Tensorflow

Machine Learning

Artificial Intelligence: The effort to automate intellectual tasks normally performed by humans.

Machine Learning: Rather than giving the program the rules, an algorithm finds the rules for us.

Neural Network: A form of machine learning that uses a layered representation of data.

What is a Tensor?

• A generalization of vectors and matrices to potentially higher dimensions. Internally, Tensorflow represents tensors as n-dimensional arrays of base datatypes.
• Each tensor represents a partially defined computation that will eventually produce a value.

Creating Tensors

string = tf.Variable("this is a string", tf.string)
number = tf.Variable(324, tf.int16)
floating = tf.Variable(3.567, tf.float64)

What we created above is a tensor of rank 0, also known as a scalar.

Rank/Degree of Tensors

rank1_tensor = tf.Variable(["Test"], tf.string)
rank2_tensor = tf.Variable([["test", "ok"], ["test", "yes"]], tf.string)
tf.rank(rank2_tensor)
>>> <tf.Tensor: shape(), dtype=int32, numpy=2>

Shape of Tensors

rank2_tensor.shape
>>> TensorShape([2, 2])

Changing Shape

# tf.ones() creates a shape [1,2,3] tensor full of ones
tensor1 = tf.ones([1,2,3])
tensor2 = tf.reshape(tensor1, [2,3,1])
# -1 tells the tensor to calculate the size of the dimension in that place -> will reshape the tensor to [3,2]
tensor3 = tf.reshape(tensor2, [3, -1])
# The number of elements in the reshaped tensor MUST match the number in the original
print(tensor1)
>>> tf.Tensor(
[[[1. 1. 1.]
  [1. 1. 1.]]], shape=(1, 2, 3), dtype=float32)

Types of Tensors

• Variable -> mutable
• Constant -> immutable
• Placeholder
• SparseTensor

Tensorflow Core Learning Algorithm

Linear Regression

• Import

from __future__ import absolute_import, division, print_function, unicode_literals
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from IPython.display import clear_output
from six.moves import urllib
import tensorflow.compat.v2.feature_column as fc

• Data

# Load dataset
dftrain = pd.read_csv('https://storage.googleapis.com/tf-datasets/titanic/train.csv')
dfeval = pd.read_csv('https://storage.googleapis.com/tf-datasets/titanic/eval.csv')
y_train = dftrain.pop('survived')
y_eval = dfeval.pop('survived')

• Visualization

dftrain.head()

dftrain.age.hist(bins=20)

The majority of passengers are in their 20’s or 30's.

dftrain.sex.value_counts().plot(kind='barh')

The majority of passengers are male.

dftrain['class'].value_counts().plot(kind='barh')

The majority of passengers are in the “Third” class.

pd.concat([dftrain, y_train], axis = 1).groupby('sex').survived.mean().plot(kind='barh').set_xlabel('% survive')

Females have a much higher chance of survival.

Training vs. Testing Data

• Training: what we feed to the model so that it can develop and learn (much larger size than the testing data).
• Testing: what we use to evaluate the model and see how well it’s performing. We must use a separate set of data that the model has not been trained on to evaluate it.