 View on TensorFlow.org
|
 Run in Google Colab
|
 View source on GitHub
|
 Download notebook
|
Overview
This notebook demonstrates how to use Moving Average Optimizer along with the Model Average Checkpoint from tensorflow addons package.
Moving Averaging
The advantage of Moving Averaging is that they are less prone to rampant loss shifts or irregular data representation in the latest batch. It gives a smooothened and a more genral idea of the model training until some point.
Stochastic Averaging
Stochastic Weight Averaging converges to wider optima. By doing so, it resembles geometric ensembeling. SWA is a simple method to improve model performance when used as a wrapper around other optimizers and averaging results from different points of trajectory of the inner optimizer.
Model Average Checkpoint
callbacks.ModelCheckpointdoesn't give you the option to save moving average weights in the middle of training, which is why Model Average Optimizers required a custom callback. Using theupdate_weightsparameter,ModelAverageCheckpointallows you to:
- Assign the moving average weights to the model, and save them.
- Keep the old non-averaged weights, but the saved model uses the average weights.
Setup
pip install -U tensorflow-addons
import tensorflow as tf
import tensorflow_addons as tfa
import numpy as np
import os
Build Model
def create_model(opt):
model = tf.keras.models.Sequential([
tf.keras.layers.Flatten(),
tf.keras.layers.Dense(64, activation='relu'),
tf.keras.layers.Dense(64, activation='relu'),
tf.keras.layers.Dense(10, activation='softmax')
])
model.compile(optimizer=opt,
loss='sparse_categorical_crossentropy',
metrics=['accuracy'])
return model
Prepare Dataset
#Load Fashion MNIST dataset
train, test = tf.keras.datasets.fashion_mnist.load_data()
images, labels = train
images = images/255.0
labels = labels.astype(np.int32)
fmnist_train_ds = tf.data.Dataset.from_tensor_slices((images, labels))
fmnist_train_ds = fmnist_train_ds.shuffle(5000).batch(32)
test_images, test_labels = test
Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/train-labels-idx1-ubyte.gz 32768/29515 [=================================] - 0s 0us/step 40960/29515 [=========================================] - 0s 0us/step Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/train-images-idx3-ubyte.gz 26427392/26421880 [==============================] - 0s 0us/step 26435584/26421880 [==============================] - 0s 0us/step Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/t10k-labels-idx1-ubyte.gz 16384/5148 [===============================================================================================] - 0s 0us/step Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/t10k-images-idx3-ubyte.gz 4423680/4422102 [==============================] - 0s 0us/step 4431872/4422102 [==============================] - 0s 0us/step
We will be comparing three optimizers here:
- Unwrapped SGD
- SGD with Moving Average
- SGD with Stochastic Weight Averaging
And see how they perform with the same model.
#Optimizers
sgd = tf.keras.optimizers.SGD(0.01)
moving_avg_sgd = tfa.optimizers.MovingAverage(sgd)
stocastic_avg_sgd = tfa.optimizers.SWA(sgd)
Both MovingAverage and StocasticAverage optimers use ModelAverageCheckpoint.
#Callback
checkpoint_path = "./training/cp-{epoch:04d}.ckpt"
checkpoint_dir = os.path.dirname(checkpoint_path)
cp_callback = tf.keras.callbacks.ModelCheckpoint(filepath=checkpoint_dir,
save_weights_only=True,
verbose=1)
avg_callback = tfa.callbacks.AverageModelCheckpoint(filepath=checkpoint_dir,
update_weights=True)
Train Model
Vanilla SGD Optimizer
#Build Model
model = create_model(sgd)
#Train the network
model.fit(fmnist_train_ds, epochs=5, callbacks=[cp_callback])
Epoch 1/5 1875/1875 [==============================] - 4s 2ms/step - loss: 0.8031 - accuracy: 0.7282 Epoch 00001: saving model to ./training Epoch 2/5 1875/1875 [==============================] - 3s 2ms/step - loss: 0.5049 - accuracy: 0.8240 Epoch 00002: saving model to ./training Epoch 3/5 1875/1875 [==============================] - 3s 2ms/step - loss: 0.4591 - accuracy: 0.8375 Epoch 00003: saving model to ./training Epoch 4/5 1875/1875 [==============================] - 3s 2ms/step - loss: 0.4328 - accuracy: 0.8492 Epoch 00004: saving model to ./training Epoch 5/5 1875/1875 [==============================] - 3s 2ms/step - loss: 0.4128 - accuracy: 0.8561 Epoch 00005: saving model to ./training <keras.callbacks.History at 0x7fc9d0262250>
#Evalute results
model.load_weights(checkpoint_dir)
loss, accuracy = model.evaluate(test_images, test_labels, batch_size=32, verbose=2)
print("Loss :", loss)
print("Accuracy :", accuracy)
313/313 - 0s - loss: 95.4645 - accuracy: 0.7796 Loss : 95.46446990966797 Accuracy : 0.7796000242233276
Moving Average SGD
#Build Model
model = create_model(moving_avg_sgd)
#Train the network
model.fit(fmnist_train_ds, epochs=5, callbacks=[avg_callback])
Epoch 1/5 1875/1875 [==============================] - 4s 2ms/step - loss: 0.8064 - accuracy: 0.7303 2021-09-02 00:35:29.787996: W tensorflow/python/util/util.cc:348] Sets are not currently considered sequences, but this may change in the future, so consider avoiding using them. INFO:tensorflow:Assets written to: ./training/assets Epoch 2/5 1875/1875 [==============================] - 4s 2ms/step - loss: 0.5114 - accuracy: 0.8223 INFO:tensorflow:Assets written to: ./training/assets Epoch 3/5 1875/1875 [==============================] - 4s 2ms/step - loss: 0.4620 - accuracy: 0.8382 INFO:tensorflow:Assets written to: ./training/assets Epoch 4/5 1875/1875 [==============================] - 4s 2ms/step - loss: 0.4345 - accuracy: 0.8470 INFO:tensorflow:Assets written to: ./training/assets Epoch 5/5 1875/1875 [==============================] - 4s 2ms/step - loss: 0.4146 - accuracy: 0.8547 INFO:tensorflow:Assets written to: ./training/assets <keras.callbacks.History at 0x7fc8e16f30d0>
#Evalute results
model.load_weights(checkpoint_dir)
loss, accuracy = model.evaluate(test_images, test_labels, batch_size=32, verbose=2)
print("Loss :", loss)
print("Accuracy :", accuracy)
313/313 - 0s - loss: 95.4645 - accuracy: 0.7796 Loss : 95.46446990966797 Accuracy : 0.7796000242233276
Stocastic Weight Average SGD
#Build Model
model = create_model(stocastic_avg_sgd)
#Train the network
model.fit(fmnist_train_ds, epochs=5, callbacks=[avg_callback])
Epoch 1/5 1875/1875 [==============================] - 5s 2ms/step - loss: 0.7896 - accuracy: 0.7350 INFO:tensorflow:Assets written to: ./training/assets Epoch 2/5 1875/1875 [==============================] - 5s 2ms/step - loss: 0.5670 - accuracy: 0.8065 INFO:tensorflow:Assets written to: ./training/assets Epoch 3/5 1875/1875 [==============================] - 5s 2ms/step - loss: 0.5345 - accuracy: 0.8142 INFO:tensorflow:Assets written to: ./training/assets Epoch 4/5 1875/1875 [==============================] - 5s 2ms/step - loss: 0.5194 - accuracy: 0.8188 INFO:tensorflow:Assets written to: ./training/assets Epoch 5/5 1875/1875 [==============================] - 5s 2ms/step - loss: 0.5089 - accuracy: 0.8235 INFO:tensorflow:Assets written to: ./training/assets <keras.callbacks.History at 0x7fc8e0538790>
#Evalute results
model.load_weights(checkpoint_dir)
loss, accuracy = model.evaluate(test_images, test_labels, batch_size=32, verbose=2)
print("Loss :", loss)
print("Accuracy :", accuracy)
313/313 - 0s - loss: 95.4645 - accuracy: 0.7796 Loss : 95.46446990966797 Accuracy : 0.7796000242233276