Kategorinbäddningar

Avancerad djupinlärning med Keras

Zach Deane Mayer

Data Scientist

Kategorinbäddningar

Indata: heltal
Utdata: flyttal
Obs: Ökad dimensionalitet – utdatalagret plattar ut till 2D

Indata

input_tensor = Input(shape=(1,))

Inbäddningslager

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)

Utplattning

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

Sätt ihop det

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)

Nu kör vi en övning!

Avancerad djupinlärning med Keras