ValueError: Shapes (None, 5) and (None, 15, 5) are incompatible

Question

I want to implement a Hierarchical attention mechanism for document classification presented by Yang. But I want to replace LSTM with Transformer.

I used Apoorv Nandan's text classification with Transformer: https://keras.io/examples/nlp/text_classification_with_transformer/

I have implemented Transformer hierarchically to classification. One for sentence representation and another one for document representation. The code is as follow:

import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import layers
from keras.utils.np_utils import to_categorical


class MultiHeadSelfAttention(layers.Layer):
    def __init__(self, embed_dim, num_heads=8):
        super(MultiHeadSelfAttention, self).__init__()
        self.embed_dim = embed_dim
        self.num_heads = num_heads
        if embed_dim % num_heads != 0:
            raise ValueError(
                f"embedding dimension = {embed_dim} should be divisible by number of heads = {num_heads}"
            )
        self.projection_dim = embed_dim // num_heads
        self.query_dense = layers.Dense(embed_dim)
        self.key_dense = layers.Dense(embed_dim)
        self.value_dense = layers.Dense(embed_dim)
        self.combine_heads = layers.Dense(embed_dim)

    def attention(self, query, key, value):
        score = tf.matmul(query, key, transpose_b=True)
        dim_key = tf.cast(tf.shape(key)[-1], tf.float32)
        scaled_score = score / tf.math.sqrt(dim_key)
        weights = tf.nn.softmax(scaled_score, axis=-1)
        output = tf.matmul(weights, value)
        return output, weights

    def separate_heads(self, x, batch_size):
        x = tf.reshape(x, (batch_size, -1, self.num_heads, self.projection_dim))
        return tf.transpose(x, perm=[0, 2, 1, 3])

    def call(self, inputs):
        # x.shape = [batch_size, seq_len, embedding_dim]
        batch_size = tf.shape(inputs)[0]
        query = self.query_dense(inputs)  # (batch_size, seq_len, embed_dim)
        key = self.key_dense(inputs)  # (batch_size, seq_len, embed_dim)
        value = self.value_dense(inputs)  # (batch_size, seq_len, embed_dim)
        query = self.separate_heads(
            query, batch_size
        )  # (batch_size, num_heads, seq_len, projection_dim)
        key = self.separate_heads(
            key, batch_size
        )  # (batch_size, num_heads, seq_len, projection_dim)
        value = self.separate_heads(
            value, batch_size
        )  # (batch_size, num_heads, seq_len, projection_dim)
        attention, weights = self.attention(query, key, value)
        attention = tf.transpose(
            attention, perm=[0, 2, 1, 3]
        )  # (batch_size, seq_len, num_heads, projection_dim)
        concat_attention = tf.reshape(
            attention, (batch_size, -1, self.embed_dim)
        )  # (batch_size, seq_len, embed_dim)
        output = self.combine_heads(
            concat_attention
        )  # (batch_size, seq_len, embed_dim)
        return output

    def compute_output_shape(self, input_shape):
        # it does not change the shape of its input
        return input_shape


class TransformerBlock(layers.Layer):
    def __init__(self, embed_dim, num_heads, ff_dim, dropout_rate, name=None):
        super(TransformerBlock, self).__init__(name=name)
        self.att = MultiHeadSelfAttention(embed_dim, num_heads)
        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(dropout_rate)
        self.dropout2 = layers.Dropout(dropout_rate)

    def call(self, inputs, training):
        attn_output = self.att(inputs)
        attn_output = self.dropout1(attn_output, training=training)
        out1 = self.layernorm1(inputs + attn_output)
        ffn_output = self.ffn(out1)
        ffn_output = self.dropout2(ffn_output, training=training)
        return self.layernorm2(out1 + ffn_output)

    def compute_output_shape(self, input_shape):
        # it does not change the shape of its input
        return input_shape


class TokenAndPositionEmbedding(layers.Layer):
    def __init__(self, maxlen, vocab_size, embed_dim, name=None):
        super(TokenAndPositionEmbedding, self).__init__(name=name)
        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 = tf.shape(x)[-1]
        positions = tf.range(start=0, limit=maxlen, delta=1)
        positions = self.pos_emb(positions)
        x = self.token_emb(x)
        return x + positions

    def compute_output_shape(self, input_shape):
        # it changes the shape from (batch_size, maxlen) to (batch_size, maxlen, embed_dim)
        return input_shape + (self.pos_emb.output_dim,)



# Lower level (produce a representation of each sentence):

embed_dim = 100  # Embedding size for each token
num_heads = 2  # Number of attention heads
ff_dim = 64  # Hidden layer size in feed forward network inside transformer
L1_dense_units = 100  # Size of the sentence-level representations output by the word-level model
dropout_rate = 0.1
vocab_size = 1000
class_number = 5
max_docs = 10000
max_sentences = 15
max_words = 60

word_input = layers.Input(shape=(max_words,), name='word_input')
word_embedding = TokenAndPositionEmbedding(maxlen=max_words, vocab_size=vocab_size,
                                           embed_dim=embed_dim, name='word_embedding')(word_input)
word_transformer = TransformerBlock(embed_dim=embed_dim, num_heads=num_heads, ff_dim=ff_dim,
                                    dropout_rate=dropout_rate, name='word_transformer')(word_embedding)
word_pool = layers.GlobalAveragePooling1D(name='word_pooling')(word_transformer)
word_drop = layers.Dropout(dropout_rate, name='word_drop')(word_pool)
word_dense = layers.Dense(L1_dense_units, activation="relu", name='word_dense')(word_drop)
word_encoder = keras.Model(word_input, word_dense)

word_encoder.summary()

# =========================================================================
# Upper level (produce a representation of each document):

L2_dense_units = 100

sentence_input = layers.Input(shape=(max_sentences, max_words), name='sentence_input')

# This is the line producing "NotImplementedError":
sentence_encoder = tf.keras.layers.TimeDistributed(word_encoder, name='sentence_encoder')(sentence_input)

sentence_transformer = TransformerBlock(embed_dim=L1_dense_units, num_heads=num_heads, ff_dim=ff_dim,
                               dropout_rate=dropout_rate, name='sentence_transformer')(sentence_encoder)
sentence_dense = layers.TimeDistributed(layers.Dense(int(L2_dense_units)),name='sentence_dense')(sentence_transformer)
sentence_out = layers.Dropout(dropout_rate)(sentence_dense)
preds = layers.Dense(class_number , activation='softmax', name='sentence_output')(sentence_out)

model = keras.Model(sentence_input, preds)
model.summary()

#==========================================================================

Everything is OK(for testing you can copy and paste it in googlecolab). But when I compile and fit the model by following codes, it throws an error:

X = tf.random.uniform(shape=(max_docs, max_sentences, max_words), minval=1, maxval=1000, dtype=tf.dtypes.int32, seed=1)

y = tf.random.uniform(shape=(max_docs, ), minval=0, maxval=class_number , dtype=tf.dtypes.int32, seed=1)
y = to_categorical(y)
    model.compile(optimizer="adam", loss="categorical_crossentropy", metrics=["accuracy"])
    history = model.fit(
        X, y, batch_size=32, epochs=25,
    )

The error is:

ValueError: Shapes (None, 5) and (None, 15, 5) are incompatible

Answer 1

When I had a similar error, I found that a Flatten() layer helped, I had incompatible shapes of (None, x, y) and (None, y).

If you try to provide a flatten layer for the part that gives you the (None, 15, 5), then it should output something like (None, 75).

The flatten layer merely removes dimensions, when I was doing this I got the output as (None, x y) and due to the way Tensorflow works, it was able to match both shapes as x y is obviously a factor of just y.