Merge pull request WZMIAOMIAO#30 from WZMIAOMIAO/dev

update read_ckpt.py

tensorflow_classification/Test3_vgg/fine_train_vgg16.py

@@ -0,0 +1,90 @@
+from tensorflow.keras.preprocessing.image import ImageDataGenerator
+import matplotlib.pyplot as plt
+from model import vgg
+import tensorflow as tf
+import json
+import os
+
+
+data_root = os.path.abspath(os.path.join(os.getcwd(), "../.."))  # get data root path
+image_path = data_root + "/data_set/flower_data/"  # flower data set path
+train_dir = image_path + "train"
+validation_dir = image_path + "val"
+
+# create direction for saving weights
+if not os.path.exists("save_weights"):
+    os.makedirs("save_weights")
+
+im_height = 224
+im_width = 224
+batch_size = 32
+epochs = 10
+
+_R_MEAN = 123.68
+_G_MEAN = 116.78
+_B_MEAN = 103.94
+
+
+def pre_function(img):
+    # img = im.open('test.jpg')
+    # img = np.array(img).astype(np.float32)
+    img = img - [_R_MEAN, _G_MEAN, _B_MEAN]
+
+    return img
+
+
+# data generator with data augmentation
+train_image_generator = ImageDataGenerator(horizontal_flip=True,
+                                           preprocessing_function=pre_function)
+validation_image_generator = ImageDataGenerator(preprocessing_function=pre_function)
+
+train_data_gen = train_image_generator.flow_from_directory(directory=train_dir,
+                                                           batch_size=batch_size,
+                                                           shuffle=True,
+                                                           target_size=(im_height, im_width),
+                                                           class_mode='categorical')
+total_train = train_data_gen.n
+
+# get class dict
+class_indices = train_data_gen.class_indices
+
+# transform value and key of dict
+inverse_dict = dict((val, key) for key, val in class_indices.items())
+# write dict into json file
+json_str = json.dumps(inverse_dict, indent=4)
+with open('class_indices.json', 'w') as json_file:
+    json_file.write(json_str)
+
+val_data_gen = train_image_generator.flow_from_directory(directory=validation_dir,
+                                                         batch_size=batch_size,
+                                                         shuffle=True,
+                                                         target_size=(im_height, im_width),
+                                                         class_mode='categorical')
+total_val = val_data_gen.n
+
+model = vgg("vgg16", 224, 224, 5)
+model.load_weights('./pretrain_weights.ckpt')
+for layer_t in model.layers:
+    if layer_t.name == 'feature':
+        layer_t.trainable = False
+        break
+
+model.summary()
+
+# using keras high level api for training
+model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=0.0001),
+              loss=tf.keras.losses.CategoricalCrossentropy(from_logits=False),
+              metrics=["accuracy"])
+
+callbacks = [tf.keras.callbacks.ModelCheckpoint(filepath='./save_weights/myAlex_{epoch}.h5',
+                                                save_best_only=True,
+                                                save_weights_only=True,
+                                                monitor='val_loss')]
+
+# tensorflow2.1 recommend to using fit
+history = model.fit(x=train_data_gen,
+                    steps_per_epoch=total_train // batch_size,
+                    epochs=epochs,
+                    validation_data=val_data_gen,
+                    validation_steps=total_val // batch_size,
+                    callbacks=callbacks)

tensorflow_classification/Test3_vgg/read_ckpt.py

@@ -0,0 +1,72 @@
+import tensorflow as tf
+
+
+def rename_var(ckpt_path, new_ckpt_path, num_classes=5):
+    with tf.Graph().as_default(), tf.compat.v1.Session().as_default() as sess:
+        var_list = tf.train.list_variables(ckpt_path)
+        new_var_list = []
+
+        for var_name, shape in var_list:
+            # print(var_name)
+            if var_name in except_list:
+                continue
+
+            var = tf.train.load_variable(ckpt_path, var_name)
+            new_var_name = var_name.replace('vgg_16', 'feature')
+            new_var_name = new_var_name.replace("weights", "kernel")
+            new_var_name = new_var_name.replace("biases", "bias")
+
+            new_var_name = new_var_name.replace("conv1/conv1_1", "conv2d")
+            new_var_name = new_var_name.replace("conv1/conv1_2", "conv2d_1")
+
+            new_var_name = new_var_name.replace("conv2/conv2_1", "conv2d_2")
+            new_var_name = new_var_name.replace("conv2/conv2_2", "conv2d_3")
+
+            new_var_name = new_var_name.replace("conv3/conv3_1", "conv2d_4")
+            new_var_name = new_var_name.replace("conv3/conv3_2", "conv2d_5")
+            new_var_name = new_var_name.replace("conv3/conv3_3", "conv2d_6")
+
+            new_var_name = new_var_name.replace("conv4/conv4_1", "conv2d_7")
+            new_var_name = new_var_name.replace("conv4/conv4_2", "conv2d_8")
+            new_var_name = new_var_name.replace("conv4/conv4_3", "conv2d_9")
+
+            new_var_name = new_var_name.replace("conv5/conv5_1", "conv2d_10")
+            new_var_name = new_var_name.replace("conv5/conv5_2", "conv2d_11")
+            new_var_name = new_var_name.replace("conv5/conv5_3", "conv2d_12")
+
+            if 'fc' in new_var_name:
+                # new_var_name = new_var_name.replace("feature/fc6", "dense")
+                # new_var_name = new_var_name.replace("feature/fc7", "dense_1")
+                # new_var_name = new_var_name.replace("fc8", "dense_2")
+                continue
+
+        #     print(new_var_name)
+            re_var = tf.Variable(var, name=new_var_name)
+            new_var_list.append(re_var)
+
+        re_var = tf.Variable(tf.keras.initializers.he_uniform()([25088, 2048]), name="dense/kernel")
+        new_var_list.append(re_var)
+        re_var = tf.Variable(tf.keras.initializers.he_uniform()([2048]), name="dense/bias")
+        new_var_list.append(re_var)
+
+        re_var = tf.Variable(tf.keras.initializers.he_uniform()([2048, 2048]), name="dense_1/kernel")
+        new_var_list.append(re_var)
+        re_var = tf.Variable(tf.keras.initializers.he_uniform()([2048]), name="dense_1/bias")
+        new_var_list.append(re_var)
+
+        re_var = tf.Variable(tf.keras.initializers.he_uniform()([2048, num_classes]), name="dense_2/kernel")
+        new_var_list.append(re_var)
+        re_var = tf.Variable(tf.keras.initializers.he_uniform()([num_classes]), name="dense_2/bias")
+        new_var_list.append(re_var)
+
+        saver = tf.compat.v1.train.Saver(new_var_list)
+        sess.run(tf.compat.v1.global_variables_initializer())
+        saver.save(sess, save_path=new_ckpt_path, write_meta_graph=False, write_state=False)
+
+
+except_list = ['global_step', 'vgg_16/mean_rgb', 'vgg_16/fc8/biases', 'vgg_16/fc8/weights']
+# http://download.tensorflow.org/models/vgg_16_2016_08_28.tar.gz
+ckpt_path = './vgg_16.ckpt'
+new_ckpt_path = './pretrain_weights.ckpt'
+num_classes = 5
+rename_var(ckpt_path, new_ckpt_path, num_classes)

tensorflow_classification/Test5_resnet/read_ckpt.py

@@ -1,9 +1,11 @@
 import tensorflow as tf
 
 
-def rename_var(ckpt_path, new_ckpt_path):
+def rename_var(ckpt_path, new_ckpt_path, num_classes=5):
     with tf.Graph().as_default(), tf.compat.v1.Session().as_default() as sess:
         var_list = tf.train.list_variables(ckpt_path)
+        new_var_list = []
+
         for var_name, shape in var_list:
             print(var_name)
             if var_name in except_list:
@@ -17,9 +19,9 @@ def rename_var(ckpt_path, new_ckpt_path):
             re_var = tf.Variable(var, name=new_var_name)
             new_var_list.append(re_var)
 
-        re_var = tf.Variable(tf.keras.initializers.he_uniform()([2048, 5]), name="logits/kernel")
+        re_var = tf.Variable(tf.keras.initializers.he_uniform()([2048, num_classes]), name="logits/kernel")
         new_var_list.append(re_var)
-        re_var = tf.Variable(tf.keras.initializers.he_uniform()([5]), name="logits/bias")
+        re_var = tf.Variable(tf.keras.initializers.he_uniform()([num_classes]), name="logits/bias")
         new_var_list.append(re_var)
         saver = tf.compat.v1.train.Saver(new_var_list)
         sess.run(tf.compat.v1.global_variables_initializer())
@@ -29,5 +31,5 @@ def rename_var(ckpt_path, new_ckpt_path):
 except_list = ['global_step', 'resnet_v1_50/mean_rgb', 'resnet_v1_50/logits/biases', 'resnet_v1_50/logits/weights']
 ckpt_path = './resnet_v1_50.ckpt'
 new_ckpt_path = './pretrain_weights.ckpt'
-new_var_list = []
-rename_var(ckpt_path, new_ckpt_path)
+num_classes = 5
+rename_var(ckpt_path, new_ckpt_path, num_classes)

tensorflow_classification/Test6_mobilenet/read_ckpt.py

@@ -1,9 +1,11 @@
 import tensorflow as tf
 
 
-def rename_var(ckpt_path, new_ckpt_path):
+def rename_var(ckpt_path, new_ckpt_path, num_classes=5):
     with tf.Graph().as_default(), tf.compat.v1.Session().as_default() as sess:
         var_list = tf.train.list_variables(ckpt_path)
+        new_var_list = []
+
         for var_name, shape in var_list:
             # print(var_name)
             if var_name in except_list:
@@ -30,9 +32,9 @@ def rename_var(ckpt_path, new_ckpt_path):
             re_var = tf.Variable(var, name=new_var_name)
             new_var_list.append(re_var)
 
-        re_var = tf.Variable(tf.keras.initializers.he_uniform()([1280, 5]), name="Logits/kernel")
+        re_var = tf.Variable(tf.keras.initializers.he_uniform()([1280, num_classes]), name="Logits/kernel")
         new_var_list.append(re_var)
-        re_var = tf.Variable(tf.keras.initializers.he_uniform()([5]), name="Logits/bias")
+        re_var = tf.Variable(tf.keras.initializers.he_uniform()([num_classes]), name="Logits/bias")
 
         new_var_list.append(re_var)
         tf.keras.initializers.he_uniform()
@@ -44,5 +46,5 @@ def rename_var(ckpt_path, new_ckpt_path):
 except_list = ['global_step', 'MobilenetV2/Logits/Conv2d_1c_1x1/biases', 'MobilenetV2/Logits/Conv2d_1c_1x1/weights']
 ckpt_path = './pretain_model/mobilenet_v2_1.0_224.ckpt'
 new_ckpt_path = './pretrain_weights.ckpt'
-new_var_list = []
-rename_var(ckpt_path, new_ckpt_path)
+num_classes = 5
+rename_var(ckpt_path, new_ckpt_path, num_classes)