Embedding kategori

Deep Learning Tingkat Lanjut dengan Keras

Zach Deane Mayer

Data Scientist

Embedding kategori

Input: bilangan bulat
Output: bilangan pecahan
Catatan: Dimensi meningkat; lapisan output diratakan kembali ke 2D

Input

input_tensor = Input(shape=(1,))

Lapisan Embedding

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)

Perataan (Flattening)

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

Gabungkan semuanya

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)

Ayo berlatih!

Deep Learning Tingkat Lanjut dengan Keras