Diving into the TensorFlow Code

by Sophia TurolFebruary 24, 2016
This blog post looks into the variables, placeholders, cost values, etc. of TensorFlow code and discusses the potential of Deep Q-Learner.

Below are the videos from the TensorFlow Seattle meetup—sponsored and organized by Altoros on January 21, 2016.

 

An overview of TensorFlow

In this session, Akshay Srivatsan of Bloomberg LP provides an overview of TensorFlow, speaking of its mechanics and taking a glimpse at TensorBoard. He also mentions the likely scenarios of TensorFlow evolution.

 

Dive into the TensorFlow code

In the second part of the session, Akshay demonstrates the actual TensorFlow code, diving into its variables, placeholders, cost values, etc. In addition, he talks about Deep Q-Learner that showcases the potential of what can be built with TensorFlow.

 

 

Fireside chat

Watch this video to find out:

  • How long does it take to become productive with TensorFlow vs. other frameworks?
  • What are the alternatives?
  • What are the uses case that TensorFlow can be applied to?
  • What change is needed to make tools like TensorFlow accessible to anyone (in terms of organizations, toolsets, education, etc.)?

 

 

Join our group to stay updated with information on the next meetups!

 

Further reading

 

About the speaker

Akshay Srivatsan has worked at Bloomberg LP and at Applied Communication Sciences. He has a strong interest in machine learning, specifically in natural language processing. Akshay is experienced in graphical model approaches, such as LDA, and deep learning techniques, such as word vector embeddings. Using TensorFlow, Akshay has implemented projects on deep Q-learning, and on deep canonical correlation analysis.