Basic AI Model that I’m going to use for developing a hyperparameter tuning template.

This is going to be the basic model that I’m going to work with for the next few days. I want to setup a template for automating hyperparameters.

import tensorflow as tf
print(tf.__version__)

2.0.0

from numpy.random import seed
seed(1)
tf.random.set_seed(1234)

mnist = tf.keras.datasets.fashion_mnist 

(training_images, training_labels), (test_images, test_labels) = mnist.load_data() 

training_images  = training_images / 255.0
test_images = test_images / 255.0

model = tf.keras.models.Sequential([tf.keras.layers.Flatten(),  
                                    tf.keras.layers.Dense(128, activation=tf.nn.relu),  
                                    tf.keras.layers.Dense(10, activation=tf.nn.softmax)]) 

model.compile(optimizer = 'adam', 
              loss = 'sparse_categorical_crossentropy', 
              metrics=['accuracy']) 

model.fit(training_images, training_labels, epochs=5)

Train on 60000 samples
Epoch 1/5
WARNING:tensorflow:Entity <function Function._initialize_uninitialized_variables.<locals>.initialize_variables at 0x7fdc1c516b70> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: module 'gast' has no attribute 'Num'
WARNING: Entity <function Function._initialize_uninitialized_variables.<locals>.initialize_variables at 0x7fdc1c516b70> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: module 'gast' has no attribute 'Num'
60000/60000 [==============================] - 3s 52us/sample - loss: 0.5063 - accuracy: 0.8210
Epoch 2/5
60000/60000 [==============================] - 3s 47us/sample - loss: 0.3784 - accuracy: 0.8648
Epoch 3/5
60000/60000 [==============================] - 3s 47us/sample - loss: 0.3404 - accuracy: 0.8765
Epoch 4/5
60000/60000 [==============================] - 3s 47us/sample - loss: 0.3149 - accuracy: 0.8848
Epoch 5/5
60000/60000 [==============================] - 3s 49us/sample - loss: 0.2949 - accuracy: 0.8916





<tensorflow.python.keras.callbacks.History at 0x7fdc1c4ca240>

model.evaluate(test_images, 
               test_labels,
               verbose=0) 

[0.3478685932993889, 0.8738]

Download