Tuesday, February 14, 2017

Tensor Flow

Tensor Flow is the open source machine learning / deep learning  platform used by google. They have implemented this in a wide variety of their products. Basically it can do numerical computation using data flow graphs. The Nodes in the graph represent mathematical operations. Edges represent the multidimensional data arrays (tensors) communicated between them hence tensor flow. Operation in tensor flow is a named abstract computation which can take input attribute and produce output attribute.

The speciality of tensor flow is that it can be implemented on android, iOS and raspberry pi so that  the mobile tensor flow can be used to process data within the device even without a network connection.

You can install tensor flow by following this guide. Once you are done, try the below program.

import tensorflow as tf
a = tf.constant(5)
b = tf.constant(6)
sess = tf.Session()
print(sess.run(a*b))

Output
30

Get the GitHub repo from here

2 comments: