Initial commit

Tensorflow/tutoriel2/MNIST_convolution.py

@@ -0,0 +1,94 @@
+import tensorflow as tf
+import numpy as np
+import matplotlib.pyplot as plot
+import cv2
+
+def convolution(couche_prec, taille_noyau, nbr_noyau):
+    w=tf.Variable(tf.random.truncated_normal(shape=(taille_noyau, taille_noyau, int(couche_prec.get_shape()[-1]), nbr_noyau)))
+    b=np.zeros(nbr_noyau)
+    result=tf.nn.conv2d(couche_prec, w, strides=[1, 1, 1, 1], padding='SAME')+b
+    return result
+        
+def fc(couche_prec, nbr_neurone):
+    w=tf.Variable(tf.random.truncated_normal(shape=(int(couche_prec.get_shape()[-1]), nbr_neurone), dtype=tf.float32))
+    b=tf.Variable(np.zeros(shape=(nbr_neurone)), dtype=tf.float32)
+    result=tf.matmul(couche_prec, w)+b
+    return result
+
+taille_batch=100
+nbr_entrainement=3
+learning_rate=0.001
+
+mnist_train_images=np.fromfile("mnist/train-images-idx3-ubyte", dtype=np.uint8)[16:].reshape(-1, 28, 28, 1)/255
+mnist_train_labels=np.eye(10)[np.fromfile("mnist/train-labels-idx1-ubyte", dtype=np.uint8)[8:]]
+mnist_test_images=np.fromfile("mnist/t10k-images-idx3-ubyte", dtype=np.uint8)[16:].reshape(-1, 28, 28, 1)/255
+mnist_test_labels=np.eye(10)[np.fromfile("mnist/t10k-labels-idx1-ubyte", dtype=np.uint8)[8:]]
+
+ph_images=tf.placeholder(shape=(None, 28, 28, 1), dtype=tf.float32)
+ph_labels=tf.placeholder(shape=(None, 10), dtype=tf.float32)
+
+result=convolution(ph_images, 5, 32)
+result=convolution(result, 5, 32)
+result=tf.nn.max_pool(result, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
+
+result=convolution(result, 5, 128)
+result=convolution(result, 5, 128)
+result=tf.nn.max_pool(result, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
+
+result=tf.contrib.layers.flatten(result)
+
+result=fc(result, 512)
+result=tf.nn.sigmoid(result)
+result=fc(result, 10)
+scso=tf.nn.softmax(result)
+
+loss=tf.nn.softmax_cross_entropy_with_logits_v2(labels=ph_labels, logits=result)
+train=tf.train.AdamOptimizer(learning_rate).minimize(loss)
+accuracy=tf.reduce_mean(tf.cast(tf.equal(tf.argmax(scso, 1), tf.argmax(ph_labels, 1)), tf.float32))
+
+with tf.Session() as s:
+    s.run(tf.global_variables_initializer())
+    tab_train=[]
+    tab_test=[]
+    for id_entrainement in np.arange(nbr_entrainement):
+        tab_accuracy_train=[]
+        tab_accuracy_test=[]
+        for batch in np.arange(0, len(mnist_train_images), taille_batch):
+            s.run(train, feed_dict={
+                ph_images: mnist_train_images[batch:batch+taille_batch],
+                ph_labels: mnist_train_labels[batch:batch+taille_batch]
+            })
+        for batch in np.arange(0, len(mnist_train_images), taille_batch):
+            precision=s.run(accuracy, feed_dict={
+                ph_images: mnist_train_images[batch:batch+taille_batch],
+                ph_labels: mnist_train_labels[batch:batch+taille_batch]
+            })
+            tab_accuracy_train.append(precision)
+        for batch in np.arange(0, len(mnist_test_images), taille_batch):
+            precision=s.run(accuracy, feed_dict={
+                ph_images: mnist_test_images[batch:batch+taille_batch],
+                ph_labels: mnist_test_labels[batch:batch+taille_batch]
+            })
+            tab_accuracy_test.append(precision)
+        print("> Entrainement", id_entrainement)
+        print("  train:", np.mean(tab_accuracy_train))
+        tab_train.append(1-np.mean(tab_accuracy_train))
+        print("  test :", np.mean(tab_accuracy_test))
+        tab_test.append(1-np.mean(tab_accuracy_test))
+
+    plot.ylim(0, 1)
+    plot.grid()
+    plot.plot(tab_train, label="Train error")
+    plot.plot(tab_test, label="Test error")
+    plot.legend(loc="upper right")
+    plot.show()
+    
+    resulat=s.run(scso, feed_dict={ph_images: mnist_test_images[0:taille_batch]})
+    np.set_printoptions(formatter={'float': '{:0.3f}'.format})
+    for image in range(taille_batch):
+        print("image", image)
+        print("sortie du réseau:", resulat[image], np.argmax(resulat[image]))
+        print("sortie attendue :", mnist_test_labels[image], np.argmax(mnist_test_labels[image]))
+        cv2.imshow('image', mnist_test_images[image])
+        if cv2.waitKey()&0xFF==ord('q'):
+            break