Course 2; Module 1: Optional Lab - Neurons and Layers

• Tensorflow is a machine learning package developed by Google.
  • In 2019, Google integrated Keras into Tensorflow and released Tensorflow 2.0.
  • Keras is a framework developed independently by François Chollet that creates a simple, layer-centric interface to Tensorflow.
• Runs thru how to do a linear regression in tensorflow, setting it up as one neuron, and assigning it a weight and a bias

• Generate a random weight and bias but feeding it some Xs from the training set (the x's were 1.0 and 2.0):

• Feed data into neuron by function linear_layer(); when fed 1.0 it results in $300 grand; just as regular np.dot method via linear regression:

When feed whole data set in, predicts as planned:

Logistic Regression in TensorFlow

• Set up a new layer, ensure it's activation for sigmoid operations

• Set weights for said layer:

• Predict output (sigmoid probability) for first training data - output matches up with non tf-methodology:

Can run and plot rest of data set thru this logistic layer:

Coffee Roasting in Tensorflow