Created
April 18, 2024 04:27
-
-
Save raphant/e64e635fdce50065600309a187365e9f to your computer and use it in GitHub Desktop.
Revisions
-
raphant created this gist
Apr 18, 2024 .There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters. Learn more about bidirectional Unicode charactersOriginal file line number Diff line number Diff line change @@ -0,0 +1,61 @@ import tensorflow as tf from tensorflow import keras from tensorflow.keras import layers class PositionalEmbedding(layers.Layer): def __init__(self, sequence_length, vocab_size, embed_dim, **kwargs): super().__init__(**kwargs) self.token_embeddings = layers.Embedding(input_dim=vocab_size, output_dim=embed_dim) self.position_embeddings = layers.Embedding(input_dim=sequence_length, output_dim=embed_dim) self.sequence_length = sequence_length self.vocab_size = vocab_size self.embed_dim = embed_dim def call(self, inputs): length = tf.shape(inputs)[-1] positions = tf.range(start=0, limit=length, delta=1) embedded_tokens = self.token_embeddings(inputs) embedded_positions = self.position_embeddings(positions) return embedded_tokens + embedded_positions class DecoderBlock(layers.Layer): def __init__(self, embed_dim, num_heads, dropout_rate, **kwargs): super().__init__(**kwargs) self.attention = layers.MultiHeadAttention(num_heads=num_heads, key_dim=embed_dim) self.dropout1 = layers.Dropout(dropout_rate) self.norm1 = layers.LayerNormalization() self.dense1 = layers.Dense(embed_dim, activation="relu") self.dense2 = layers.Dense(embed_dim) self.dropout2 = layers.Dropout(dropout_rate) self.norm2 = layers.LayerNormalization() def call(self, inputs, training=False): attn_output = self.attention(inputs, inputs) attn_output = self.dropout1(attn_output, training=training) out1 = self.norm1(inputs + attn_output) dense_output = self.dense1(out1) dense_output = self.dense2(dense_output) dense_output = self.dropout2(dense_output, training=training) return self.norm2(out1 + dense_output) class TransformerDecoder(layers.Layer): def __init__(self, embed_dim, num_heads, num_blocks, dropout_rate, **kwargs): super().__init__(**kwargs) self.num_blocks = num_blocks self.blocks = [DecoderBlock(embed_dim, num_heads, dropout_rate) for _ in range(num_blocks)] def call(self, inputs, training=False): x = inputs for block in self.blocks: x = block(x, training=training) return x def transformer_model(vocab_size, sequence_length, embed_dim, num_heads, num_blocks, dropout_rate, num_classes): inputs = keras.Input(shape=(None,), dtype="int64") x = PositionalEmbedding(sequence_length, vocab_size, embed_dim)(inputs) x = TransformerDecoder(embed_dim, num_heads, num_blocks, dropout_rate)(x, training=True) last_step_output = layers.Lambda(lambda x: x[:, -1, :])(x) # # Output Layer # outputs = Dense(vocab_size, activation='softmax')(last_step_output) outputs = layers.Dense(num_classes, activation="softmax")(last_step_output) return keras.Model(inputs, outputs)