Continue training of the graph after restoring the session from a local checkpoint (this can be useful if we have to interrupt out computational session)
Plot the decision boundary for this TensorFlow perceptron. Why do you think the TensorFlow implementation performs better than our NumPy implementation on the test set?
Hint 1: you can re-use the code that we used in the NumPy section
Hint 2: since the bias is a 2D array, you need to access the float value via modelparams['bias'][0]
In [19]:
# %load solutions/03_tensorflow-boundary.py
Theoretically, we could restart the Jupyter notebook now (we would just have to prepare the dataset again then, though)
We are going to restore the session from a meta graph (notice "tf.Session()")