In this project, we utilize Tensorflow's animation capabilities to visualize a Linear Regression model built for predicting House Prices.
As outlined in Jerry Kurata's wonderful course from Pluralsight, Training a Model with TensorFlow has two prime aspects:
- Concept
- Implementation
We achieve the Conceptualization phase by following these steps:
- Preparation of the Data
- Inference
- Loss Measurement
- Optimizer to Minimize Loss
The corresponding Implementation steps followed are:
- Generation house size and price data
- Defining inference as : Price = (sizeFactor * size) + priceOffset
- Using Mean Square Error for loss measurement
- using Gradient Descent Optimizer
- tensorflow, numpy, math, matplotlib.pyplot, matplotlib.animation, numpy.random, plt.plot, plt.xlabel, plt.ylabel, plt.show
- matplotlib.animation used for animation of the plot to watch fitting of the line IN ACTION!