I have run this example and I got the following error when I try to save the model.
import tensorflow as tf
import h5py
mnist = tf.keras.datasets.mnist
(x_train, y_train),(x_test, y_test) = mnist.load_data()
x_train, x_test = x_train / 255.0, x_test / 255.0
model = tf.keras.models.Sequential([
tf.keras.layers.Flatten(),
tf.keras.layers.Dense(512, activation=tf.nn.relu),
tf.keras.layers.Dropout(0.2),
tf.keras.layers.Dense(10, activation=tf.nn.softmax)
])
model.compile(optimizer='adam',
loss='sparse_categorical_crossentropy',
metrics=['accuracy'])
model.fit(x_train, y_train, epochs=2)
val_loss, val_acc = model.evaluate(x_test, y_test)
print(val_loss, val_acc)
model.save('model.h5')
new_model = tf.keras.models.load_model('model.h5')
I get this error:
Traceback (most recent call last):
File "/home/zneic/PycharmProjects/test/venv/test.py", line 23, in <module>
model.save('model.h5')
File "/home/zneic/.local/lib/python3.6/site-packages/tensorflow/python/keras/engine/network.py", line 1359, in save
'Currently `save` requires model to be a graph network. Consider '
NotImplementedError: Currently `save` requires model to be a graph network. Consider using `save_weights`, in order to save the weights of the model.
Your weights don't seem to be saved or loaded back into the session. Can you try saving the graph and the weights separately and loading them separately?
model_json = model.to_json()
with open("model.json", "w") as json_file:
json_file.write(model_json)
model.save_weights("model.h5")
Then you can load them:
def loadModel(jsonStr, weightStr):
json_file = open(jsonStr, 'r')
loaded_nnet = json_file.read()
json_file.close()
serve_model = tf.keras.models.model_from_json(loaded_nnet)
serve_model.load_weights(weightStr)
serve_model.compile(optimizer=tf.train.AdamOptimizer(),
loss='categorical_crossentropy',
metrics=['accuracy'])
return serve_model
model = loadModel('model.json', 'model.h5')
I have the same problem, and I solved it. I don't konw why, but it works. You can modify as this:
model = tf.keras.Sequential([
layers.Flatten(input_shape=(28, 28)),
layers.Dense(512, activation=tf.nn.relu, input_shape=(784,)),
layers.Dropout(0.2),
layers.Dense(10, activation=tf.nn.softmax)
])
Related
The code bellow build two tensorboard graphs for the same model, while using Keras API build nice simple graph, using tf.summary.trace_export() add for each variable define in the graph a node in the external scope with the suffix "readvariableop_resource", which make the graph be really messy as the number of the parameters increase.
(In the example below we have 2 dense layer each one have 2 variable (kernel and bias) total 4 variables (4 nodes))
from datetime import datetime
import tensorflow as tf
from tensorflow import keras
# Define the model.
model = keras.models.Sequential([
keras.layers.Flatten(input_shape=(28, 28)),
keras.layers.Dense(32, activation='relu'),
keras.layers.Dropout(0.2),
keras.layers.Dense(10, activation='softmax')
])
model.compile(
optimizer='adam',
loss='sparse_categorical_crossentropy',
metrics=['accuracy'])
(train_images, train_labels), _ = keras.datasets.mnist.load_data()
train_images = train_images / 255.0
# Define the Keras TensorBoard callback.
logdir="logs/fit/" + datetime.now().strftime("%Y%m%d-%H%M%S")
tensorboard_callback = keras.callbacks.TensorBoard(log_dir=logdir)
# Train the model.
model.fit(
train_images,
train_labels,
batch_size=64,
epochs=1,
callbacks=[tensorboard_callback])
#tf.function
def traceme(x):
return model(x)
logdir="logs/fit1/" + datetime.now().strftime("%Y%m%d-%H%M%S")
writer = tf.summary.create_file_writer(logdir)
tf.summary.trace_on(graph=True)
# Forward pass
traceme(tf.zeros((1, 28, 28, 1)))
with writer.as_default():
tf.summary.trace_export(name="model_trace", step=0)
While trying to implement a standard MNIST digit recognizer that many tutorials use to introduce you to neural networks, I'm encountering the error
ValueError: Shape mismatch: The shape of labels (received (1,)) should equal the shape of logits except for the last dimension (received (28, 10)).
I would like to use from_tensor_slices to process the data, since I want to apply the code to another problem where the data comes from a CSV file. Anyway, here is the code producing the error in the line model.fit(...)
import tensorflow as tf
train_dataset, test_dataset = tf.keras.datasets.mnist.load_data()
train_images, train_labels = train_dataset
train_images = train_images/255.0
train_dataset_tensor = tf.data.Dataset.from_tensor_slices((train_images, train_labels))
num_of_validation_data = 10000
validation_data = train_dataset_tensor.take(num_of_validation_data)
train_data = train_dataset_tensor.skip(num_of_validation_data)
model = tf.keras.Sequential([
tf.keras.layers.Flatten(),
tf.keras.layers.Dense(100, activation='sigmoid'),
tf.keras.layers.Dense(10, activation='softmax')
])
model.compile(
optimizer=tf.keras.optimizers.Adam(learning_rate=1e-3),
loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
metrics=['accuracy']
)
model.fit(train_data, batch_size=50, epochs=5)
performance = model.evaluate(validation_data)
I don't understand where the shape (28, 10) of the logits comes from, I thought I was flattening the image, essentially making a 1D vector out of the 2D image? How can I prevent the error?
You can use the following code
mnist = tf.keras.datasets.mnist
(x_train, y_train), (x_test, y_test) = mnist.load_data()
x_train, x_test = x_train / 255.0, x_test / 255.0
x_train = x_train[..., tf.newaxis]
x_test = x_test[..., tf.newaxis]
train_ds = tf.data.Dataset.from_tensor_slices(
(x_train, y_train)).shuffle(10000).batch(32)
test_ds = tf.data.Dataset.from_tensor_slices((x_test, y_test)).batch(32)
model = tf.keras.models.Sequential([
tf.keras.layers.Flatten(),
tf.keras.layers.Dense(100, activation='sigmoid'),
tf.keras.layers.Dense(10, activation='softmax')
])
model.compile(
optimizer=tf.keras.optimizers.Adam(learning_rate=1e-3),
loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
metrics=['accuracy']
)
model.fit(train_ds)
When running the code below I get an error, saying there is a shape mismatch, after searching on google I found this is commonly caused by using cross_entropy as a loss function when the target is not one-hot encoded. Since mine isn't one-hot encoded I used sparse_categorical_crossentropy instead, but the error is still there.
Here is the error I recieved :
Train on 60000 samples
Epoch 1/100
32/60000 [..............................] - ETA: 1:18
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-123-4ce1d4d047d4> in <module>()
----> 1 model.fit(x_train, y_train, batch_size=32, epochs=100, verbose=1)
25 frames
/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/ops/nn_ops.py in sparse_softmax_cross_entropy_with_logits(_sentinel, labels, logits, name)
3391 "should equal the shape of logits except for the last "
3392 "dimension (received %s)." % (labels_static_shape,
-> 3393 logits.get_shape()))
3394 # Check if no reshapes are required.
3395 if logits.get_shape().ndims == 2:
ValueError: Shape mismatch: The shape of labels (received (32,)) should equal the shape of logits except
for the last dimension (received (5408, 10)).
This is my code:
import numpy as np
import tensorflow as tf
from tensorflow.keras.datasets import fashion_mnist
(x_train, y_train), (x_test, y_test) = fashion_mnist.load_data()
x_train = x_train / 255.0
x_test = x_test / 255.0
x_train = np.reshape(x_train, (-1,28,28,1))
model = tf.keras.models.Sequential()
model.add(tf.keras.layers.Conv2D(filters=64, kernel_size=3, strides=2, input_shape=[28,28,1], activation='relu'))
model.add(tf.keras.layers.Dense(units = 10, activation='softmax'))
model.compile(optimizer='adam',
loss='sparse_categorical_crossentropy',
metrics=['sparse_categorical_accuracy'])
model.fit(x_train, y_train, batch_size=32, epochs=100, verbose=1)
I only just realized I had forgotten to add 'flatten' layer to the model:
model.add(tf.keras.layers.Flatten())
which has resolved the issue.
model = tf.keras.models.Sequential()
model.add(tf.keras.layers.Conv2D(filters=64, kernel_size=3, strides=2, input_shape= [28,28,1], activation='relu'))
model.add(tf.keras.layers.Flatten())
model.add(tf.keras.layers.Dense(units = 10, activation='softmax'))
I am having some trouble understanding how to implement a fully connected layer. Can someone share an example with real values?
Ex. Flattening the input, multiplying by the weights, adding the bias, etc.
TensorFlow does most of the low-level stuff for you.
Have a look at one of the getting started tutorials where there are complete examples in just a few lines of code:
import tensorflow as tf
mnist = tf.keras.datasets.mnist
(x_train, y_train),(x_test, y_test) = mnist.load_data()
x_train, x_test = x_train / 255.0, x_test / 255.0
model = tf.keras.models.Sequential([
tf.keras.layers.Flatten(input_shape=(28, 28)),
tf.keras.layers.Dense(512, activation=tf.nn.relu),
tf.keras.layers.Dropout(0.2),
tf.keras.layers.Dense(10, activation=tf.nn.softmax)
])
model.compile(optimizer='adam',
loss='sparse_categorical_crossentropy',
metrics=['accuracy'])
model.fit(x_train, y_train, epochs=5)
model.evaluate(x_test, y_test)
link: https://www.tensorflow.org/tutorials
I am trying to freeze the weights of certain layer in a prediction model with Keras and mnist dataset, but it does not work. The code is like:
from keras.layers import Dense, Flatten
from keras.utils import to_categorical
from keras.models import Sequential, load_model
from keras.datasets import mnist
from keras.losses import categorical_crossentropy
import numpy as np
def load_data():
(x_train, y_train), (x_test, y_test) = mnist.load_data()
x_train = x_train.astype('float32')
x_test = x_test.astype('float32')
x_train /= 255
x_test /= 255
y_train = to_categorical(y_train, num_classes=10)
y_test = to_categorical(y_test, num_classes=10)
return x_train, y_train, x_test, y_test
def run():
x_train, y_train, x_test, y_test = load_data()
model = Sequential([Flatten(input_shape=(28, 28)),
Dense(300, name='dense1', activation='relu'),
Dense(100, name='dense2', activation='relu'),
Dense(10, name='dense3', activation='softmax')])
model.trainable = True
model.compile(optimizer='Adam',
metrics=['accuracy'],
loss=categorical_crossentropy)
print(model.summary())
model.fit(x_train, y_train, epochs=5, verbose=2)
print(model.evaluate(x_test, y_test))
return model
def freeze(model):
x_train, y_train, x_test, y_test = load_data()
name = 'dense1'
weightsAndBias = model.get_layer(name=name).get_weights()
# freeze the weights of this layer
model.get_layer(name=name).trainable = False
# record the weights before retrain
weights_before = weightsAndBias[0]
# retrain
model.fit(x_train, y_train, verbose=2, epochs=1)
weights_after = model.get_layer(name=name).get_weights()[0]
if (weights_before == weights_after).all():
print('the weights did not change!!!')
else:
print('the weights changed!!!!')
if __name__ == '__main__':
model = run()
freeze(model)
The program outputs 'the weights changed!!!!'.
I do not understand why the weights of the layer named 'dense1' changes after setting model.get_layer(name=name).trainable = False.
You can do it by using:
model=Sequential()
layer=Dense(64,init='glorot_uniform',input_shape=(784,))
layer.trainable=False
model.add(layer)
layer2=Dense(784, activation='sigmoid',init='glorot_uniform')
layer2.trainable=True
model.add(layer2)
model.compile(loss='relu', optimizer=sgd,metrics = ['mae'])
You need to compile the graph after setting 'trainable'.
more info here
let me keep my layers freezed upto 5th layer, rest i will keep trainable
Here is more simple & more efficient code
for layer in model.layers[:5]:
layer.trainable=False
for layer in model.layers[5:]:
layer.trainable=True