Added convolutional net results

Author: Gripnook

src/csv/convolutional.csv

@@ -0,0 +1,101 @@
+epoch,test_accuracy
+0,0.0934
+1,0.9623
+2,0.9759
+3,0.9807
+4,0.9834
+5,0.9867
+6,0.9861
+7,0.9881
+8,0.9887
+9,0.9899
+10,0.9891
+11,0.991
+12,0.9913
+13,0.9889
+14,0.992
+15,0.9925
+16,0.9909
+17,0.9919
+18,0.9918
+19,0.9922
+20,0.9923
+21,0.9924
+22,0.9923
+23,0.9918
+24,0.9917
+25,0.9904
+26,0.9921
+27,0.9928
+28,0.9911
+29,0.993
+30,0.9927
+31,0.992
+32,0.9916
+33,0.9926
+34,0.992
+35,0.9917
+36,0.9935
+37,0.9927
+38,0.9924
+39,0.9925
+40,0.9923
+41,0.9925
+42,0.9925
+43,0.9923
+44,0.992
+45,0.9924
+46,0.9923
+47,0.9928
+48,0.9935
+49,0.993
+50,0.9934
+51,0.9931
+52,0.9925
+53,0.9929
+54,0.9933
+55,0.9929
+56,0.9929
+57,0.9936
+58,0.9934
+59,0.9929
+60,0.9926
+61,0.9935
+62,0.9929
+63,0.9934
+64,0.9939
+65,0.9935
+66,0.993
+67,0.9931
+68,0.9928
+69,0.9925
+70,0.9929
+71,0.9937
+72,0.993
+73,0.9932
+74,0.994
+75,0.9933
+76,0.9928
+77,0.9924
+78,0.9928
+79,0.9919
+80,0.9931
+81,0.9932
+82,0.993
+83,0.9932
+84,0.9934
+85,0.9926
+86,0.9927
+87,0.9934
+88,0.9924
+89,0.9934
+90,0.9934
+91,0.9936
+92,0.9933
+93,0.9932
+94,0.9935
+95,0.9934
+96,0.9933
+97,0.9931
+98,0.993
+99,0.993

src/mnist_convolutional.py

@@ -1,3 +1,5 @@
+from __future__ import division
+
 import numpy as np
 import tensorflow as tf
 
@@ -60,18 +62,18 @@ def main():
     correct_prediction = tf.equal(tf.argmax(y_conv, 1), tf.argmax(y_, 1))
     accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
 
+    num_examples = mnist.train.images.shape[0]
+
     with tf.Session() as sess:
         sess.run(tf.global_variables_initializer())
-        for i in range(20000):
+        print('epoch,test_accuracy')
+        for i in range(100 * (num_examples // 100)):
+            if i % (num_examples // 100) == 0:
+                test_accuracy = accuracy.eval(feed_dict={x: mnist.test.images, y_: mnist.test.labels, keep_prob: 1.0})
+                print('%d,%g' % (i // (num_examples // 100), test_accuracy))
             batch = mnist.train.next_batch(100)
-            if i % 100 == 0:
-                train_accuracy = accuracy.eval(feed_dict={x: batch[0], y_: batch[1], keep_prob: 1.0})
-                print('step %d, training accuracy: %g' % (i, train_accuracy))
             train_step.run(feed_dict={x: batch[0], y_: batch[1], keep_prob: 0.5})
 
-        print('test accuracy %g' %
-              accuracy.eval(feed_dict={x: mnist.test.images, y_: mnist.test.labels, keep_prob: 1.0}))
-
 
 if __name__ == '__main__':
     main()