TensorFlow in 60 seconds

Say hello to machine learning with TensorFlow and thanks to Google for providing free resources (including Chrome browser, Colab notebooks, and powerful GPUs in the cloud). Here is a quick refresher in 60 seconds in just 3 steps and 3 clicks.

Python with TensorFlow in Google Colaboratory

Photo Credit: Machine Learning in Design Automation

This article will show you how to run a sample machine learning app using Python, Tensorflow and Google Colab as software infrastructure and Google Cloud as hardware backbone.

Click and Install Colaboratory in Chrome Browser

Click on the following url to install colab extension in chrome browser— https://chrome.google.com/webstore/detail/colaboratory/flckfnigdgnmmidlohfbfccgpakpeagd. This extension acts as IDE for Machine Learning app.

Photo Credit: Machine Learning in Design Automation

Open TensorFlow Notebook

Click on following url in chrome browser — https://colab.research.google.com/drive/1HmIpZcjstcNwAjVj6ZN-LxNCajfELDD6. This will open Colab IDE. Hit “OPEN IN PLAYGROUND” followed by “COPY TO DRIVE” to create a local copy in your google drive.

Photo Credit: Machine Learning in Design Automation

Let’s Run

Once you’ve a local copy, Press <CTRL-F9> or Click “Runtime -> Runall” to run simple machine learning app in your browser.

Photo Credit: Machine Learning in Design Automation

Results

Once you run, you’ll see Machine Learning model training, generation and analysis kick in once cell at a time. See the picture below:

Photo Credit: Machine Learning in Design Automation

Summary:

Once you’ve successfully run this sample app, you can delete the code or import a new jupyter notebook and start writing your latest machine learning app. Happy Coding !

References:

1. Machine Intelligence in Design Automation
2. Developing EDA applications using Machine Learning
3. Webinar on Machine Learning in EDA.