Training settings
Please provide a valid training processor option
Neural network architecture
import tensorflow as tf
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, InputLayer, Dropout, Conv1D, Conv2D, Flatten, Reshape, MaxPooling1D, MaxPooling2D, AveragePooling2D, BatchNormalization, Permute, ReLU, Softmax, Activation
from tensorflow.keras.optimizers.legacy import Adam
from ei_tensorflow.velo import train_keras_model_with_velo
EPOCHS = args.epochs or 50
# If True, non-deterministic functions (e.g. shuffling batches) are not used.
# This is False by default.
ENSURE_DETERMINISM = args.ensure_determinism
# this controls the batch size, or you can manipulate the tf.data.Dataset objects yourself
BATCH_SIZE = args.batch_size or 32
if not ENSURE_DETERMINISM:
train_dataset = train_dataset.shuffle(buffer_size=BATCH_SIZE*4)
train_dataset=train_dataset.batch(BATCH_SIZE, drop_remainder=False)
validation_dataset = validation_dataset.batch(BATCH_SIZE, drop_remainder=False)
# model architecture
model = Sequential()
model.add(Conv2D(16, kernel_size=3, kernel_constraint=tf.keras.constraints.MaxNorm(1), padding='same', activation='relu'))
model.add(MaxPooling2D(pool_size=2, strides=2, padding='same'))
model.add(Conv2D(32, kernel_size=3, kernel_constraint=tf.keras.constraints.MaxNorm(1), padding='same', activation='relu'))
model.add(MaxPooling2D(pool_size=2, strides=2, padding='same'))
model.add(Flatten())
model.add(Dropout(0.25))
model.add(Dense(classes, name='y_pred', activation='softmax'))
callbacks.append(BatchLoggerCallback(BATCH_SIZE, train_sample_count, epochs=EPOCHS, ensure_determinism=ENSURE_DETERMINISM))
# train the neural network
train_keras_model_with_velo(
keras_model=model,
training_data=train_dataset,
validation_data=validation_dataset,
loss_fn=tf.keras.metrics.categorical_crossentropy,
num_epochs=EPOCHS,
callbacks=callbacks
)
# Use this flag to disable per-channel quantization for a model.
# This can reduce RAM usage for convolutional models, but may have
# an impact on accuracy.
disable_per_channel_quantization = False
Input layer (9,216 features)
2D conv / pool layer (16 filters, 3 kernel size, 1 layer)
2D conv / pool layer (32 filters, 3 kernel size, 1 layer)
Flatten layer
Dropout (rate 0.25)
Output layer (7 classes)
Model
Model version:
