def test_fastnlp_1min_tutorial(self):
# tutorials/fastnlp_1min_tutorial.ipynb
data_path = "test/data_for_tests/tutorial_sample_dataset.csv"
ds = DataSet.read_csv(data_path,
headers=('raw_sentence', 'label'),
sep='\t')
print(ds[1])
# 将所有数字转为小写
ds.apply(lambda x: x['raw_sentence'].lower(),
new_field_name='raw_sentence')
# label转int
ds.apply(lambda x: int(x['label']),
new_field_name='target',
is_target=True)
def split_sent(ins):
return ins['raw_sentence'].split()
ds.apply(split_sent, new_field_name='words', is_input=True)
# 分割训练集/验证集
train_data, dev_data = ds.split(0.3)
print("Train size: ", len(train_data))
print("Test size: ", len(dev_data))
from fastNLP import Vocabulary
vocab = Vocabulary(min_freq=2)
train_data.apply(lambda x: [vocab.add(word) for word in x['words']])
# index句子, Vocabulary.to_index(word)
train_data.apply(
lambda x: [vocab.to_index(word) for word in x['words']],
new_field_name='words',
is_input=True)
dev_data.apply(lambda x: [vocab.to_index(word) for word in x['words']],
new_field_name='words',
is_input=True)
from fastNLP.models import CNNText
model = CNNText((len(vocab), 50),
num_classes=5,
padding=2,
dropout=0.1)
from fastNLP import Trainer, CrossEntropyLoss, AccuracyMetric, Adam
trainer = Trainer(model=model,
train_data=train_data,
dev_data=dev_data,
loss=CrossEntropyLoss(),
optimizer=Adam(),
metrics=AccuracyMetric(target='target'))
trainer.train()
print('Train finished!')
def train(config):
train_data = pickle.load(open(os.path.join(config.data_path, config.train_name), "rb"))
dev_data = pickle.load(open(os.path.join(config.data_path, config.dev_name), "rb"))
test_data = pickle.load(open(os.path.join(config.data_path, config.test_name), "rb"))
vocabulary = pickle.load(open(os.path.join(config.data_path, config.vocabulary_name), "rb"))
# load w2v data
weight = pickle.load(open(os.path.join(config.data_path, config.weight_name), "rb"))
if config.task_name == "lstm":
text_model = LSTM(vocab_size=len(vocabulary), embed_dim=config.embed_dim,
output_dim=config.class_num, hidden_dim=config.hidden_dim,
num_layers=config.num_layers, dropout=config.dropout)
elif config.task_name == "lstm_maxpool":
text_model = LSTM_maxpool(vocab_size=len(vocabulary), embed_dim=config.embed_dim,
output_dim=config.class_num, hidden_dim=config.hidden_dim,
num_layers=config.num_layers, dropout=config.dropout)
elif config.task_name == "rnn":
text_model = RNN(vocab_size=len(vocabulary), embed_dim=config.embed_dim,
output_dim=config.class_num, hidden_dim=config.hidden_dim,
num_layers=config.num_layers, dropout=config.dropout)
elif config.task_name == "cnn":
text_model = CNN(vocab_size=len(vocabulary), embed_dim=config.embed_dim,
class_num=config.class_num, kernel_num=config.kernel_num,
kernel_sizes=config.kernel_sizes, dropout=config.dropout,
static=config.static, in_channels=config.in_channels)
elif config.task_name == "cnn_w2v":
text_model = CNN_w2v(vocab_size=len(vocabulary), embed_dim=config.embed_dim,
class_num=config.class_num, kernel_num=config.kernel_num,
kernel_sizes=config.kernel_sizes, dropout=config.dropout,
static=config.static, in_channels=config.in_channels,
weight=weight)
elif config.task_name == "rcnn":
text_model = RCNN(vocab_size=len(vocabulary), embed_dim=config.embed_dim,
output_dim=config.class_num, hidden_dim=config.hidden_dim,
num_layers=config.num_layers, dropout=config.dropout)
optimizer = Adam(lr=config.lr, weight_decay=config.weight_decay)
timing = TimingCallback()
early_stop = EarlyStopCallback(config.patience)
accuracy = AccuracyMetric(pred='output', target='target')
trainer = Trainer(train_data=train_data, model=text_model, loss=CrossEntropyLoss(),
batch_size=config.batch_size, check_code_level=0,
metrics=accuracy, n_epochs=config.epoch,
dev_data=dev_data, save_path=config.save_path,
print_every=config.print_every, validate_every=config.validate_every,
optimizer=optimizer, use_tqdm=False,
device=config.device, callbacks=[timing, early_stop])
trainer.train()
# test result
tester = Tester(test_data, text_model, metrics=accuracy)
tester.test()
def train_variational_ae():
n_epochs = 500
trainer = Trainer(train_data=fast_data.train_data,
dev_data=fast_data.test_data,
model=vae,
loss=LossFunc(VariationalAE.loss),
metrics=AccuracyMetric(target='dec_target'),
n_epochs=n_epochs,
batch_size=batch_size,
optimizer=Adam(lr=0.001,
weight_decay=0,
model_params=vae.parameters()))
trainer.train()
print('VAE Train finished!')
def train_classifier():
n_epochs = 50
trainer = Trainer(train_data=fast_data.train_data,
dev_data=fast_data.test_data,
model=disc,
loss=CrossEntropyLoss(target='label_seq'),
metrics=AccuracyMetric(target='label_seq'),
n_epochs=n_epochs,
batch_size=batch_size,
optimizer=Adam(lr=0.001,
weight_decay=0,
model_params=disc.parameters()))
trainer.train()
print('Disc Train finished!')
def train():
config = Config()
train_data, dev_data, vocabulary = get_dataset(config.data_path)
poetry_model = PoetryModel(vocabulary_size=len(vocabulary),
embedding_size=config.embedding_size,
hidden_size=config.hidden_size)
loss = Loss(pred='output', target='target')
perplexity = Perplexity(pred='output', target='target')
print("optimizer:", config.optimizer)
print("momentum:", config.momentum)
if config.optimizer == 'adam':
optimizer = Adam(lr=config.lr, weight_decay=config.weight_decay)
elif config.optimizer == 'sgd':
optimizer = SGD(lr=config.lr, momentum=config.momentum)
elif config.optimizer == 'adagrad':
optimizer = Adagrad(lr=config.lr, weight_decay=config.weight_decay)
elif config.optimizer == 'adadelta':
optimizer = Adadelta(lr=config.lr,
rho=config.rho,
eps=config.eps,
weight_decay=config.weight_decay)
timing = TimingCallback()
early_stop = EarlyStopCallback(config.patience)
trainer = Trainer(train_data=train_data,
model=poetry_model,
loss=loss,
metrics=perplexity,
n_epochs=config.epoch,
batch_size=config.batch_size,
print_every=config.print_every,
validate_every=config.validate_every,
dev_data=dev_data,
save_path=config.save_path,
optimizer=optimizer,
check_code_level=config.check_code_level,
metric_key="-PPL",
sampler=RandomSampler(),
prefetch=False,
use_tqdm=True,
device=config.device,
callbacks=[timing, early_stop])
trainer.train()
def train(epochs=10, lr=0.001):
global model
for i in range(epochs):
print('----------------- ', str(i + 1), ' ------------------')
trainer = Trainer(model=model,
train_data=train_set,
dev_data=test_set,
loss=CrossEntropyLoss(pred='output',
target='target'),
metrics=AccuracyMetric(pred='pred', target='target'),
optimizer=Adam(lr=lr),
save_path=None,
batch_size=1,
n_epochs=1)
trainer.train()
model.load_state_dict(copy.deepcopy(trainer.model.state_dict()))
# save('../model/cnn-' + str(kernel_sizes) + '-' + str(keep_proba) + '-' + str(i+1))
save('../model/lstm-' + str(input_dim) + '-' + str(hidden_dim) + '-' +
str(i + 1))
# training
device = 0 if torch.cuda.is_available() else 'cpu'
'''
EMBED_DIM = 100
model = CNNText((len(vocab),EMBED_DIM), num_classes=len(vocab_target), dropout=0.1)
metrics=AccuracyMetric()
loss = CrossEntropyLoss()
optimizer=optim.RMSprop(model.parameters(), lr=0.01, alpha=0.99, eps=1e-08, weight_decay=0, momentum=0, centered=False)
N_EPOCHS = 10
BATCH_SIZE = 16
trainer = Trainer(model=model, train_data=train_data, dev_data=dev_data, loss=loss, metrics=metrics,optimizer=optimizer,n_epochs=N_EPOCHS, batch_size=BATCH_SIZE, device=device)
trainer.train()
'''
embed = BertEmbedding(vocab, model_dir_or_name='en', include_cls_sep=True)
model = BertForSequenceClassification(embed, len(vocab_target))
trainer = Trainer(train_data,
model,
optimizer=Adam(model_params=model.parameters(), lr=2e-5),
loss=CrossEntropyLoss(),
device=device,
batch_size=8,
dev_data=dev_data,
metrics=AccuracyMetric(),
n_epochs=2,
print_every=1)
trainer.train()
saver = ModelSaver("save_model/bert2021.1.19.pkl")
saver.save_pytorch(model)
def test_Adam(self):
optim = Adam(model_params=torch.nn.Linear(10, 3).parameters())
self.assertTrue("lr" in optim.__dict__["settings"])
self.assertTrue("weight_decay" in optim.__dict__["settings"])
res = optim.construct_from_pytorch(torch.nn.Linear(10, 3).parameters())
self.assertTrue(isinstance(res, torch.optim.Adam))
optim = Adam(lr=0.001)
self.assertEqual(optim.__dict__["settings"]["lr"], 0.001)
res = optim.construct_from_pytorch(torch.nn.Linear(10, 3).parameters())
self.assertTrue(isinstance(res, torch.optim.Adam))
optim = Adam(lr=0.002, weight_decay=0.989)
self.assertEqual(optim.__dict__["settings"]["lr"], 0.002)
self.assertEqual(optim.__dict__["settings"]["weight_decay"], 0.989)
optim = Adam(0.001)
self.assertEqual(optim.__dict__["settings"]["lr"], 0.001)
res = optim.construct_from_pytorch(torch.nn.Linear(10, 3).parameters())
self.assertTrue(isinstance(res, torch.optim.Adam))
embeddings=embeddings,
dropout=0.5,
num_classes=3,
device=device).to(device)
trainer = Trainer(train_data=train_data,
model=model,
loss=CrossEntropyLoss(pred='pred', target='label'),
metrics=AccuracyMetric(),
n_epochs=10,
batch_size=32,
print_every=-1,
validate_every=-1,
dev_data=dev_data,
use_cuda=True,
optimizer=Adam(lr=0.0004, weight_decay=0),
check_code_level=-1,
metric_key='acc',
use_tqdm=False)
trainer.train()
# 训练结束后model为dev的最佳模型，保存
torch.save(model.state_dict(), '../data/checkpoints/best_model.pkl')
tester = Tester(
data=test_data,
model=model,
metrics=AccuracyMetric(),
batch_size=32,
)
tester.test()
self.p = []
def evaluate(self, target, pred):
self.l.extend(target.tolist())
pred = pred.argmax(dim=-1).tolist()
self.p.extend(pred)
def get_metric(self, reset=True):
f1_score = m.f1_score(self.l, self.p, average="macro")
if reset:
self.l = []
self.p = []
return {"f1:": f1_score}
optimizer = Adam(lr=args.lr, weight_decay=0)
acc = AccuracyMetric()
f1 = f1metric()
loss = CrossEntropyLoss()
trainer = Trainer(
trainset,
cla,
optimizer=optimizer,
loss=loss,
batch_size=args.batch_size,
n_epochs=args.num_epoch,
dev_data=testset,
metrics=[acc, f1],
save_path=args.save_dir,
callbacks=[FitlogCallback(log_loss_every=5)],
)
model = SLSTM(nemb=300, nhid=300, num_layer=1, Tar_emb=embedding)
trainer = Trainer(
train_data=train_dataset,
model=model,
loss=CrossEntropyLoss(pred='predict', target='label'),
metrics=AccuracyMetric(),
n_epochs=20,
batch_size=arg.batch_size,
print_every=1,
validate_every=-1,
dev_data=dev_dataset,
use_cuda=True,
save_path=save_dir,
optimizer=Adam(1e-3, weight_decay=0),
check_code_level=-1,
metric_key='acc',
# sampler=default,
use_tqdm=True,
)
results = trainer.train(load_best_model=True)
print(results)
torch.save(model, os.path.join(save_dir,"best_model.pkl"))
tester = Tester(
data=test_dataset,
model=model,
def test_fastnlp_advanced_tutorial(self):
import os
os.chdir("tutorials/fastnlp_advanced_tutorial")
from fastNLP import DataSet
from fastNLP import Instance
from fastNLP import Vocabulary
from fastNLP import Trainer
from fastNLP import Tester
# ### Instance
# Instance表示一个样本，由一个或者多个field（域、属性、特征）组成，每个field具有自己的名字以及值
# 在初始化Instance的时候可以定义它包含的field，使用"field_name=field_value"的写法
# In[2]:
# 组织一个Instance，这个Instance由premise、hypothesis、label三个field组成
instance = Instance(premise='an premise example .',
hypothesis='an hypothesis example.',
label=1)
instance
# In[3]:
data_set = DataSet([instance] * 5)
data_set.append(instance)
data_set[-2:]
# In[4]:
# 如果某一个field的类型与dataset对应的field类型不一样仍可被加入dataset中
instance2 = Instance(premise='the second premise example .',
hypothesis='the second hypothesis example.',
label='1')
try:
data_set.append(instance2)
except:
pass
data_set[-2:]
# In[5]:
# 如果某一个field的名字不对，则该instance不能被append到dataset中
instance3 = Instance(premises='the third premise example .',
hypothesis='the third hypothesis example.',
label=1)
try:
data_set.append(instance3)
except:
print('cannot append instance')
pass
data_set[-2:]
# In[6]:
# 除了文本以外，还可以将tensor作为其中一个field的value
import torch
tensor_ins = Instance(image=torch.randn(5, 5), label=0)
ds = DataSet()
ds.append(tensor_ins)
ds
from fastNLP import DataSet
from fastNLP import Instance
# 从csv读取数据到DataSet
# 类csv文件，即每一行为一个example的文件，都可以使用这种方法进行数据读取
dataset = DataSet.read_csv('tutorial_sample_dataset.csv',
headers=('raw_sentence', 'label'),
sep='\t')
# 查看DataSet的大小
len(dataset)
# In[8]:
# 使用数字索引[k]，获取第k个样本
dataset[0]
# In[9]:
# 获取的样本是一个Instance
type(dataset[0])
# In[10]:
# 使用数字索引[a: b]，获取第a到第b个样本
dataset[0:3]
# In[11]:
# 索引也可以是负数
dataset[-1]
data_path = ['premise', 'hypothesis', 'label']
# 读入文件
with open(data_path[0]) as f:
premise = f.readlines()
with open(data_path[1]) as f:
hypothesis = f.readlines()
with open(data_path[2]) as f:
label = f.readlines()
assert len(premise) == len(hypothesis) and len(hypothesis) == len(
label)
# 组织DataSet
data_set = DataSet()
for p, h, l in zip(premise, hypothesis, label):
p = p.strip() # 将行末空格去除
h = h.strip() # 将行末空格去除
data_set.append(Instance(premise=p, hypothesis=h, truth=l))
data_set[0]
# ### DataSet的其他操作
# 在构建完毕DataSet后，仍然可以对DataSet的内容进行操作，函数接口为DataSet.apply()
# In[13]:
# 将premise域的所有文本转成小写
data_set.apply(lambda x: x['premise'].lower(),
new_field_name='premise')
data_set[-2:]
# In[14]:
# label转int
data_set.apply(lambda x: int(x['truth']), new_field_name='truth')
data_set[-2:]
# In[15]:
# 使用空格分割句子
def split_sent(ins):
return ins['premise'].split()
data_set.apply(split_sent, new_field_name='premise')
data_set.apply(lambda x: x['hypothesis'].split(),
new_field_name='hypothesis')
data_set[-2:]
# In[16]:
# 筛选数据
origin_data_set_len = len(data_set)
data_set.drop(lambda x: len(x['premise']) <= 6)
origin_data_set_len, len(data_set)
# In[17]:
# 增加长度信息
data_set.apply(lambda x: [1] * len(x['premise']),
new_field_name='premise_len')
data_set.apply(lambda x: [1] * len(x['hypothesis']),
new_field_name='hypothesis_len')
data_set[-1]
# In[18]:
# 设定特征域、标签域
data_set.set_input("premise", "premise_len", "hypothesis",
"hypothesis_len")
data_set.set_target("truth")
# In[19]:
# 重命名field
data_set.rename_field('truth', 'label')
data_set[-1]
# In[20]:
# 切分训练、验证集、测试集
train_data, vad_data = data_set.split(0.5)
dev_data, test_data = vad_data.split(0.4)
len(train_data), len(dev_data), len(test_data)
# In[21]:
# 深拷贝一个数据集
import copy
train_data_2, dev_data_2 = copy.deepcopy(train_data), copy.deepcopy(
dev_data)
del copy
# 初始化词表，该词表最大的vocab_size为10000，词表中每个词出现的最低频率为2，'<unk>'表示未知词语，'<pad>'表示padding词语
# Vocabulary默认初始化参数为max_size=None, min_freq=None, unknown='<unk>', padding='<pad>'
vocab = Vocabulary(max_size=10000,
min_freq=2,
unknown='<unk>',
padding='<pad>')
# 构建词表
train_data.apply(lambda x: [vocab.add(word) for word in x['premise']])
train_data.apply(
lambda x: [vocab.add(word) for word in x['hypothesis']])
vocab.build_vocab()
# In[23]:
# 根据词表index句子
train_data.apply(
lambda x: [vocab.to_index(word) for word in x['premise']],
new_field_name='premise')
train_data.apply(
lambda x: [vocab.to_index(word) for word in x['hypothesis']],
new_field_name='hypothesis')
dev_data.apply(
lambda x: [vocab.to_index(word) for word in x['premise']],
new_field_name='premise')
dev_data.apply(
lambda x: [vocab.to_index(word) for word in x['hypothesis']],
new_field_name='hypothesis')
test_data.apply(
lambda x: [vocab.to_index(word) for word in x['premise']],
new_field_name='premise')
test_data.apply(
lambda x: [vocab.to_index(word) for word in x['hypothesis']],
new_field_name='hypothesis')
train_data[-1], dev_data[-1], test_data[-1]
# 读入vocab文件
with open('vocab.txt') as f:
lines = f.readlines()
vocabs = []
for line in lines:
vocabs.append(line.strip())
# 实例化Vocabulary
vocab_bert = Vocabulary(unknown=None, padding=None)
# 将vocabs列表加入Vocabulary
vocab_bert.add_word_lst(vocabs)
# 构建词表
vocab_bert.build_vocab()
# 更新unknown与padding的token文本
vocab_bert.unknown = '[UNK]'
vocab_bert.padding = '[PAD]'
# In[25]:
# 根据词表index句子
train_data_2.apply(
lambda x: [vocab_bert.to_index(word) for word in x['premise']],
new_field_name='premise')
train_data_2.apply(
lambda x: [vocab_bert.to_index(word) for word in x['hypothesis']],
new_field_name='hypothesis')
dev_data_2.apply(
lambda x: [vocab_bert.to_index(word) for word in x['premise']],
new_field_name='premise')
dev_data_2.apply(
lambda x: [vocab_bert.to_index(word) for word in x['hypothesis']],
new_field_name='hypothesis')
train_data_2[-1], dev_data_2[-1]
# step 1：加载模型参数（非必选）
from fastNLP.io.config_io import ConfigSection, ConfigLoader
args = ConfigSection()
ConfigLoader().load_config("./data/config", {"esim_model": args})
args["vocab_size"] = len(vocab)
args.data
# In[27]:
# step 2：加载ESIM模型
from fastNLP.models import ESIM
model = ESIM(**args.data)
model
# In[28]:
# 另一个例子：加载CNN文本分类模型
from fastNLP.models import CNNText
cnn_text_model = CNNText(embed_num=len(vocab),
embed_dim=50,
num_classes=5,
padding=2,
dropout=0.1)
cnn_text_model
from fastNLP import CrossEntropyLoss
from fastNLP import Adam
from fastNLP import AccuracyMetric
trainer = Trainer(
train_data=train_data,
model=model,
loss=CrossEntropyLoss(pred='pred', target='label'),
metrics=AccuracyMetric(),
n_epochs=3,
batch_size=16,
print_every=-1,
validate_every=-1,
dev_data=dev_data,
use_cuda=False,
optimizer=Adam(lr=1e-3, weight_decay=0),
check_code_level=-1,
metric_key='acc',
use_tqdm=False,
)
trainer.train()
tester = Tester(
data=test_data,
model=model,
metrics=AccuracyMetric(),
batch_size=args["batch_size"],
)
tester.test()
os.chdir("../..")
def train(config, task_name):
train_data = pickle.load(
open(os.path.join(config.data_path, config.train_name), "rb"))
# debug
if config.debug:
train_data = train_data[0:100]
dev_data = pickle.load(
open(os.path.join(config.data_path, config.dev_name), "rb"))
print(len(train_data), len(dev_data))
# test_data = pickle.load(open(os.path.join(config.data_path, config.test_name), "rb"))
# load w2v data
# weight = pickle.load(open(os.path.join(config.data_path, config.weight_name), "rb"))
word_vocab = pickle.load(
open(os.path.join(config.data_path, config.word_vocab_name), "rb"))
char_vocab = pickle.load(
open(os.path.join(config.data_path, config.char_vocab_name), "rb"))
pos_vocab = pickle.load(
open(os.path.join(config.data_path, config.pos_vocab_name), "rb"))
# spo_vocab = pickle.load(open(os.path.join(config.data_path, config.spo_vocab_name), "rb"))
tag_vocab = pickle.load(
open(os.path.join(config.data_path, config.tag_vocab_name), "rb"))
print('word vocab', len(word_vocab))
print('char vocab', len(char_vocab))
print('pos vocab', len(pos_vocab))
# print('spo vocab', len(spo_vocab))
print('tag vocab', len(tag_vocab))
schema = get_schemas(config.source_path)
if task_name == 'bilstm_crf':
model = AdvSeqLabel(
char_init_embed=(len(char_vocab), config.char_embed_dim),
word_init_embed=(len(word_vocab), config.word_embed_dim),
pos_init_embed=(len(pos_vocab), config.pos_embed_dim),
spo_embed_dim=len(schema),
sentence_length=config.sentence_length,
hidden_size=config.hidden_dim,
num_classes=len(tag_vocab),
dropout=config.dropout,
id2words=tag_vocab.idx2word,
encoding_type=config.encoding_type)
elif task_name == 'trans_crf':
model = TransformerSeqLabel(
char_init_embed=(len(char_vocab), config.char_embed_dim),
word_init_embed=(len(word_vocab), config.word_embed_dim),
pos_init_embed=(len(pos_vocab), config.pos_embed_dim),
spo_embed_dim=len(schema),
num_classes=len(tag_vocab),
id2words=tag_vocab.idx2word,
encoding_type=config.encoding_type,
num_layers=config.num_layers,
inner_size=config.inner_size,
key_size=config.key_size,
value_size=config.value_size,
num_head=config.num_head,
dropout=config.dropout)
optimizer = Adam(lr=config.lr, weight_decay=config.weight_decay)
timing = TimingCallback()
early_stop = EarlyStopCallback(config.patience)
# loss = NLLLoss()
logs = FitlogCallback(dev_data)
metrics = SpanFPreRecMetric(tag_vocab,
pred='pred',
seq_len='seq_len',
target='tag')
train_data.set_input('tag')
dev_data.set_input('tag')
dev_data.set_target('seq_len')
#print(train_data.get_field_names())
trainer = Trainer(
train_data=train_data,
model=model,
# loss=loss,
metrics=metrics,
metric_key='f',
batch_size=config.batch_size,
n_epochs=config.epoch,
dev_data=dev_data,
save_path=config.save_path,
check_code_level=-1,
print_every=config.print_every,
validate_every=config.validate_every,
optimizer=optimizer,
use_tqdm=False,
device=config.device,
callbacks=[timing, early_stop, logs])
trainer.train()
# test result
tester = Tester(dev_data,
model,
metrics=metrics,
device=config.device,
batch_size=config.batch_size)
tester.test()
def train(config, task_name):
train_data = pickle.load(
open(os.path.join(config.data_path, config.train_name), "rb"))
# debug
if config.debug:
train_data = train_data[0:30]
dev_data = pickle.load(
open(os.path.join(config.data_path, config.dev_name), "rb"))
# test_data = pickle.load(open(os.path.join(config.data_path, config.test_name), "rb"))
vocabulary = pickle.load(
open(os.path.join(config.data_path, config.vocabulary_name), "rb"))
# load w2v data
# weight = pickle.load(open(os.path.join(config.data_path, config.weight_name), "rb"))
if task_name == "lstm":
text_model = LSTM(vocab_size=len(vocabulary),
embed_dim=config.embed_dim,
output_dim=config.class_num,
hidden_dim=config.hidden_dim,
num_layers=config.num_layers,
dropout=config.dropout)
elif task_name == "lstm_maxpool":
text_model = LSTM_maxpool(vocab_size=len(vocabulary),
embed_dim=config.embed_dim,
output_dim=config.class_num,
hidden_dim=config.hidden_dim,
num_layers=config.num_layers,
dropout=config.dropout)
elif task_name == "cnn":
text_model = CNN(vocab_size=len(vocabulary),
embed_dim=config.embed_dim,
class_num=config.class_num,
kernel_num=config.kernel_num,
kernel_sizes=config.kernel_sizes,
dropout=config.dropout,
static=config.static,
in_channels=config.in_channels)
elif task_name == "rnn":
text_model = RNN(vocab_size=len(vocabulary),
embed_dim=config.embed_dim,
output_dim=config.class_num,
hidden_dim=config.hidden_dim,
num_layers=config.num_layers,
dropout=config.dropout)
# elif task_name == "cnn_w2v":
# text_model = CNN_w2v(vocab_size=len(vocabulary), embed_dim=config.embed_dim,
# class_num=config.class_num, kernel_num=config.kernel_num,
# kernel_sizes=config.kernel_sizes, dropout=config.dropout,
# static=config.static, in_channels=config.in_channels,
# weight=weight)
elif task_name == "rcnn":
text_model = RCNN(vocab_size=len(vocabulary),
embed_dim=config.embed_dim,
output_dim=config.class_num,
hidden_dim=config.hidden_dim,
num_layers=config.num_layers,
dropout=config.dropout)
#elif task_name == "bert":
# text_model = BertModel.from_pretrained(config.bert_path)
optimizer = Adam(lr=config.lr, weight_decay=config.weight_decay)
timing = TimingCallback()
early_stop = EarlyStopCallback(config.patience)
logs = FitlogCallback(dev_data)
f1 = F1_score(pred='output', target='target')
trainer = Trainer(train_data=train_data,
model=text_model,
loss=BCEWithLogitsLoss(),
batch_size=config.batch_size,
check_code_level=-1,
metrics=f1,
metric_key='f1',
n_epochs=config.epoch,
dev_data=dev_data,
save_path=config.save_path,
print_every=config.print_every,
validate_every=config.validate_every,
optimizer=optimizer,
use_tqdm=False,
device=config.device,
callbacks=[timing, early_stop, logs])
trainer.train()
# test result
tester = Tester(
dev_data,
text_model,
metrics=f1,
device=config.device,
batch_size=config.batch_size,
)
tester.test()
print("Total Number of Words:", m)
rnn_text_model = RNN.RNN_Text(vocab_size=m, input_size=50, hidden_layer_size=128, target_size=k, dropout=0.1)
cnn_text_model = CNN.CNN_Text(vocab_size=m, input_size=50, target_size=k, dropout=0.05)
model = rnn_text_model
# ModelLoader.load_pytorch(model, "model_ckpt_large_CNN.pkl")
trainer = Trainer(
train_data=train_set,
model=model,
loss=CrossEntropyLoss(pred='pred', target='label'),
n_epochs=50,
batch_size=16,
metrics=AccuracyMetric(pred='pred', target='label'),
dev_data=dev_set,
optimizer=Adam(lr=1e-3),
callbacks=[FitlogCallback(data=test_set)]
)
trainer.train()
# saver = ModelSaver("model_ckpt_large_RNN.pkl")
# saver.save_pytorch(model)
tester = Tester(
data=train_set,
model=model,
metrics=AccuracyMetric(pred='pred', target='label'),
batch_size=16,
)
tester.test()
embed_dim=128,
hidden_dim=hidden_units,
output_dim=8)
# mymodel = load_model(mymodel, './model/best_TextModel_acc_2019-06-28-09-07-50')
trainer = Trainer(
train_data=train_data,
model=mymodel,
loss=CrossEntropyLoss(pred='pred', target='target'),
# loss=SkipBudgetLoss(pred='pred', target='target', updated_states='updated_states'),
metrics=[AccuracyMetric(), UsedStepsMetric()],
n_epochs=30,
batch_size=batch_size,
print_every=-1,
validate_every=-1,
dev_data=test_data,
save_path='./model',
optimizer=Adam(lr=learning_rate, weight_decay=0),
check_code_level=0,
device="cuda",
metric_key='acc',
use_tqdm=False,
callbacks=[FitlogCallback(test_data)])
start = time.clock()
trainer.train()
end = time.clock()
training_time = end - start
print('total training time:%fs' % (end - start))
fitlog.add_hyper({'time': training_time})
fitlog.finish()