lstm_seq2seq_edit_ferdosi.py

'''
Sequence to sequence example in Keras (character-level).

Edits by Morteza Zakeri:
 - add support for UTF-8 inputs
 - add save model
 - add plot model
 -
 
Date: 1396-10-15
 
Note: Run this script on GPU on massive train set

# Data download
     ./trainset/ferdosi_poem_1.txt

# References

- Sequence to Sequence Learning with Neural Networks
    https://arxiv.org/abs/1409.3215
- Learning Phrase Representations using
    RNN Encoder-Decoder for Statistical Machine Translation
    https://arxiv.org/abs/1406.1078
    
'''
from __future__ import print_function

from keras.models import Model, load_model
from keras.layers import Input, LSTM, Dense
import numpy as np


from keras.utils import plot_model
import datetime
from random import randint

batch_size = 64  # Batch size for training.
epochs = 10  # Number of epochs to train for. # last time I try it!!!
latent_dim = 256  # Latent dimensionality of the encoding space.
num_samples = 100000  # Number of samples to train on. One element of a dataset.

# Path to the txt dataset file on disk, inside the code directory.
#data_path = './trainset/hafez_poem_1.txt' # 419594 character , 12824 mesra and 6412 beyt 
data_path = './trainset/ferdosi_poem_3.txt' # 95884 mesra and 47942 beyt

# Vectorize the data.
input_texts = []
target_texts = []
input_characters = set()
target_characters = set()

# I add support to UTF-8 (for persian (not-enlish) inputs)
lines = open(data_path, encoding="utf8").read().split('\n')
for line in lines[: min(num_samples, len(lines) - 1)]:
    input_text, target_text = line.split('\t')
    # We use "tab" as the "start sequence" character
    # for the targets, and "\n" as "end sequence" character.
    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')

# bulid one-hot vector
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.
    for t, char in enumerate(target_text):
        # decoder_target_data is ahead of decoder_input_data by one timestep
        decoder_input_data[i, t, target_token_index[char]] = 1.
        if t > 0:
            # decoder_target_data will be ahead by one timestep
            # and will not include the start character.
            decoder_target_data[i, t - 1, target_token_index[char]] = 1.

# Define an input sequence and process it.
encoder_inputs = Input(shape=(None, num_encoder_tokens))
#print(type(encoder_inputs))
encoder = LSTM(latent_dim, return_state=True)
#print(type(encoder))

encoder_outputs, state_h, state_c = encoder(encoder_inputs)
#print(type(encoder_outputs))

#encoder_outputs, state_h, state_c = LSTM(latent_dim, return_state=True)(encoder_inputs)

# We discard `encoder_outputs` and only keep the states.
encoder_states = [state_h, state_c]
#input()
#quit()
# Set up the decoder, using `encoder_states` as initial state.
decoder_inputs = Input(shape=(None, num_decoder_tokens))
# We set up our decoder to return full output sequences,
# and to return internal states as well. We don't use the
# return states in the training model, but we will use them in inference.
decoder_lstm = LSTM(latent_dim, return_sequences=True, return_state=True)
decoder_outputs, _, _ = decoder_lstm(decoder_inputs,
                                     initial_state = encoder_states)
decoder_dense = Dense(num_decoder_tokens, activation = 'softmax')
decoder_outputs = decoder_dense(decoder_outputs)

# Define the model that will turn
# `encoder_input_data` & `decoder_input_data` into `decoder_target_data`
model = Model([encoder_inputs, decoder_inputs], decoder_outputs)

# Run training
model.compile(optimizer='rmsprop', loss='categorical_crossentropy')

#del model  # deletes the existing model

# returns a compiled model
# identical to the previous one
#model = load_model('./model_ferdosi/s2s_model_2.h5')

model.fit([encoder_input_data, decoder_input_data], decoder_target_data,
          batch_size=batch_size,
          epochs=epochs,
          validation_split=0.2)


# Save model (requires HDF5 and h5py)
dt = datetime.datetime.now().strftime('%Y%m%d_%H%M%S')
#model.save('./model_ferdosi/s2s_'+ dt + '_edit_ferdosi_ep' + str(epochs) +  '.h5')
model.save('./model_ferdosi/s2s_model_2.h5')

# plot model
#plot_model(model, to_file = './modelpic/seq2seq_model_' + dt + '.png', show_shapes=True, show_layer_names=True)


# ********************************
# Next: inference mode (sampling).
# Here's the drill:
# 1) encode input and retrieve initial decoder state
# 2) run one step of decoder with this initial state
# and a "start of sequence" token as target.
# Output will be the next target token
# 3) Repeat with the current target token and current states

# Define sampling models
## encoder model
encoder_model = Model(encoder_inputs, encoder_states)

## decoder model
decoder_state_input_h = Input(shape=(latent_dim,))
decoder_state_input_c = Input(shape=(latent_dim,))
decoder_states_inputs = [decoder_state_input_h, decoder_state_input_c]
decoder_outputs, state_h, state_c = decoder_lstm(
    decoder_inputs, initial_state = decoder_states_inputs)
decoder_states = [state_h, state_c]
decoder_outputs = decoder_dense(decoder_outputs)
decoder_model = Model(
    [decoder_inputs] + decoder_states_inputs,
    [decoder_outputs] + decoder_states)

## plot sampling models
#plot_model(encoder_model, to_file = './modelpic/sampling_encoder_model_' + dt + '.png', show_shapes=True, show_layer_names=True)
#plot_model(decoder_model, to_file = './modelpic/sampling_decoder_model_' + dt + '.png', show_shapes=True, show_layer_names=True)

# Reverse-lookup token index to decode sequences back to
# something readable.
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_sequence(input_seq):
    # Encode the input as state vectors.
    states_value = encoder_model.predict(input_seq)

    # Generate empty target sequence of length 1.
    target_seq = np.zeros((1, 1, num_decoder_tokens))
    # Populate the first character of target sequence with the start character.
    target_seq[0, 0, target_token_index['\t']] = 1.

    # Sampling loop for a batch of sequences
    # (to simplify, here we assume a batch of size 1).
    stop_condition = False
    decoded_sentence = ''
    while not stop_condition:
        output_tokens, h, c = decoder_model.predict(
            [target_seq] + states_value)

        # Sample a token
        sampled_token_index = np.argmax(output_tokens[0, -1, :])
        sampled_char = reverse_target_char_index[sampled_token_index]
        decoded_sentence += sampled_char

        # Exit condition: either hit max length
        # or find stop character.
        if (sampled_char == '\n' or
           len(decoded_sentence) > max_decoder_seq_length-1):
            stop_condition = True

        # Update the target sequence (of length 1).
        target_seq = np.zeros((1, 1, num_decoder_tokens))
        target_seq[0, 0, sampled_token_index] = 1.

        # Update states
        states_value = [h, c]

    return decoded_sentence


index = randint(0,len(input_texts)-1)
input_seq = encoder_input_data[index: index+1]
decoded_sentence = decode_sequence(input_seq)
gen_poem = ''
#print('-----')
print('Input verse:', input_texts[index])
gen_poem += input_texts[0] + '\n'
#print('Next verses:', decoded_sentence)
gen_poem += decoded_sentence + '\n'


encoder_input_data_test = np.zeros(
          (1, max_encoder_seq_length, num_encoder_tokens), dtype='float32')

for seq_index in range(64):   
    print('Input verse:', decoded_sentence)
    decoded_sentence = decoded_sentence.replace('\n', '')
    decoded_sentence = decoded_sentence.replace('\t', '')
    print('Input verse:', decoded_sentence)
    for t, char in enumerate(decoded_sentence):
        encoder_input_data_test[0, t, input_token_index[char]] = 1.
        
    input_seq = encoder_input_data_test[0: 1]
    
    decoded_sentence = decode_sequence(input_seq)
    print('-----')
    print('Next verses:', decoded_sentence)
    gen_poem +=  decoded_sentence + '\n'
    
new_poem = './result_ferdosi/new_poem_ferdosi' + dt + '_ep' + str(epochs) + '.txt'    
with open(new_poem, 'w', encoding='utf8') as nhp:
     nhp.write(str(gen_poem))