A simple example that shows how to implement your own simple linear regression using TensorFlow in Python. Below is an overiew about the existing files:

• Points.csv: Data file that will be used to train and test our model
• DataHelper.py: A helper script that:
  • Reads the data file
  • Separate data from labels
  • Split into training and test using sklearn.cross_validation
• MyLinearRegression.py: The main script that contains the actual model. The code has been inspired from this repo. Once the script executes, it creates a directory called 'tf_model' that contais:
  • checkpoint
  • model.ckpt.data-00000-of-00001
  • model.ckpt.index
  • model.ckpt.meta
  • model.pbtxt
  • frozen.pb
  • optimized.pb