カテゴリ埋め込み

Keras で学ぶ高度な Deep Learning

Zach Deane Mayer

Data Scientist

カテゴリ埋め込み

入力：整数
出力：浮動小数点数
注：次元数が増加：出力層で2Dに平坦化

入力

input_tensor = Input(shape=(1,))

埋め込み層

from tensorflow.keras.layers import Embedding
input_tensor = Input(shape=(1,))
n_teams = 10887
embed_layer = Embedding(input_dim=n_teams,
                        input_length=1,
                        output_dim=1,
                        name='Team-Strength-Lookup')
embed_tensor = embed_layer(input_tensor)

平坦化

from tensorflow.keras.layers import Flatten
flatten_tensor = Flatten()(embed_tensor)

まとめ

input_tensor = Input(shape=(1,))
n_teams =  10887
embed_layer = Embedding(input_dim=n_teams,
                        input_length=1,
                        output_dim=1,
                        name='Team-Strength-Lookup')
embed_tensor = embed_layer(input_tensor)
flatten_tensor = Flatten()(embed_tensor)
model = Model(input_tensor, flatten_tensor)

練習しましょう！

Keras で学ぶ高度な Deep Learning