-
Caroline DE POURTALES authoredCaroline DE POURTALES authored
Linker.py 18.33 KiB
import os
import sys
import datetime
import time
import torch
import torch.nn.functional as F
from torch.nn import Sequential, LayerNorm, Dropout
from torch.optim import AdamW
from torch.utils.data import TensorDataset, random_split
from torch.utils.tensorboard import SummaryWriter
from tqdm import tqdm
from Configuration import Configuration
from Linker.AtomEmbedding import AtomEmbedding
from Linker.AtomTokenizer import AtomTokenizer
from Linker.MHA import AttentionDecoderLayer
from Linker.Sinkhorn import sinkhorn_fn_no_exp as sinkhorn
from Linker.atom_map import atom_map
from Linker.eval import mesure_accuracy, SinkhornLoss
from Linker.utils_linker import find_pos_neg_idexes, get_atoms_batch, FFN, get_axiom_links, get_pos_encoding_for_s_idx, \
get_neg_encoding_for_s_idx
from Supertagger import *
from utils import pad_sequence
def format_time(elapsed):
'''
Takes a time in seconds and returns a string hh:mm:ss
'''
# Round to the nearest second.
elapsed_rounded = int(round(elapsed))
# Format as hh:mm:ss
return str(datetime.timedelta(seconds=elapsed_rounded))
def output_create_dir():
"""
Create le output dir for tensorboard and checkpoint
@return: output dir, tensorboard writter
"""
from datetime import datetime
outpout_path = 'TensorBoard'
training_dir = os.path.join(outpout_path, 'Tranning_' + datetime.today().strftime('%d-%m_%H-%M'))
logs_dir = os.path.join(training_dir, 'logs')
writer = SummaryWriter(log_dir=logs_dir)
return training_dir, writer
class Linker(Module):
def __init__(self, supertagger_path_model):
super(Linker, self).__init__()
self.dim_embedding_atoms = int(Configuration.modelLinkerConfig['dim_embedding_atoms'])
self.nhead = int(Configuration.modelDecoderConfig['nhead'])
dim_pre_sinkhorn_transfo = int(Configuration.modelLinkerConfig['dim_pre_sinkhorn_transfo'])
dim_polarity_transfo = int(Configuration.modelLinkerConfig['dim_polarity_transfo'])
self.sinkhorn_iters = int(Configuration.modelLinkerConfig['sinkhorn_iters'])
self.max_atoms_in_sentence = int(Configuration.datasetConfig['max_atoms_in_sentence'])
self.max_atoms_in_one_type = int(Configuration.datasetConfig['max_atoms_in_one_type'])
atom_vocab_size = int(Configuration.datasetConfig['atom_vocab_size'])
learning_rate = float(Configuration.modelTrainingConfig['learning_rate'])
self.dropout = Dropout(0.1)
self.device = "cpu"
supertagger = SuperTagger()
supertagger.load_weights(supertagger_path_model)
self.Supertagger = supertagger
self.atom_map = atom_map
self.sub_atoms_type_list = ['cl_r', 'pp', 'n', 'np', 'cl_y', 'txt', 's']
self.padding_id = self.atom_map['[PAD]']
self.atoms_tokenizer = AtomTokenizer(atom_map, self.max_atoms_in_sentence)
self.inverse_map = self.atoms_tokenizer.inverse_atom_map
self.atoms_embedding = AtomEmbedding(self.dim_embedding_atoms, atom_vocab_size, self.padding_id)
self.linker_encoder = AttentionDecoderLayer()
self.pos_transformation = Sequential(
FFN(self.dim_embedding_atoms, dim_polarity_transfo, 0.1, d_out=dim_pre_sinkhorn_transfo),
LayerNorm(dim_pre_sinkhorn_transfo, eps=1e-12)
)
self.neg_transformation = Sequential(
FFN(self.dim_embedding_atoms, dim_polarity_transfo, 0.1, d_out=dim_pre_sinkhorn_transfo),
LayerNorm(dim_pre_sinkhorn_transfo, eps=1e-12)
)
self.cross_entropy_loss = SinkhornLoss()
self.optimizer = AdamW(self.parameters(),
lr=learning_rate)
self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
self.to(self.device)
def __preprocess_data(self, batch_size, df_axiom_links, validation_rate=0.1):
r"""
Args:
batch_size : int
df_axiom_links pandas DataFrame
validation_rate
Returns:
the training dataloader and the validation dataloader. They contains the list of atoms, their polarities, the axiom links, the sentences tokenized, sentence mask
"""
print("Start preprocess Data")
sentences_batch = df_axiom_links["X"].tolist()
sentences_tokens, sentences_mask = self.Supertagger.sent_tokenizer.fit_transform_tensors(sentences_batch)
atoms_batch = get_atoms_batch(df_axiom_links["Z"])
atoms_batch_tokenized = self.atoms_tokenizer.convert_batchs_to_ids(atoms_batch)
atoms_polarity_batch = find_pos_neg_idexes(self.max_atoms_in_sentence, df_axiom_links["Z"])
truth_links_batch = get_axiom_links(self.max_atoms_in_one_type, self.sub_atoms_type_list, atoms_polarity_batch,
df_axiom_links["Y"])
truth_links_batch = truth_links_batch.permute(1, 0, 2)
# Construction tensor dataset
dataset = TensorDataset(atoms_batch_tokenized, atoms_polarity_batch, truth_links_batch, sentences_tokens,
sentences_mask)
if validation_rate > 0.0:
train_size = int(0.9 * len(dataset))
val_size = len(dataset) - train_size
train_dataset, val_dataset = random_split(dataset, [train_size, val_size])
validation_dataloader = torch.utils.data.DataLoader(val_dataset, batch_size=batch_size, shuffle=False)
else:
validation_dataloader = None
train_dataset = dataset
training_dataloader = torch.utils.data.DataLoader(train_dataset, batch_size=batch_size, shuffle=False)
print("End preprocess Data")
return training_dataloader, validation_dataloader
def make_decoder_mask(self, atoms_token):
decoder_attn_mask = torch.ones_like(atoms_token, dtype=torch.float64, device=self.device)
decoder_attn_mask[atoms_token.eq(self.padding_id)] = 0.0
return decoder_attn_mask.unsqueeze(1).repeat(1, atoms_token.shape[1], 1).repeat(self.nhead, 1, 1)
def forward(self, atoms_batch_tokenized, atoms_polarity_batch, sents_embedding, sents_mask=None):
r"""
Args:
atoms_batch_tokenized : (batch_size, max_atoms_in_one_sentence) flattened categories
atoms_polarity_batch : (batch_size, max_atoms_in_one_sentence) flattened categories polarities
sents_embedding : (batch_size, len_sentence, dim_encoder) output of BERT for context
sents_mask : mask from BERT tokenizer
Returns:
link_weights : atom_vocab_size, batch-size, max_atoms_in_one_cat, max_atoms_in_one_cat) log probabilities
"""
# atoms embedding
atoms_embedding = self.atoms_embedding(atoms_batch_tokenized)
# MHA ou LSTM avec sortie de BERT
sents_mask = sents_mask.unsqueeze(1).repeat(self.nhead, self.max_atoms_in_sentence, 1).to(torch.float64)
atoms_encoding = self.linker_encoder(atoms_embedding, sents_embedding, sents_mask,
self.make_decoder_mask(atoms_batch_tokenized))
link_weights = []
for atom_type in self.sub_atoms_type_list:
pos_encoding = pad_sequence(
[get_pos_encoding_for_s_idx(self.dim_embedding_atoms, atoms_encoding, atoms_batch_tokenized,
atoms_polarity_batch, atom_type, self.inverse_map, s_idx)
for s_idx in range(len(atoms_polarity_batch))], padding_value=0,
max_len=self.max_atoms_in_one_type // 2)
neg_encoding = pad_sequence(
[get_neg_encoding_for_s_idx(self.dim_embedding_atoms, atoms_encoding, atoms_batch_tokenized,
atoms_polarity_batch, atom_type, self.inverse_map, s_idx)
for s_idx in range(len(atoms_polarity_batch))], padding_value=0,
max_len=self.max_atoms_in_one_type // 2)
pos_encoding = self.pos_transformation(pos_encoding)
neg_encoding = self.neg_transformation(neg_encoding)
weights = torch.bmm(pos_encoding, neg_encoding.transpose(2, 1))
link_weights.append(sinkhorn(weights, iters=self.sinkhorn_iters))
total_link_weights = torch.stack(link_weights)
link_weights_per_batch = total_link_weights.permute(1, 0, 2, 3)
return F.log_softmax(link_weights_per_batch, dim=3)
def train_linker(self, df_axiom_links, validation_rate=0.1, epochs=20,
batch_size=32, checkpoint=True, tensorboard=False):
r"""
Args:
df_axiom_links : pandas dataFrame containing the atoms anoted with _i
validation_rate : float
epochs : int
batch_size : int
checkpoint : boolean
tensorboard : boolean
Returns:
Final accuracy and final loss
"""
training_dataloader, validation_dataloader = self.__preprocess_data(batch_size, df_axiom_links,
validation_rate)
if checkpoint or tensorboard:
checkpoint_dir, writer = output_create_dir()
for epoch_i in range(epochs):
print("")
print('======== Epoch {:} / {:} ========'.format(epoch_i + 1, epochs))
print('Training...')
avg_train_loss, avg_accuracy_train, training_time = self.train_epoch(training_dataloader)
print("")
print(f'Epoch: {epoch_i + 1:02} | Epoch Time: {training_time}')
print(f'\tTrain Loss: {avg_train_loss:.3f} | Train Acc: {avg_accuracy_train * 100:.2f}%')
if validation_rate > 0.0:
loss_test, accuracy_test = self.eval_epoch(validation_dataloader, self.cross_entropy_loss)
print(f'\tVal Loss: {loss_test:.3f} | Val Acc: {accuracy_test * 100:.2f}%')
if checkpoint:
self.__checkpoint_save(
path=os.path.join("Output", 'linker' + datetime.today().strftime('%d-%m_%H-%M') + '.pt'))
if tensorboard:
writer.add_scalars(f'Accuracy', {
'Train': avg_accuracy_train}, epoch_i)
writer.add_scalars(f'Loss', {
'Train': avg_train_loss}, epoch_i)
if validation_rate > 0.0:
writer.add_scalars(f'Accuracy', {
'Validation': accuracy_test}, epoch_i)
writer.add_scalars(f'Loss', {
'Validation': loss_test}, epoch_i)
print('\n')
def train_epoch(self, training_dataloader):
r""" Train epoch
Args:
training_dataloader : DataLoader from torch , contains atoms, polarities, axiom_links, sents_tokenized, sents_masks
validation_dataloader : DataLoader from torch , contains atoms, polarities, axiom_links, sents_tokenized, sents_masks
Returns:
accuracy on validation set
loss on train set
"""
self.train()
# Reset the total loss for this epoch.
epoch_loss = 0
accuracy_train = 0
t0 = time.time()
# For each batch of training data...
with tqdm(training_dataloader, unit="batch") as tepoch:
for batch in tepoch:
# Unpack this training batch from our dataloader
batch_atoms = batch[0].to(self.device)
batch_polarity = batch[1].to(self.device)
batch_true_links = batch[2].to(self.device)
batch_sentences_tokens = batch[3].to(self.device)
batch_sentences_mask = batch[4].to(self.device)
self.optimizer.zero_grad()
# get sentence embedding from BERT which is already trained
logits, sentences_embedding = self.Supertagger.forward(batch_sentences_tokens, batch_sentences_mask)
# Run the kinker on the categories predictions
logits_predictions = self(batch_atoms, batch_polarity, sentences_embedding, batch_sentences_mask)
linker_loss = self.cross_entropy_loss(logits_predictions, batch_true_links)
# Perform a backward pass to calculate the gradients.
epoch_loss += float(linker_loss)
linker_loss.backward()
# This is to help prevent the "exploding gradients" problem.
# torch.nn.utils.clip_grad_norm_(model.parameters(), max_norm=5.0, norm_type=2)
# Update parameters and take a step using the computed gradient.
self.optimizer.step()
pred_axiom_links = torch.argmax(logits_predictions, dim=3)
accuracy_train += mesure_accuracy(batch_true_links, pred_axiom_links)
# Measure how long this epoch took.
training_time = format_time(time.time() - t0)
avg_train_loss = epoch_loss / len(training_dataloader)
avg_accuracy_train = accuracy_train / len(training_dataloader)
return avg_train_loss, avg_accuracy_train, training_time
def predict(self, categories, sents_embedding, sents_mask=None):
r"""Prediction from categories output by BERT and hidden_state from BERT
Args:
categories : (batch_size, len_sentence)
sents_embedding : (batch_size, len_sentence, dim_encoder) output of BERT for context
sents_mask
Returns:
axiom_links : atom_vocab_size, batch-size, max_atoms_in_one_cat)
"""
self.eval()
with torch.no_grad():
# get atoms
atoms_batch = get_atoms_batch(categories)
atoms_tokenized = self.atoms_tokenizer.convert_batchs_to_ids(atoms_batch)
# get polarities
polarities = find_pos_neg_idexes(self.max_atoms_in_sentence, categories)
# atoms embedding
atoms_embedding = self.atoms_embedding(atoms_tokenized)
# MHA ou LSTM avec sortie de BERT
atoms_encoding = self.linker_encoder(atoms_embedding, sents_embedding, sents_mask,
self.make_decoder_mask(atoms_tokenized))
link_weights = []
for atom_type in self.sub_atoms_type_list:
pos_encoding = pad_sequence(
[get_pos_encoding_for_s_idx(self.dim_embedding_atoms, atoms_encoding, atoms_tokenized,
polarities, atom_type, self.inverse_map, s_idx)
for s_idx in range(len(polarities))], padding_value=0,
max_len=self.max_atoms_in_one_type // 2)
neg_encoding = pad_sequence(
[get_neg_encoding_for_s_idx(self.dim_embedding_atoms, atoms_encoding, atoms_tokenized,
polarities, atom_type, self.inverse_map, s_idx)
for s_idx in range(len(polarities))], padding_value=0,
max_len=self.max_atoms_in_one_type // 2)
pos_encoding = self.pos_transformation(pos_encoding)
neg_encoding = self.neg_transformation(neg_encoding)
weights = torch.bmm(pos_encoding, neg_encoding.transpose(2, 1))
link_weights.append(sinkhorn(weights, iters=3))
logits_predictions = torch.stack(link_weights).permute(1, 0, 2, 3)
axiom_links = torch.argmax(F.log_softmax(logits_predictions, dim=3), dim=3)
return axiom_links
def eval_batch(self, batch, cross_entropy_loss):
batch_atoms = batch[0].to(self.device)
batch_polarity = batch[1].to(self.device)
batch_true_links = batch[2].to(self.device)
batch_sentences_tokens = batch[3].to(self.device)
batch_sentences_mask = batch[4].to(self.device)
logits, sentences_embedding = self.Supertagger.forward(batch_sentences_tokens, batch_sentences_mask)
logits_axiom_links_pred = self(batch_atoms, batch_polarity, sentences_embedding,
batch_sentences_mask)
axiom_links_pred = torch.argmax(logits_axiom_links_pred, dim=3)
accuracy = mesure_accuracy(batch_true_links, axiom_links_pred)
loss = cross_entropy_loss(logits_axiom_links_pred, batch_true_links)
return loss, accuracy
def eval_epoch(self, dataloader, cross_entropy_loss):
r"""Average the evaluation of all the batch.
Args:
dataloader: contains all the batch which contain the tokenized sentences, their masks and the true symbols
"""
self.eval()
accuracy_average = 0
loss_average = 0
with torch.no_grad():
for step, batch in enumerate(dataloader):
loss, accuracy = self.eval_batch(batch, cross_entropy_loss)
accuracy_average += accuracy
loss_average += float(loss)
return loss_average / len(dataloader), accuracy_average / len(dataloader)
def load_weights(self, model_file):
print("#" * 15)
try:
params = torch.load(model_file, map_location=self.device)
args = params['args']
self.atom_map = args['atom_map']
self.max_atoms_in_sentence = args['max_atoms_in_sentence']
self.atoms_tokenizer = AtomTokenizer(self.atom_map, self.max_atoms_in_sentence)
self.atoms_embedding.load_state_dict(params['atoms_embedding'])
self.linker_encoder.load_state_dict(params['linker_encoder'])
self.pos_transformation.load_state_dict(params['pos_transformation'])
self.neg_transformation.load_state_dict(params['neg_transformation'])
self.optimizer.load_state_dict(params['optimizer'])
print("\n The loading checkpoint was successful ! \n")
except Exception as e:
print("\n/!\ Can't load checkpoint model /!\ because :\n\n " + str(e), file=sys.stderr)
raise e
print("#" * 15)
def __checkpoint_save(self, path='/linker.pt'):
"""
@param path:
"""
self.cpu()
torch.save({
'args': dict(atom_map=self.atom_map, max_atoms_in_sentence=self.max_atoms_in_sentence),
'atoms_embedding': self.atoms_embedding.state_dict(),
'linker_encoder': self.linker_encoder.state_dict(),
'pos_transformation': self.pos_transformation.state_dict(),
'neg_transformation': self.neg_transformation.state_dict(),
'optimizer': self.optimizer,
}, path)
self.to(self.device)