#!/usr/bin/env python
# coding: utf-8

# # Eager Mode

# In[1]:


import tensorflow as tf
import numpy as np


# Everything is eager by default in tensorflow 2.0

# In[2]:


a = np.array([1., 2.])
b = np.array([2., 5.])

tf.add(a, b)


# ## Graph Model : AutoGraph
# As tensorflow suggest (https://www.youtube.com/watch?v=jh4ITuOytE4), we can use Autograph to automatically convert  eager function to graph function by using the tf.function decorator
# 
# The following code first compiles inside the function into graph code before to execute it. Note that if you remove the @tf.function decorator the code will work perfectly but will not be executed in graph mode.

# In[3]:


@tf.function
def add_fc(a, b):
    return tf.add(a, b)

print (add_fc(a, b))


# The following code show the code generated to create the graph

# In[4]:


def add_fc(a, b):
    return tf.add(a, b)

print(tf.autograph.to_code(add_fc))


# # Graph mode and eager mode  with a keras model

# In[5]:


# Flatten
model = tf.keras.models.Sequential()
# Add layers
model.add(tf.keras.layers.Dense(256, activation="relu"))
model.add(tf.keras.layers.Dense(128, activation="relu"))
model.add(tf.keras.layers.Dense(10, activation="softmax"))


# By calling model.predict you get a numpy array but the output is not differentiable

# In[6]:


model_output = model.predict(np.zeros((1, 30)))
model_output


# By calling model(), eager execution is used and the output is differentiable

# In[7]:


model_output = model(np.zeros((1, 30)))
model_output


# Of course, you can execute the model in graph mode by using the @tf.function decorator

# In[8]:


@tf.function
def predict(x):
    return model(x)

model_output = predict(np.zeros((1, 30)))
print(model_output)


# In[ ]:





# In[ ]: