#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Thu Jan 23 09:59:50 2020

@author: tanu
"""
#%%
# LIDO ML: tensorflow
#%%

from sklearn.datasets import load_boston
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import MinMaxScaler
import jrpytensorflow
import tensorflow as tf

import matplotlib.pyplot as plt

boston = load_boston()
#X, y = boston.data, boston.target


from sklearn.model_selection import train_test_split

from sklearn.datasets import load_digits

digits = load_digits()

X_train, X_test, y_train, y_test = train_test_split(
        X, y, test_size = 0.2
)
#%%
# P1
X,y = jrpytensorflow.datasets.load_circles() 

plt.figure()
plt.scatter(X[:,0], X[:,1], c = y, edgecolor = 'black')

#2)
from sklearn.model_selection import train_test_split

X_train, X_test, y_train, y_test = train_test_split(
        X, y, test_size = 0.2
)


preprocess = Pipeline(
        steps = [
              ('rescale', MinMaxScaler())
              ]
)

preprocess.fit(X_train)

X_train = preprocess.transform(X_train)
X_test = preprocess.transform(X_test)

import tensorflow as tf
logModel = tf.keras.models.Sequential([
        tf.keras.layers.Dense(1, activation = 'sigmoid')
        
])

logModel.compile(optimizer = 'sgd',
                 loss = 'binary_crossentropy')
history = logModel.fit(X, y, epochs = 100)

logModel.summary()


#4) predicted probability: for X
model(X) # tf object
model.predict(X) # array


model(X).numpy().ravel() > 0.5

sum(model(X).numpy().ravel() > 0.5)
#101

sum( ( model(X).numpy().ravel() > 0.5) == y)
#97

# 4*) predicted probability: for X_test

logModel.compile(optimizer = 'sgd',
                 loss = 'binary_crossentropy')
history = logModel.fit(X_test, y_test, epochs = 100)

logModel.summary()



model(X_test) # tf object
model.predict(X_test) # array


model(X_test).numpy().ravel() > 0.5

sum( model(X_test).numpy().ravel() > 0.5) 
#22

sum((model(X_test).numpy().ravel() > 0.5)  == y_test)
#21

#%%
# Practical 2
#%%

from tensorflow import keras

def smallModel():
    model = tf.keras.models.Sequential([
            tf.keras.layers.Dense(20, activation = 'relu'),
            tf.keras.layers.Dense(10, activation = 'relu'),
            tf.keras.layers.Dense(1, activation = 'softmax')
            ])
    model.compile(optimizer = 'sgd',
                  loss = 'binary_crossentropy',
                  metrics = ['accuracy'])

    return model

model = smallModel()
model.summary()