What is TensorFlow and why do we use it

Robert John
GDE (GCP, ML)
@robert_thas
TensorFlow
What is TensorFlow and why do we use it?

TensorFlow is a numerical
computation library. It lets us
perform mathematical and
statistical operations on our
data.

Example
Predict the selling price of a house given the number of rooms

● Medv => y
● Rooms => x
● Constant => b
● Gradients => w
Formulate a Hypothesis
This hypothesis is called a model

import pandas as pd
import numpy as np
train_df = pd.read_csv('/content/gdrive/My
Drive/boston/train.csv', index_col='ID')
train_df.head()

train_df[['rm', 'medv']].head()

y = train_df['medv'].values
train_df['constant'] = 1
columns = ['constant', 'rm']
x = train_df[columns].values
w = np.zeros((x.shape[1], 1))
y_pred = np.dot(x, w)

error = y - y_pred
print(error.shape)
squared_error = np.power(error, 2)
root_mean_squared_error = sqrt(squared_error.sum()) /
y_pred.shape[0]
print(root_mean_squared_error)

costs = []
w_0_s = []
w_1_s = []
learning_rate = 1e-3
steps = 20

for a in range(steps):
w_0 = w[0][0]
w_1 = w[1][0]
# make prediction
y_pred = np.dot(x, w)
error = y - y_pred
error_squared = np.power(error, 2)

# cost function is Least Mean Squares
LMS = error_squared.sum() / (2 * y.shape[0])
costs.append(LMS)
w_0_s.append(w_0)
w_1_s.append(w_1)
# update
w_0 = w_0 + learning_rate/y.shape[0] * error.sum()
w_1 = w_1 + learning_rate/y.shape[0] * (error * x[1]).sum()
w[0][0] = w_0
w[1][0] = w_1

cost_df = pd.DataFrame({'cost': pd.Series(costs),
'w_0': pd.Series(w_0_s), 'w_1': pd.Series(w_1_s)})
cost_df['cost'].plot()

_w = [w_0, w_1]
_w = np.asarray(_w)
_x = train_df[['constant', 'rm']].values
y_pred = np.dot(_x, _w)
_p = pd.DataFrame(dict(actual=train_df['medv'].values,
predicted=y_pred.reshape(-1)))
_p.head()

class Model(object):
def __init__(self):
self.W = None
self.b = None
def __call__(self, x):
if self.W == None:
self.W = tf.Variable(tf.random.normal(shape=(1, x.shape[1])))
if self.b == None:
self.b = tf.Variable(tf.random.normal(shape=(x.shape[0], 1)))
return tf.matmul(x, self.W, transpose_b=True) + self.b

model = Model()
output = model(tf.constant([3.0, 3.1, 1.9, 2.0, 2.5, 2.9],
shape=(3,2)))
print(output)

@tf.function
def loss(y_pred, y):
return tf.reduce_mean(tf.square(y-y_pred))

def train(model, x, y, alpha):
x = tf.convert_to_tensor(x, np.float32)
y = tf.convert_to_tensor(y, np.float32)
with tf.GradientTape() as t:
t.watch(x)
current_loss = loss(model(x), y)
#print(current_loss)
dW, db = t.gradient(current_loss, [model.W, model.b])
#print(dW, db)
model.W.assign_sub(alpha * dW)
model.b.assign_sub(alpha * db)

train_df = df.sample(frac=0.8,random_state=0)
test_df = df.drop(train_df.index)
columns = ['nox', 'rm', 'chas', 'dis', 'ptratio', 'lstat', 'rad']
X_train = train_df[columns].values
X_test = test_df[columns].values
y_train = train_df[['medv']].values
y_test = test_df[['medv']].values

epochs = 10
model = Model()
for i in range(epochs):
train(model, X_train, y_train, alpha=0.1)
print(model.W)

● Vectors & Matrices
● Matrix Dot Products
● Differentiation
What was all that?
We got introduced to
● Learning Rates
● Gradient Descent
● Training Epochs

All of that was Linear Regression. How
about Neural Networks?

import tensorflow as tf
from tensorflow import keras
model = keras.Sequential([
keras.layers.Dense(50, input_shape=(7,), activation='relu'),
keras.layers.Dense(50, activation='relu'),
keras.layers.Dense(50, activation='relu'),
keras.layers.Dropout(0.5),
keras.layers.Dense(1)
])
print(model.summary())

adam = keras.optimizers.Adam(0.001)
model.compile(optimizer=adam, loss='mse')
model.fit(X_train, y_train, epochs=2000, validation_split=0.1)

● Optimized operations
● Hardware acceleration
● Multiple languages/platforms
Benefits of using TensorFlow
● Layers
● Optimizers
● Loss Functions
● Feature Engineering
● Hyperparameter Tuning
● Model Optimization
● Transfer Learning

● Keras Preprocessing Layers
● Recommendations
● New features in tf.data
● Experimental Numpy support
What’s New in TensorFlow

● API Types (Keras, Estimators, etc)
● Originally graph-based vs. “pythonic”
Differences Between TensorFlow and Others

Learn More
Neural Networks and Deep Learning
http://neuralnetworksanddeeplearning.com/
Machine Learning Crash Course
https://developers.google.com/machine-learning/crash-course
TensorFlow
https://www.tensorﬂow.org/

Robert John
GDE (GCP, ML)
@robert_thas
Thank You!

What is TensorFlow and why do we use it

More Related Content

What's hot (19)

Similar to What is TensorFlow and why do we use it (20)

More from Robert John (7)

Recently uploaded (20)

What is TensorFlow and why do we use it