Hacktoberfest 2026: the issues maintainers tagged for October, open and beginner-friendly. Browse Hacktoberfest issues

QAT aware training for mobilenetV2 not working

Open
#1,086 1 comment 0 reactions 1 assignee View on GitHub

@Xhark is already working on this.

Since Aug 28, 2023.

Assessment

This issue has not been assessed yet.

Description

Hello I am trying to apply quantization aware training for mobilenet, and I am testing on the mnist dataset, the floating point models works very well but the moment I add the quantization I have a very huge loss and the model isn't converging.

definition of the model :

`
def build_model(target_size):
input_tensor = Input(shape=(target_size, target_size, 3))
base_model = applications.MobileNetV2(
include_top=False,
weights='imagenet',
input_tensor=input_tensor,
input_shape=(target_size, target_size, 3),
pooling='avg')

for layer in base_model.layers:
    layer.trainable = False  # trainable has to be false in order to freeze the layers

op = Dense(256, activation='relu')(base_model.output)
op = Dropout(.25)(op)

output_tensor = Dense(10, activation='softmax')(op)

model = Model(inputs=input_tensor, outputs=output_tensor)

return model`

quantization and training:

def quantized_model(model_fp32): quantize_model = tfmot.quantization.keras.quantize_model q_aware_model = quantize_model(model_fp32) return q_aware_model

`q_aware_model = quantized_model(model_fp32)
q_aware_model.compile(optimizer='adam',
loss='categorical_crossentropy',
metrics=['categorical_accuracy'])

print("training model with int8 precision")

train_images_subset = train_images[0:1000] # out of 60000
train_labels_subset = train_labels[0:1000]
encoded_y_quant_train = to_categorical(train_labels_subset, num_classes=10, dtype='float32')

train_quant_generator = load_data_generator(train_images_subset, encoded_y_quant_train, batch_size=64)

q_aware_model.fit(train_quant_generator,
                batch_size=500, epochs=1, steps_per_epoch=900)`
Dominant language
Python
Stars
1.6k
Forks
349
Avg merge
3d 2h
Merged PRs (30d)
1

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from tensorflow/model-optimization

All issues in tensorflow/model-optimization

Similar issues

More Python issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.