947
Norm: 446.22, NNZs: 10, Bias: 151.793280, T: 335238, Avg. loss: 1604.401232
Total training time: 0.02 seconds.
-- Epoch 948
Norm: 446.41, NNZs: 10, Bias: 151.799624, T: 335592, Avg. loss: 1604.131318
Total training time: 0.02 seconds.
-- Epoch 949
Norm: 446.59, NNZs: 10, Bias: 151.774922, T: 335946, Avg. loss: 1603.839503
Total training time: 0.02 seconds.
-- Epoch 950
Norm: 446.78, NNZs: 10, Bias: 151.782760, T: 336300, Avg. loss: 1603.596146
Total training time: 0.02 seconds.
-- Epoch 951
Norm: 446.96, NNZs: 10, Bias: 151.793946, T: 336654, Avg. loss: 1603.314522
Total training time: 0.02 seconds.
-- Epoch 952
Norm: 447.15, NNZs: 10, Bias: 151.803114, T: 337008, Avg. loss: 1603.056664
Total training time: 0.02 seconds.
-- Epoch 953
Norm: 447.33, NNZs: 10, Bias: 151.807621, T: 337362, Avg. loss: 1602.793313
Total training time: 0.02 seconds.
-- Epoch 954
Norm: 447.51, NNZs: 10, Bias: 151.825209, T: 337716, Avg. loss: 1602.515442
Total training time: 0.02 seconds.
-- Epoch 955
Norm: 447.70, NNZs: 10, Bias: 151.819726, T: 338070, Avg. loss: 1602.255275
Total training time: 0.02 seconds.
-- Epoch 956
Norm: 447.88, NNZs: 10, Bias: 151.827808, T: 338424, Avg. loss: 1601.992286
Total training time: 0.02 seconds.
-- Epoch 957
Norm: 448.06, NNZs: 10, Bias: 151.807303, T: 338778, Avg. loss: 1601.728913
Total training time: 0.02 seconds.
-- Epoch 958
Norm: 448.24, NNZs: 10, Bias: 151.787678, T: 339132, Avg. loss: 1601.461478
Total training time: 0.02 seconds.
-- Epoch 959
Norm: 448.43, NNZs: 10, Bias: 151.764960, T: 339486, Avg. loss: 1601.178509
Total training time: 0.02 seconds.
-- Epoch 960
Norm: 448.61, NNZs: 10, Bias: 151.751153, T: 339840, Avg. loss: 1600.935408
Total training time: 0.02 seconds.
-- Epoch 961
Norm: 448.79, NNZs: 10, Bias: 151.756832, T: 340194, Avg. loss: 1600.677730
Total training time: 0.02 seconds.
-- Epoch 962
Norm: 448.97, NNZs: 10, Bias: 151.765629, T: 340548, Avg. loss: 1600.412748
Total training time: 0.02 seconds.
-- Epoch 963
Norm: 449.15, NNZs: 10, Bias: 151.769923, T: 340902, Avg. loss: 1600.149939
Total training time: 0.02 seconds.
-- Epoch 964
Norm: 449.34, NNZs: 10, Bias: 151.766783, T: 341256, Avg. loss: 1599.888817
Total training time: 0.02 seconds.
-- Epoch 965
Norm: 449.52, NNZs: 10, Bias: 151.775891, T: 341610, Avg. loss: 1599.625520
Total training time: 0.02 seconds.
-- Epoch 966
Norm: 449.70, NNZs: 10, Bias: 151.755849, T: 341964, Avg. loss: 1599.345438
Total training time: 0.02 seconds.
-- Epoch 967
Norm: 449.88, NNZs: 10, Bias: 151.759400, T: 342318, Avg. loss: 1599.106189
Total training time: 0.02 seconds.
-- Epoch 968
Norm: 450.06, NNZs: 10, Bias: 151.762488, T: 342672, Avg. loss: 1598.843076
Total training time: 0.02 seconds.
-- Epoch 969
Norm: 450.24, NNZs: 10, Bias: 151.771248, T: 343026, Avg. loss: 1598.582275
Total training time: 0.02 seconds.
-- Epoch 970
Norm: 450.42, NNZs: 10, Bias: 151.768565, T: 343380, Avg. loss: 1598.324582
Total training time: 0.02 seconds.
-- Epoch 971
Norm: 450.60, NNZs: 10, Bias: 151.772711, T: 343734, Avg. loss: 1598.064172
Total training time: 0.02 seconds.
-- Epoch 972
Norm: 450.78, NNZs: 10, Bias: 151.743357, T: 344088, Avg. loss: 1597.786140
Total training time: 0.02 seconds.
-- Epoch 973
Norm: 450.96, NNZs: 10, Bias: 151.737972, T: 344442, Avg. loss: 1597.546754
Total training time: 0.02 seconds.
-- Epoch 974
Norm: 451.14, NNZs: 10, Bias: 151.774842, T: 344796, Avg. loss: 1597.257458
Total training time: 0.02 seconds.
-- Epoch 975
Norm: 451.32, NNZs: 10, Bias: 151.774413, T: 345150, Avg. loss: 1597.030959
Total training time: 0.02 seconds.
-- Epoch 976
Norm: 451.50, NNZs: 10, Bias: 151.790882, T: 345504, Avg. loss: 1596.766605
Total training time: 0.02 seconds.
-- Epoch 977
Norm: 451.68, NNZs: 10, Bias: 151.802927, T: 345858, Avg. loss: 1596.505575
Total training time: 0.02 seconds.
-- Epoch 978
Norm: 451.86, NNZs: 10, Bias: 151.779111, T: 346212, Avg. loss: 1596.251683
Total training time: 0.02 seconds.
-- Epoch 979
Norm: 452.04, NNZs: 10, Bias: 151.785138, T: 346566, Avg. loss: 1596.004895
Total training time: 0.02 seconds.
-- Epoch 980
Norm: 452.21, NNZs: 10, Bias: 151.788504, T: 346920, Avg. loss: 1595.749456
Total training time: 0.02 seconds.
-- Epoch 981
Norm: 452.39, NNZs: 10, Bias: 151.785855, T: 347274, Avg. loss: 1595.497644
Total training time: 0.02 seconds.
-- Epoch 982
Norm: 452.57, NNZs: 10, Bias: 151.800724, T: 347628, Avg. loss: 1595.231212
Total training time: 0.02 seconds.
-- Epoch 983
Norm: 452.75, NNZs: 10, Bias: 151.798246, T: 347982, Avg. loss: 1594.985625
Total training time: 0.02 seconds.
-- Epoch 984
Norm: 452.93, NNZs: 10, Bias: 151.797478, T: 348336, Avg. loss: 1594.731607
Total training time: 0.02 seconds.
-- Epoch 985
Norm: 453.10, NNZs: 10, Bias: 151.800365, T: 348690, Avg. loss: 1594.473042
Total training time: 0.02 seconds.
-- Epoch 986
Norm: 453.28, NNZs: 10, Bias: 151.786884, T: 349044, Avg. loss: 1594.217369
Total training time: 0.02 seconds.
-- Epoch 987
Norm: 453.46, NNZs: 10, Bias: 151.762368, T: 349398, Avg. loss: 1593.950474
Total training time: 0.02 seconds.
-- Epoch 988
Norm: 453.64, NNZs: 10, Bias: 151.765496, T: 349752, Avg. loss: 1593.719781
Total training time: 0.02 seconds.
-- Epoch 989
Norm: 453.81, NNZs: 10, Bias: 151.771007, T: 350106, Avg. loss: 1593.470282
Total training time: 0.02 seconds.
-- Epoch 990
Norm: 453.99, NNZs: 10, Bias: 151.769235, T: 350460, Avg. loss: 1593.217237
Total training time: 0.02 seconds.
-- Epoch 991
Norm: 454.17, NNZs: 10, Bias: 151.750945, T: 350814, Avg. loss: 1592.949117
Total training time: 0.02 seconds.
-- Epoch 992
Norm: 454.34, NNZs: 10, Bias: 151.762665, T: 351168, Avg. loss: 1592.712976
Total training time: 0.02 seconds.
-- Epoch 993
Norm: 454.52, NNZs: 10, Bias: 151.727238, T: 351522, Avg. loss: 1592.427194
Total training time: 0.02 seconds.
-- Epoch 994
Norm: 454.69, NNZs: 10, Bias: 151.729536, T: 351876, Avg. loss: 1592.214650
Total training time: 0.02 seconds.
-- Epoch 995
Norm: 454.87, NNZs: 10, Bias: 151.721239, T: 352230, Avg. loss: 1591.947619
Total training time: 0.02 seconds.
-- Epoch 996
Norm: 455.05, NNZs: 10, Bias: 151.741390, T: 352584, Avg. loss: 1591.709884
Total training time: 0.02 seconds.
-- Epoch 997
Norm: 455.22, NNZs: 10, Bias: 151.779267, T: 352938, Avg. loss: 1591.422542
Total training time: 0.02 seconds.
-- Epoch 998
Norm: 455.40, NNZs: 10, Bias: 151.764717, T: 353292, Avg. loss: 1591.211323
Total training time: 0.02 seconds.
-- Epoch 999
Norm: 455.57, NNZs: 10, Bias: 151.762485, T: 353646, Avg. loss: 1590.970197
Total training time: 0.02 seconds.
-- Epoch 1000
Norm: 455.75, NNZs: 10, Bias: 151.735620, T: 354000, Avg. loss: 1590.699242
Total training time: 0.02 seconds.
Model Score: 0.44