Write a TensorFlow program for the following Task.

a.

Implement linear regression for dataset that is not covered

in class (e.g. Boston Dataset -

https://archive.ics.uci.edu/ml/datasets/Housing

).

b.

Plot training cost using Matplotlib in python.

c.

Implement cross-validation (Optional)

In the Wiki Page, include a brief description of your datase

t and your approach/results for this task.

Here's the link to the source code:

https://github.com/nancyynx88/lab7/blob/master/LinearRegression.py

Here's the program result:

In this python based tensorflow program, I learned how to implement the linear regression in it: This is to generate the training data function and to create some noice:

The training dataset is randomly generated:

Linear Regression

trX = np.linspace(-1, 1, 101)

create a y value which is approximately linear but with some random noise

Result for this task:

Training cost= 0.000622037 W= 2.00299 b= 3.99981

Testing... (Mean square loss Comparison) Testing cost= 2.29777e-07 Absolute mean square loss difference: 0.000621807

trY = 2 * trX + 4+np.random.randn(*trX.shape) * 0.033