Author: Apoorv Nandan
Date created: 2020/05/10
Last modified: 2024/01/18
Description: Implement a Transformer block as a Keras layer and use it for text classification.
View in Colab • GitHub source
import keras
from keras import ops
from keras import layers
class TransformerBlock(layers.Layer):
def __init__(self, embed_dim, num_heads, ff_dim, rate=0.1):
super().__init__()
self.att = layers.MultiHeadAttention(num_heads=num_heads, key_dim=embed_dim)
self.ffn = keras.Sequential(
[layers.Dense(ff_dim, activation="relu"), layers.Dense(embed_dim),]
)
self.layernorm1 = layers.LayerNormalization(epsilon=1e-6)
self.layernorm2 = layers.LayerNormalization(epsilon=1e-6)
self.dropout1 = layers.Dropout(rate)
self.dropout2 = layers.Dropout(rate)
def call(self, inputs):
attn_output = self.att(inputs, inputs)
attn_output = self.dropout1(attn_output)
out1 = self.layernorm1(inputs + attn_output)
ffn_output = self.ffn(out1)
ffn_output = self.dropout2(ffn_output)
return self.layernorm2(out1 + ffn_output)
Two separate embedding layers, one for tokens, one for token index (positions).
class TokenAndPositionEmbedding(layers.Layer):
def __init__(self, maxlen, vocab_size, embed_dim):
super().__init__()
self.token_emb = layers.Embedding(input_dim=vocab_size, output_dim=embed_dim)
self.pos_emb = layers.Embedding(input_dim=maxlen, output_dim=embed_dim)
def call(self, x):
maxlen = ops.shape(x)[-1]
positions = ops.arange(start=0, stop=maxlen, step=1)
positions = self.pos_emb(positions)
x = self.token_emb(x)
return x + positions
vocab_size = 20000 # Only consider the top 20k words
maxlen = 200 # Only consider the first 200 words of each movie review
(x_train, y_train), (x_val, y_val) = keras.datasets.imdb.load_data(num_words=vocab_size)
print(len(x_train), "Training sequences")
print(len(x_val), "Validation sequences")
x_train = keras.utils.pad_sequences(x_train, maxlen=maxlen)
x_val = keras.utils.pad_sequences(x_val, maxlen=maxlen)
Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/imdb.npz
17465344/17464789 [==============================] - 0s 0us/step
<string>:6: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/datasets/imdb.py:159: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
x_train, y_train = np.array(xs[:idx]), np.array(labels[:idx])
/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/datasets/imdb.py:160: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
x_test, y_test = np.array(xs[idx:]), np.array(labels[idx:])
25000 Training sequences
25000 Validation sequences
Transformer layer outputs one vector for each time step of our input sequence. Here, we take the mean across all time steps and use a feed forward network on top of it to classify text.
embed_dim = 32 # Embedding size for each token
num_heads = 2 # Number of attention heads
ff_dim = 32 # Hidden layer size in feed forward network inside transformer
inputs = layers.Input(shape=(maxlen,))
embedding_layer = TokenAndPositionEmbedding(maxlen, vocab_size, embed_dim)
x = embedding_layer(inputs)
transformer_block = TransformerBlock(embed_dim, num_heads, ff_dim)
x = transformer_block(x)
x = layers.GlobalAveragePooling1D()(x)
x = layers.Dropout(0.1)(x)
x = layers.Dense(20, activation="relu")(x)
x = layers.Dropout(0.1)(x)
outputs = layers.Dense(2, activation="softmax")(x)
model = keras.Model(inputs=inputs, outputs=outputs)
model.compile(optimizer="adam", loss="sparse_categorical_crossentropy", metrics=["accuracy"])
history = model.fit(
x_train, y_train, batch_size=32, epochs=2, validation_data=(x_val, y_val)
)
Epoch 1/2
782/782 [==============================] - 15s 18ms/step - loss: 0.5112 - accuracy: 0.7070 - val_loss: 0.3598 - val_accuracy: 0.8444
Epoch 2/2
782/782 [==============================] - 13s 17ms/step - loss: 0.1942 - accuracy: 0.9297 - val_loss: 0.2977 - val_accuracy: 0.8745