TensorFlow Programming:

Write a TensorFlow program for the following Task.

a. Implement SoftMax Classification for Image Dataset that is not covered in class. Report accuracy,

b. Visualizations(Tensor Board): training and testing both.

Reference: https://www.tensorflow.org/programmers_guide/reading_data

In the Wiki Page, include a brief description of your dataset and your approach/results for this task.

The dataset is MINST dataset for handwriting images. The training has 60,000 examples and the testing has 10,000 examples. It is a subset of a larger set from NIST. The format is IDX FILE format. It is a very good format for storing vectors and multidimensional matrics of various numerial types. The data in the file actually looks like:

The program upload URL: https://github.com/nancyynx88/lab8/blob/master/mnist_train.py https://github.com/nancyynx88/lab8/blob/master/mnist_test.py

The testing result:

The accuracy as shown in the picture is 90.57%

The tensor board result:

Graph:

Distribution:

Histogram: