179 lines
6.5 KiB
Python
179 lines
6.5 KiB
Python
import numpy as np
|
|
import keras
|
|
from nltk.translate.bleu_score import corpus_bleu
|
|
|
|
# Configuration
|
|
batch_size = 64
|
|
epochs = 350
|
|
latent_dim = 256
|
|
num_samples = 50000
|
|
data_path = "fra.txt"
|
|
|
|
# Preparing data
|
|
input_texts = []
|
|
target_texts = []
|
|
input_characters = set()
|
|
target_characters = set()
|
|
with open(data_path, "r", encoding="utf-8") as f:
|
|
lines = f.read().split("\n")
|
|
for line in lines[: min(num_samples, len(lines) - 1)]:
|
|
input_text, target_text, _ = line.split("\t")
|
|
target_text = "\t" + target_text + "\n"
|
|
input_texts.append(input_text)
|
|
target_texts.append(target_text)
|
|
for char in input_text:
|
|
if char not in input_characters:
|
|
input_characters.add(char)
|
|
for char in target_text:
|
|
if char not in target_characters:
|
|
target_characters.add(char)
|
|
|
|
input_characters = sorted(list(input_characters))
|
|
target_characters = sorted(list(target_characters))
|
|
num_encoder_tokens = len(input_characters)
|
|
num_decoder_tokens = len(target_characters)
|
|
max_encoder_seq_length = max([len(txt) for txt in input_texts])
|
|
max_decoder_seq_length = max([len(txt) for txt in target_texts])
|
|
|
|
print("Number of samples:", len(input_texts))
|
|
print("Number of unique input tokens:", num_encoder_tokens)
|
|
print("Number of unique output tokens:", num_decoder_tokens)
|
|
print("Max sequence length for inputs:", max_encoder_seq_length)
|
|
print("Max sequence length for outputs:", max_decoder_seq_length)
|
|
|
|
input_token_index = dict([(char, i) for i, char in enumerate(input_characters)])
|
|
target_token_index = dict([(char, i) for i, char in enumerate(target_characters)])
|
|
|
|
encoder_input_data = np.zeros(
|
|
(len(input_texts), max_encoder_seq_length, num_encoder_tokens),
|
|
dtype="float32",
|
|
)
|
|
decoder_input_data = np.zeros(
|
|
(len(input_texts), max_decoder_seq_length, num_decoder_tokens),
|
|
dtype="float32",
|
|
)
|
|
decoder_target_data = np.zeros(
|
|
(len(input_texts), max_decoder_seq_length, num_decoder_tokens),
|
|
dtype="float32",
|
|
)
|
|
|
|
for i, (input_text, target_text) in enumerate(zip(input_texts, target_texts)):
|
|
for t, char in enumerate(input_text):
|
|
encoder_input_data[i, t, input_token_index[char]] = 1.0
|
|
encoder_input_data[i, t + 1 :, input_token_index[" "]] = 1.0
|
|
for t, char in enumerate(target_text):
|
|
decoder_input_data[i, t, target_token_index[char]] = 1.0
|
|
if t > 0:
|
|
decoder_target_data[i, t - 1, target_token_index[char]] = 1.0
|
|
decoder_input_data[i, t + 1 :, target_token_index[" "]] = 1.0
|
|
decoder_target_data[i, t:, target_token_index[" "]] = 1.0
|
|
|
|
# Creating model
|
|
encoder_inputs = keras.Input(shape=(None, num_encoder_tokens))
|
|
encoder = keras.layers.LSTM(latent_dim, return_state=True)
|
|
encoder_outputs, state_h, state_c = encoder(encoder_inputs)
|
|
encoder_states = [state_h, state_c]
|
|
|
|
decoder_inputs = keras.Input(shape=(None, num_decoder_tokens))
|
|
decoder_lstm = keras.layers.LSTM(latent_dim, return_sequences=True, return_state=True)
|
|
decoder_outputs, _, _ = decoder_lstm(decoder_inputs, initial_state=encoder_states)
|
|
decoder_dense = keras.layers.Dense(num_decoder_tokens, activation="softmax")
|
|
decoder_outputs = decoder_dense(decoder_outputs)
|
|
|
|
model = keras.Model([encoder_inputs, decoder_inputs], decoder_outputs)
|
|
|
|
# Creating the training model
|
|
model.compile(
|
|
optimizer="rmsprop", loss="categorical_crossentropy", metrics=["accuracy"]
|
|
)
|
|
model.fit(
|
|
[encoder_input_data, decoder_input_data],
|
|
decoder_target_data,
|
|
batch_size=batch_size,
|
|
epochs=epochs,
|
|
validation_split=0.2,
|
|
)
|
|
model.save("s2s_model.keras")
|
|
|
|
# Sampling
|
|
model = keras.models.load_model("s2s_model.keras")
|
|
|
|
encoder_inputs = model.input[0]
|
|
encoder_outputs, state_h_enc, state_c_enc = model.layers[2].output
|
|
encoder_states = [state_h_enc, state_c_enc]
|
|
encoder_model = keras.Model(encoder_inputs, encoder_states)
|
|
|
|
decoder_inputs = model.input[1]
|
|
decoder_state_input_h = keras.Input(shape=(latent_dim,))
|
|
decoder_state_input_c = keras.Input(shape=(latent_dim,))
|
|
decoder_states_inputs = [decoder_state_input_h, decoder_state_input_c]
|
|
decoder_lstm = model.layers[3]
|
|
decoder_outputs, state_h_dec, state_c_dec = decoder_lstm(
|
|
decoder_inputs, initial_state=decoder_states_inputs
|
|
)
|
|
decoder_states = [state_h_dec, state_c_dec]
|
|
decoder_dense = model.layers[4]
|
|
decoder_outputs = decoder_dense(decoder_outputs)
|
|
decoder_model = keras.Model(
|
|
[decoder_inputs] + decoder_states_inputs, [decoder_outputs] + decoder_states
|
|
)
|
|
|
|
reverse_input_char_index = dict((i, char) for char, i in input_token_index.items())
|
|
reverse_target_char_index = dict((i, char) for char, i in target_token_index.items())
|
|
|
|
def decode_sequences(input_seqs):
|
|
states_values = encoder_model.predict(input_seqs, verbose=0)
|
|
target_seqs = np.zeros((len(input_seqs), 1, num_decoder_tokens))
|
|
target_seqs[:, 0, target_token_index["\t"]] = 1.0
|
|
decoded_sentences = [""] * len(input_seqs)
|
|
stop_conditions = np.zeros(len(input_seqs), dtype=bool)
|
|
|
|
while not np.all(stop_conditions):
|
|
output_tokens, h, c = decoder_model.predict(
|
|
[target_seqs] + states_values, verbose=0
|
|
)
|
|
|
|
sampled_token_indices = np.argmax(output_tokens[:, -1, :], axis=1)
|
|
sampled_chars = [
|
|
reverse_target_char_index[idx] for idx in sampled_token_indices
|
|
]
|
|
|
|
for i, char in enumerate(sampled_chars):
|
|
decoded_sentences[i] += char
|
|
if char == "\n" or len(decoded_sentences[i]) > max_decoder_seq_length:
|
|
stop_conditions[i] = True
|
|
|
|
target_seqs = np.zeros((len(input_seqs), 1, num_decoder_tokens))
|
|
for i, token_index in enumerate(sampled_token_indices):
|
|
target_seqs[i, 0, token_index] = 1.0
|
|
|
|
states_values = [h, c]
|
|
|
|
return decoded_sentences
|
|
|
|
# BLEU score evaluation
|
|
def calculate_bleu_score(input_texts, target_texts, num_samples=500):
|
|
input_seqs = np.zeros(
|
|
(num_samples, max_encoder_seq_length, num_encoder_tokens), dtype="float32"
|
|
)
|
|
for i, input_text in enumerate(input_texts[:num_samples]):
|
|
for t, char in enumerate(input_text):
|
|
input_seqs[i, t, input_token_index[char]] = 1.0
|
|
input_seqs[i, t + 1 :, input_token_index[" "]] = 1.0
|
|
|
|
decoded_sentences = decode_sequences(input_seqs)
|
|
|
|
references = [[list(text.strip())] for text in target_texts[:num_samples]]
|
|
hypotheses = [list(text.strip()) for text in decoded_sentences]
|
|
bleu = corpus_bleu(references, hypotheses)
|
|
print("BLEU Score:", bleu)
|
|
|
|
print("\nExample Translations:")
|
|
for i in range(10):
|
|
print("Input:", input_texts[i])
|
|
print("Target:", target_texts[i])
|
|
print("Translation:", decoded_sentences[i])
|
|
print()
|
|
|
|
calculate_bleu_score(input_texts, target_texts)
|