parser.add_argument('--model-dir', type=str, default=os.environ['SM_MODEL_DIR'])
parser.add_argument('--data-dir', type=str, default=os.environ['SM_CHANNEL_TRAINING'])
parser.add_argument('--num-gpus', type=int, default=os.environ['SM_NUM_GPUS'])
args = parser.parse_args()
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
print("Using device {}.".format(device))
torch.manual_seed(args.seed)
# Load the training data.
train_loader = _get_train_data_loader(args.batch_size, args.data_dir)
# Build the model.
model = LSTMClassifier(args.embedding_dim, args.hidden_dim, args.vocab_size).to(device)
with open(os.path.join(args.data_dir, "word_dict.pkl"), "rb") as f:
model.word_dict = pickle.load(f)
print("Model loaded with embedding_dim {}, hidden_dim {}, vocab_size {}.".format(
args.embedding_dim, args.hidden_dim, args.vocab_size
))
# Train the model.
optimizer = optim.Adam(model.parameters())
loss_fn = torch.nn.BCELoss()
train(model, train_loader, args.epochs, optimizer, loss_fn, device)
# Save the parameters used to construct the model
parser.add_argument('--num-gpus',
type=int,
default=os.environ['SM_NUM_GPUS'])
args = parser.parse_args()
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
print("Using device {}.".format(device))
torch.manual_seed(args.seed)
# Load the training data.
train_loader = _get_train_data_loader(args.batch_size, args.data_dir)
# Build the model.
model = LSTMClassifier(args.embedding_dim, args.hidden_dim,
args.vocab_size).to(device)
with open(os.path.join(args.data_dir, "word_dict.pkl"), "rb") as f:
model.word_dict = pickle.load(f)
print("Model loaded with embedding_dim {}, hidden_dim {}, vocab_size {}.".
format(args.embedding_dim, args.hidden_dim, args.vocab_size))
# Train the model.
optimizer = optim.Adam(model.parameters(), lr=0.001)
loss_fn = torch.nn.BCELoss()
print(model)
print(len(list(model.parameters())))
for i in range(len(list(model.parameters()))):
print(list(model.parameters())[i].size())
import torch.nn as nn
import pandas as pd
# 元データを7:3に分ける
train_data, test_data = train_test_split(datasets, train_size=0.7)
# 単語のベクトル次元数
EMBEDDING_DIM = 10
# 隠れ層の次元数
HIDDEN_DIM = 128
# データ全体の単語数
VOCAB_SIZE = len(word2index)
# 分類先のカテゴリの数
TAG_SIZE = len(categories)
# モデル宣言
model = LSTMClassifier(EMBEDDING_DIM, HIDDEN_DIM, VOCAB_SIZE, TAG_SIZE)
# 損失関数はNLLLoss()。LogSoftmaxにはこれを使うことが多い。
loss_function = nn.NLLLoss()
# 最適化はSGD。lossの減りに多少時間がかかる。
optimizer = optim.SGD(model.parameters(), lr=0.01)
# 各エポックの合計loss値を格納
losses = []
for epoch in range(100):
all_loss = 0
for title, cat in zip(train_data["title"], train_data["category"]):
# モデルが持っている勾配の情報をリセット
model.zero_grad()
# 文章を単語IDの系列に変換（modelが読み込めるように）
inputs = sentence2index(title)
def main():
data, labels = read_imdb_data()
train_X, test_X, train_y, test_y = prepare_imdb_data(data, labels)
#storing the preprocess data as cache
cache_dir = os.path.join(
"cache", "sentiment_analysis") # where to store cache files
os.makedirs(cache_dir, exist_ok=True) # ensure cache directory exists
# Preprocess data
train_X, test_X, train_y, test_y = preprocess_data(train_X, test_X,
train_y, test_y,
cache_dir)
#building word dict from reviews
word_dict = build_dict(train_X)
#now we store word dict for future references
data_dir = 'data/pytorch' # The folder we will use for storing data
if not os.path.exists(data_dir): # Make sure that the folder exists
os.makedirs(data_dir)
with open(os.path.join(data_dir, 'word_dict.pkl'), "wb") as f:
pickle.dump(word_dict, f)
train_X, train_X_len = convert_and_pad_data(word_dict, train_X)
test_X, test_X_len = convert_and_pad_data(word_dict, test_X)
#store processed data
pd.concat([pd.DataFrame(train_y), pd.DataFrame(train_X_len), pd.DataFrame(train_X)], axis=1) \
.to_csv(os.path.join(data_dir, 'train.csv'), header=False, index=False)
loadEnv()
# Accessing variables.
access_key_id = os.getenv('ACCESS_KEY_ID')
secret_key = os.getenv('SECRET_KEY')
region = os.getenv('AWS_REGION')
execution_role = os.getenv('EXEC_ROLE')
# create sagemaker session
session = boto3.Session(aws_access_key_id=access_key_id,
aws_secret_access_key=secret_key,
region_name=region)
sagemaker_session = sagemaker.Session(boto_session=session)
#update data to s3 bucket
bucket = sagemaker_session.default_bucket()
prefix = 'sagemaker/sentiment_rnn'
role = execution_role
input_data = sagemaker_session.upload_data(path=data_dir,
bucket=bucket,
key_prefix=prefix)
# Read in only the first 250 rows
train_sample = pd.read_csv(os.path.join(data_dir, 'train.csv'),
header=None,
names=None,
nrows=250)
# Turn the input pandas dataframe into tensors
train_sample_y = torch.from_numpy(
train_sample[[0]].values).float().squeeze()
train_sample_X = torch.from_numpy(train_sample.drop([0],
axis=1).values).long()
# Build the dataset
train_sample_ds = torch.utils.data.TensorDataset(train_sample_X,
train_sample_y)
# Build the dataloader
train_sample_dl = torch.utils.data.DataLoader(train_sample_ds,
batch_size=50)
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
lstm_model = LSTMClassifier(32, 100, 5000).to(device)
optimizer = optim.Adam(lstm_model.parameters())
loss_fn = torch.nn.BCELoss()
train(lstm_model, train_sample_dl, 5, optimizer, loss_fn, device)
estimator = PyTorch(entry_point="train.py",
source_dir="train",
role=role,
framework_version='0.4.0',
train_instance_count=1,
train_instance_type='ml.m4.xlarge',
hyperparameters={
'epochs': 10,
'hidden_dim': 200,
})
estimator.fit({'training': input_data})
# Deploy the trained model
class StringPredictor(RealTimePredictor):
def __init__(self, endpoint_name, sagemaker_session):
super(StringPredictor, self).__init__(endpoint_name,
sagemaker_session,
content_type='text/plain')
py_model = PyTorchModel(model_data=estimator.model_data,
role=role,
framework_version='0.4.0',
entry_point='predict.py',
source_dir='serve',
predictor_cls=StringPredictor)
pytorch_predictor = py_model.deploy(initial_instance_count=1,
instance_type='ml.m4.xlarge')
print(pytorch_predictor.endpoint)
return
def main():
parser = argparse.ArgumentParser(
description='PyTorch Gambler\'s Loss Runner')
parser.add_argument('--result_dir',
type=str,
help='directory to save result txt files',
default='results')
parser.add_argument('--noise_rate',
type=float,
help='corruption rate, should be less than 1',
default=0.5)
parser.add_argument('--noise_type',
type=str,
help='[pairflip, symmetric]',
default='symmetric')
parser.add_argument('--dataset',
type=str,
help='mnist, cifar10, or imdb',
default='mnist')
parser.add_argument('--n_epoch', type=int, default=10)
parser.add_argument('--seed', type=int, default=1)
parser.add_argument('--num_workers',
type=int,
default=4,
help='how many subprocesses to use for data loading')
parser.add_argument('--epoch_decay_start', type=int, default=80)
parser.add_argument('--load_model', type=str, default="")
parser.add_argument('--model', type=str, default='default')
parser.add_argument('--batch_size',
type=int,
default=128,
metavar='N',
help='input batch size for training (default: 128)')
parser.add_argument(
'--log-interval',
type=int,
default=100,
metavar='N',
help=
'how many batches to wait before logging training status (default: 100)'
)
parser.add_argument('--lr',
type=float,
default=0.001,
metavar='LR',
help='learning rate (default: 0.001)')
parser.add_argument('--eps',
type=float,
help='set lambda for lambda type \'gmblers\' only',
default=1000.0)
parser.add_argument('--lambda_type',
type=str,
help='[nll, euc, mid, exp, gmblers]',
default="euc")
parser.add_argument('--start_gamblers',
type=int,
help='number of epochs before starting gamblers',
default=0)
# label smoothing args
parser.add_argument('--smoothing',
type=float,
default=1.0,
help='smoothing parameter (default: 1)')
args = parser.parse_args()
args.use_scheduler = False
torch.manual_seed(args.seed)
torch.cuda.manual_seed(args.seed)
if args.dataset == 'mnist':
input_channel = 1
num_classes = 10
train_dataset = MNIST(root='./data/',
download=True,
train=True,
transform=transforms.ToTensor(),
noise_type=args.noise_type,
noise_rate=args.noise_rate)
test_dataset = MNIST(root='./data/',
download=True,
train=False,
transform=transforms.ToTensor(),
noise_type=args.noise_type,
noise_rate=args.noise_rate)
print('loading dataset...')
train_loader = torch.utils.data.DataLoader(
dataset=train_dataset,
batch_size=args.batch_size,
num_workers=args.num_workers,
drop_last=True,
shuffle=True)
test_loader = torch.utils.data.DataLoader(dataset=test_dataset,
batch_size=args.batch_size,
num_workers=args.num_workers,
drop_last=True,
shuffle=False)
if args.dataset == 'cifar10':
input_channel = 3
num_classes = 10
transform_train = transforms.Compose([
transforms.RandomCrop(32, padding=4),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize((0.4914, 0.4822, 0.4465),
(0.2023, 0.1994, 0.2010)),
])
transform_test = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.4914, 0.4822, 0.4465),
(0.2023, 0.1994, 0.2010)),
])
train_dataset = CIFAR10(root='./data/',
download=True,
train=True,
transform=transform_train,
noise_type=args.noise_type,
noise_rate=args.noise_rate)
test_dataset = CIFAR10(root='./data/',
download=True,
train=False,
transform=transform_test,
noise_type=args.noise_type,
noise_rate=args.noise_rate)
print('loading dataset...')
train_loader = torch.utils.data.DataLoader(
dataset=train_dataset,
batch_size=args.batch_size,
num_workers=args.num_workers,
drop_last=True,
shuffle=True)
test_loader = torch.utils.data.DataLoader(dataset=test_dataset,
batch_size=args.batch_size,
num_workers=args.num_workers,
drop_last=True,
shuffle=False)
if args.dataset == 'cifar100':
input_channel = 3
num_classes = 100
transform_train = transforms.Compose([
transforms.RandomCrop(32, padding=4),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize((0.4914, 0.4822, 0.4465),
(0.2023, 0.1994, 0.2010)),
])
transform_test = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.4914, 0.4822, 0.4465),
(0.2023, 0.1994, 0.2010)),
])
train_dataset = CIFAR100(root='./data/',
download=True,
train=True,
transform=transform_train,
noise_type=args.noise_type,
noise_rate=args.noise_rate)
test_dataset = CIFAR100(root='./data/',
download=True,
train=False,
transform=transform_test,
noise_type=args.noise_type,
noise_rate=args.noise_rate)
print('loading dataset...')
train_loader = torch.utils.data.DataLoader(
dataset=train_dataset,
batch_size=args.batch_size,
num_workers=args.num_workers,
drop_last=True,
shuffle=True)
test_loader = torch.utils.data.DataLoader(dataset=test_dataset,
batch_size=args.batch_size,
num_workers=args.num_workers,
drop_last=True,
shuffle=False)
if args.dataset == 'imdb':
num_classes = 2
embedding_length = 300
hidden_size = 256
print('loading dataset...')
TEXT, vocab_size, word_embeddings, train_loader, valid_iter, test_loader = load_data.load_dataset(
rate=args.noise_rate, batch_size=args.batch_size)
use_cuda = torch.cuda.is_available()
device = torch.device("cuda" if use_cuda else "cpu")
print("using {}".format(device))
print('building model...')
if args.dataset == 'mnist':
model = CNN_basic(num_classes=num_classes).to(device)
optimizer = torch.optim.SGD(model.parameters(), lr=args.lr)
if args.dataset == 'cifar10':
if args.model == 'small':
model = CNN_small(num_classes=num_classes).to(device)
optimizer = torch.optim.SGD(model.parameters(), lr=args.lr)
else:
model = resnet.ResNet18(num_classes=num_classes).to(device)
change_lr = lambda epoch: 0.1 if epoch >= 50 else 1.0
optimizer = LaProp(filter(lambda p: p.requires_grad,
model.parameters()),
lr=4e-4)
scheduler = torch.optim.lr_scheduler.LambdaLR(optimizer,
lr_lambda=change_lr)
args.use_scheduler = True
if args.dataset == 'cifar100':
if args.model == 'small':
model = CNN_small(num_classes=num_classes).to(device)
optimizer = torch.optim.SGD(model.parameters(), lr=args.lr)
else:
model = resnet.ResNet18(num_classes=num_classes).to(device)
change_lr = lambda epoch: 0.1 if epoch >= 50 else 1.0
optimizer = LaProp(filter(lambda p: p.requires_grad,
model.parameters()),
lr=4e-4)
scheduler = torch.optim.lr_scheduler.LambdaLR(optimizer,
lr_lambda=change_lr)
args.use_scheduler = True
if args.dataset == 'imdb':
model = LSTMClassifier(args.batch_size, num_classes, hidden_size,
vocab_size, embedding_length,
word_embeddings).to(device)
optimizer = LaProp(filter(lambda p: p.requires_grad,
model.parameters()),
lr=args.lr)
test_accs = []
train_losses = []
test_losses = []
out = []
name = "{}_{}_{:.2f}_{:.2f}_{}_{}".format(args.dataset, args.noise_type,
args.smoothing, args.noise_rate,
args.eps, args.seed)
if not os.path.exists(args.result_dir):
os.system('mkdir -p %s' % args.result_dir)
save_file = args.result_dir + "/" + name + ".json"
if os.path.exists(save_file):
print('case processed')
exit()
for epoch in range(1, args.n_epoch + 1):
for param_group in optimizer.param_groups:
print(epoch, param_group['lr'])
print(name)
train_loss = train(args,
model,
device,
train_loader,
optimizer,
epoch,
num_classes=num_classes,
use_gamblers=(epoch >= args.start_gamblers),
text=(args.dataset == 'imdb'))
train_losses.append(train_loss)
test_acc, test_loss = test(args,
model,
device,
test_loader,
num_classes,
text=(args.dataset == 'imdb'))
test_accs.append(test_acc)
test_losses.append(test_loss)
if (args.use_scheduler):
scheduler.step()
# torch.save({
# 'model_state_dict': model.state_dict(),
# 'optimizer_state_dict': optimizer.state_dict(),
# 'loss': loss,
# 'test_acc': acc
# }, args.result_dir + "/" + name + "_model.npy")
save_data = {
"train_loss": train_losses,
"test_loss": test_losses,
"test_acc": test_accs
}
json.dump(save_data, open(save_file, 'w'))
#!/usr/bin/env python
# coding: utf-8
from prepare_data import PrepareData
from model import LSTMClassifier
import pandas as pd
from sklearn.model_selection import train_test_split
ppd = PrepareData()
data = ppd.get_data()
lstm = LSTMClassifier()
X = lstm.get_matrix(data)
Y = pd.get_dummies(data['label']).values
X_train, X_test, Y_train, Y_test = train_test_split(X,
Y,
test_size=0.33,
random_state=42)
print(X_train.shape, Y_train.shape)
print(X_test.shape, Y_test.shape)
model = lstm.get_model(X.shape[1])
history = lstm.fit_model(model, X_train, Y_train)
validation_size = 1500
X_validate = X_test[-validation_size:]
Y_validate = Y_test[-validation_size:]
X_test = X_test[:-validation_size]
Y_test = Y_test[:-validation_size]
score, acc = model.evaluate(X_test, Y_test, verbose=2, batch_size=batch_size)
shuffle=True,
num_workers=0)
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
#define hyperparameters
size_of_vocab = len(all_words)
embedding_dim = 20
num_hidden_nodes = 8
num_output_nodes = len(tags)
num_layers = 2
bidirection = True
dropout = 0.2
#instantiate the model
model = LSTMClassifier(12, 20, len(all_words), len(tags)).to(device)
#architecture
print(model)
#No. of trianable parameters
# def count_parameters(model):
# return sum(p.numel() for p in model.parameters() if p.requires_grad)
# print(f'The model has {count_parameters(model):,} trainable parameters')
# Loss and optimizer
criterion = nn.BCELoss()
optimizer = torch.optim.AdamW(model.parameters(), lr=learning_rate)
# Train the model
#Load data generators
train_data_loader = get_train_loader(cities=cities,
labels=labels,
batch_size=batch_size,
shuffle=False,
collate_fn=collate_fn,
sampler=train_sampler)
valid_data_loader = get_train_loader(cities=cities,
labels=labels,
batch_size=batch_size,
shuffle=False,
collate_fn=collate_fn,
sampler=valid_sampler)
#Initialize the model to train
model = LSTMClassifier(27, 10, 14)
# Loss and Optimizer
criterion = nn.NLLLoss()
learning_rate = 0.8
optimizer = torch.optim.SGD(model.parameters(), lr=learning_rate)
# train
losses = []
num_epochs = 10
# Train the Model
for epoch in range(num_epochs):
print("##### epoch {:2d}".format(epoch + 1))
for i, batch in enumerate(train_data_loader):
def main():
os.chdir('./')
global args, word2vec, batch_size, train_set_idx
global weight_scale, phenotypedictinverse
phenotypedict = dict({
"Cancer": 11,
"Heart": 4,
"Lung": 5,
"Neuro": 10,
"Pain": 9,
"Alcohol": 7,
"Substance": 8,
"Obesity": 1,
"Disorders": 6,
"Depression": 12
})
parser = argparse.ArgumentParser(
description=__doc__,
formatter_class=argparse.RawDescriptionHelpFormatter)
#parser.add_argument('clean_summaries0209.csv', help="Source Input file", type=str)
#parser.add_argument('word2vec_50d.txt', help="word2vec file", type=str)
parser.add_argument('--padding',
help="padding around each text",
type=int,
default=4)
parser.add_argument(
'--max_note_len',
help="Cut off all notes longer than this (0 = no cutoff).",
type=int,
default=0)
parser.add_argument('--filename',
help="File name for output file",
type=str,
default="data.h5")
parser.add_argument('-predict_label',
type=int,
default=phenotypedict["Depression"],
help='Choose which type of phenotyping to detect')
parser.add_argument('-topred',
type=str,
default="Depression",
help='Choose which type of phenotyping to detect')
parser.add_argument('-epochs',
type=int,
default=10,
help='number of epochs for train [default: 10]')
parser.add_argument('-batch_size',
type=int,
default=8,
help='batch size for training [default: 64]')
parser.add_argument('-output_size',
type=int,
default=2,
help='final output dim [default: 2]')
parser.add_argument('-hidden_size',
type=int,
default=256,
help='output dim of the cell [default: 256]')
parser.add_argument('-embedding_length',
type=int,
default=50,
help='number of embedding dimension [default: 50]')
parser.add_argument('-learning_rate',
type=float,
default=0.005,
help='initial learning rate [default: 0.5]')
parser.add_argument('-vocab_size',
type=float,
default=48849,
help='initial learning rate [default: 0.5]')
parser.add_argument(
'-optimizer',
type=str,
default='Adam',
help='optimizer for the gradient descent: Adadelta, Adam')
parser.add_argument('-cuda', type=int, default=-1, help='CUUUUUUUUUUUUDA')
parser.add_argument('-debug',
type=int,
default=0,
help='debug mode to print')
parser.add_argument('-l2s', type=float, default=3, help='l2 norm')
# with open("conditions.dict", 'w') as f:
# for i, c in enumerate(conditions):
# print (f, i + 1, c)
args = parser.parse_args()
phenotypedictinverse = dict({
11: "Cancer",
4: "Heart",
5: "Lung",
10: "Neuro",
9: "Pain",
7: "Alcohol",
8: "Substance",
1: "Obesity",
6: "Disorders",
12: "Depression"
})
phenotypedictsamples = dict({
"Cancer": 161,
"Heart": 275,
"Lung": 167,
"Neuro": 368,
"Pain": 321,
"Alcohol": 196,
"Substance": 155,
"Obesity": 126,
"Disorders": 295,
"Depression": 460
})
weight_scale = [
1 / (1610 -
phenotypedictsamples[phenotypedictinverse[args.predict_label]]),
1 / phenotypedictsamples[phenotypedictinverse[args.predict_label]]
]
#weight_scale = [ phenotypedictsamples[phenotypedictinverse[args.predict_label]]/1610*10, (1610 - phenotypedictsamples[phenotypedictinverse[args.predict_label]])/1610*10]
if args.cuda > -1:
weight_scale = torch.FloatTensor(weight_scale).cuda()
print('Weight Scale is: ', weight_scale)
# LOAD THE WORD2VEC FILE
word2vec, emb_size, v_large = load_bin_vec(
"word2vec_50d.txt") # word2vec whole dataset（label+unlabeled） 470260
print('WORD2VEC POINTS:', v_large)
# first step
# lbl, targets, ids, subj, time, embed = preprocess(args, emb_size, word2vec)
# lbl_train, lbl_train_target, lbl_test, lbl_test_target, phenotypedict = cross_validation(lbl, targets, ids, subj, time, args.topred, phenotypedict, phenotypedictsamples)
fold = 1
# put data of each fold in to a .h5py file
'''
for i in range(0,fold):
with h5py.File('data_biased_'+args.topred+'_cv{0}_occ'.format(i+1) + '0'+'.h5',"w") as f:
xtrain = np.array(lbl_train[i], dtype=int)
xtraintarget = np.array(lbl_train_target[i], dtype=int)
xtest = np.array(lbl_test[i], dtype=int)
xtesttarget = np.array(lbl_test_target[i], dtype=int)
f["w2v"] = np.array(embed)
f['train'] = xtrain
f['train_label'] = xtraintarget[:,phenotypedict[args.topred]]
f['test'] = xtest
f['test_label'] = xtesttarget[:,phenotypedict[args.topred]]
'''
if args.cuda > -1:
torch.cuda.set_device(args.cuda)
torch.backends.cudnn.benchmark = True
for i in range(0, fold):
train, test, y_test, w2v = readh5todata(
args, 'data_biased_' + phenotypedictinverse[args.predict_label] +
'_cv{0}'.format(i + 1) + '_occ' + '0' + '.h5')
args.w2v = w2v
train_loader = torch.utils.data.DataLoader(train,
batch_size=args.batch_size,
sampler=None,
shuffle=False)
test_loader = torch.utils.data.DataLoader(test,
batch_size=args.batch_size,
sampler=None,
shuffle=False)
LSTM = LSTMClassifier(args)
print(LSTM)
train_model(args, LSTM, args.learning_rate, args.batch_size,
args.epochs, train_loader)
class Trainer:
def __init__(self,
config,
n_gpu,
vocab,
train_loader=None,
val_loader=None):
self.config = config
self.vocab = vocab
self.n_gpu = n_gpu
self.train_loader = train_loader
self.val_loader = val_loader
# Build model
vocab_size = self.vocab.vocab_size()
self.model = LSTMClassifier(self.config, vocab_size,
self.config.n_label)
self.model.to(device)
if self.n_gpu > 1:
self.model = nn.DataParallel(self.model)
# Build optimizer
self.optimizer = optim.Adam(self.model.parameters(), lr=self.config.lr)
# Build criterion
self.criterion = nn.CrossEntropyLoss()
def train(self):
best_f1 = 0.0
best_acc = 0.0
global_step = 0
batch_f1 = []
batch_acc = []
for epoch in range(self.config.num_epoch):
batch_loss = []
for step, batch in enumerate(self.train_loader):
self.model.train()
batch = tuple(t.to(device) for t in batch)
batch = sort_batch(batch)
input_ids, input_lengths, labels = batch
outputs = self.model(input_ids, input_lengths)
loss = self.criterion(
outputs['logits'].view(-1, self.config.n_label),
labels.view(-1))
f1, acc = ic_metric(labels.cpu(),
outputs['predicted_intents'].cpu())
if self.n_gpu > 1:
loss = loss.mean()
loss.backward()
self.optimizer.step()
self.optimizer.zero_grad()
global_step += 1
batch_loss.append(loss.float().item())
batch_f1.append(f1)
batch_acc.append(acc)
if (global_step
== 1) or (global_step % self.config.log_interval == 0):
mean_loss = np.mean(batch_loss)
mean_f1 = np.mean(batch_f1)
mean_acc = np.mean(batch_acc)
batch_loss = []
nsml.report(summary=True,
scope=locals(),
epoch=epoch,
train_loss=mean_loss,
step=global_step)
if (global_step > 0) and (global_step %
self.config.val_interval == 0):
val_loss, val_f1, val_acc = self.evaluation()
nsml.report(summary=True,
scope=locals(),
epoch=epoch,
val_loss=val_loss,
val_f1=val_f1,
val_acc=val_acc,
step=global_step)
if val_f1 > best_f1:
best_f1 = val_f1
best_acc = val_acc
nsml.save(global_step)
def evaluation(self):
self.model.eval()
total_loss = []
preds = []
targets = []
with torch.no_grad():
for step, batch in enumerate(self.val_loader):
batch = tuple(t.to(device) for t in batch)
batch = sort_batch(batch)
input_ids, input_lengths, labels = batch
outputs = self.model(input_ids, input_lengths)
loss = self.criterion(
outputs['logits'].view(-1, self.config.n_label),
labels.view(-1))
pred = outputs['predicted_intents'].squeeze(
-1).cpu().numpy().tolist()
target = labels.cpu().numpy().tolist()
preds.extend(pred)
targets.extend(target)
total_loss.append(loss.float().item())
mean_loss = np.mean(total_loss)
mean_f1, mean_acc = ic_metric(targets, preds)
return mean_loss, mean_f1, mean_acc
def main():
parser = argparse.ArgumentParser("Script to train model on a GPU")
parser.add_argument(
"--checkpoint",
type=str,
default=None,
help=
"Optional path to saved model, if none provided, the model is trained from scratch."
)
parser.add_argument("--n_epochs",
type=int,
default=5,
help="Number of training epochs.")
args = parser.parse_args()
sampling_rate = 125
n_velocity_bins = 32
seq_length = 1024
n_tokens = 256 + sampling_rate + n_velocity_bins
#early_stopping = 100000 # very high value to basically turn it off
early_stopping = 200 # regular value
# transformer = MusicTransformer(n_tokens, seq_length,
# d_model = 64, n_heads = 8, d_feedforward=256,
# depth = 4, positional_encoding=True, relative_pos=True, xavier_init=True)
# set xavier_init = True to run xavier_init optimization
# transformer = LongMusicTransformer(n_tokens, seq_length,
# d_model=64, n_heads=8, d_feedforward=256,
# depth=4, positional_encoding=True, relative_pos=False,
# xavier_init=True)
transformer = LSTMClassifier(input_dim=1,
hidden_dim=413,
label_size=413,
n_tokens=n_tokens,
xavier_init=True)
if args.checkpoint is not None:
state = torch.load(args.checkpoint)
transformer.load_state_dict(state)
print(f"Successfully loaded checkpoint at {args.checkpoint}")
#rule of thumb: 1 minute is roughly 2k tokens
pipeline = PreprocessingPipeline(input_dir="data",
stretch_factors=[0.975, 1, 1.025],
split_size=30,
sampling_rate=sampling_rate,
n_velocity_bins=n_velocity_bins,
transpositions=range(-2, 3),
training_val_split=0.9,
max_encoded_length=seq_length + 1,
min_encoded_length=257)
pipeline_start = time.time()
pipeline.run()
runtime = time.time() - pipeline_start
print(f"MIDI pipeline runtime: {runtime / 60 : .1f}m")
today = datetime.date.today().strftime('%m%d%Y')
t = str(time.time())
# checkpoint = f"saved_models/tf_{today}_{t}"
checkpoint = f"saved_models/tf_lstm_both"
training_sequences = pipeline.encoded_sequences['training']
validation_sequences = pipeline.encoded_sequences['validation']
batch_size = 16
train(transformer,
training_sequences,
validation_sequences,
epochs=args.n_epochs,
evaluate_per=1,
batch_size=batch_size,
batches_per_print=100,
padding_index=0,
checkpoint_path=checkpoint,
early_stopping_value=early_stopping)
help='number of layers (default: 2)')
# args holds all passed-in arguments
args = parser.parse_args()
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
print("Using device {}.".format(device))
torch.manual_seed(args.seed)
# Load the training data.
train_loader = _get_train_data_loader(args.batch_size, args.data_dir)
# To get params from the parser, call args.argument_name, ex. args.epochs or ards.hidden_dim
# Don't forget to move your model .to(device) to move to GPU , if appropriate
model = LSTMClassifier(args.input_dim, args.hidden_dim, args.num_layers,
args.output_dim).to(device)
## TODO: Define an optimizer and loss function for training
optimizer = optim.Adam(model.parameters())
criterion = torch.nn.MSELoss()
# Trains the model (given line of code, which calls the above training function)
# Keep the keys of this dictionary as they are
model_info_path = os.path.join(args.model_dir, 'model_info.pth')
with open(model_info_path, 'wb') as f:
model_info = {
'num_layers': args.num_layers,
'hidden_dim': args.hidden_dim,
'output_dim': args.output_dim,
'input_dim': args.input_dim,
if not os.path.exists(save_root):
os.makedirs(save_root)
#print('writing results to '+save_root)
# create data generators
generator_train = get_batch_transform(model, train_data)
generator_val = get_batch_transform(model, val_data)
generator_test = get_batch_transform(model, test_data)
# train a predictor model
if exp_model == 'LR':
model_snt = LRClassifier()
elif exp_model == 'LSTM':
model_snt = LSTMClassifier()
else:
raise NotImplementedError
if args.cuda:
model_snt.cuda()
iters_max = 4000
lr_base = 0.001
lr_final = 0.00005
lr_new = lr_base
criterion = nn.BCELoss()
optimizer = torch.optim.Adam(model_snt.parameters(), lr=lr_base)
iters_val = []
accus_val = []
loss_val = []
def test(args):
dataset_test = FirmaData_select_subjects(args.data_dir, 30, args.subset_par[0], args.subset_par[1], args.subset_par[2],args.subjects_list,
subset='test', pre_process=False)
dat_loader_test = DataLoader(dataset_test, batch_size=args.batch_size, shuffle=True)
if args.test_all:
for loadid in range(args.num_epochs):
saved_model = os.path.join(args.save_path, 'model_' + str(loadid) + '.tar')
checkpoint = torch.load(saved_model)
model = LSTMClassifier(dataset_test[0][0].shape[1], args.hidden_dim, output_size=3)
model.cuda()
model.load_state_dict(checkpoint['model_state_dict'])
acc,f1,_= evaluate_test_set(model, dat_loader_test)
print('model {} test_accuracy:{:5.4f}, f1_score:{:5.4f}'.format(loadid,acc,f1))
else:
loadid=args.test_id
saved_model = os.path.join(args.save_path, 'model_' + str(loadid) + '.tar')
checkpoint = torch.load(saved_model)
model = LSTMClassifier(dataset_test[0][0].shape[1], args.hidden_dim, output_size=3)
model.cuda()
model.load_state_dict(checkpoint['model_state_dict'])
acc,f1, _ = evaluate_test_set(model, dat_loader_test)
print('model {} test_accuracy:{:5.4f}, f1_score:{:5.4f}'.format(loadid,acc,f1))
parser.add_argument('--model-dir', type=str, default=os.environ['SM_MODEL_DIR'])
parser.add_argument('--data-dir', type=str, default=os.environ['SM_CHANNEL_TRAINING'])
parser.add_argument('--num-gpus', type=int, default=os.environ['SM_NUM_GPUS'])
args = parser.parse_args()
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
print("Using device {}.".format(device))
torch.manual_seed(args.seed)
# Load the training data.
train_loader = _get_train_data_loader(args.batch_size, args.data_dir)
# Build the model.
model = LSTMClassifier(args.embedding_dim, args.hidden_dim, args.vocab_size).to(device)
with open(os.path.join(args.data_dir, "word_dict.pkl"), "rb") as f:
model.word_dict = pickle.load(f)
print("Model loaded with embedding_dim {}, hidden_dim {}, vocab_size {}.".format(
args.embedding_dim, args.hidden_dim, args.vocab_size
))
# Train the model.
optimizer = optim.Adam(model.parameters())
loss_fn = torch.nn.BCELoss()
train(model, train_loader, args.epochs, optimizer, loss_fn, device)
# Save the parameters used to construct the model
def train(model, train_loader, epochs, optimizer, loss_fn, device):
"""
This is the training method that is called by the PyTorch training script. The parameters
passed are as follows:
model - The PyTorch model that we wish to train.
train_loader - The PyTorch DataLoader that should be used during training.
epochs - The total number of epochs to train for.
optimizer - The optimizer to use during training.
loss_fn - The loss function used for training.
device - Where the model and data should be loaded (gpu or cpu).
"""
# TODO: Paste the train() method developed in the notebook here.
def train(model, train_loader, epochs, optimizer, loss_fn, device):
for epoch in range(1, epochs + 1):
model.train()
total_loss = 0
for batch in train_loader:
batch_X, batch_y = batch
batch_X = batch_X.to(device)
batch_y = batch_y.to(device)
# TODO: Complete this train method to train the model provided.
optimizer.zero_grad()
out = model.forward(batch_X)
loss = loss_fn(out, batch_y)
loss.backward()
optimizer.step()
total_loss += loss.data.item()
print("Epoch: {}, BCELoss: {}".format(epoch, total_loss / len(train_loader)))
pass
if __name__ == '__main__':
# All of the model parameters and training parameters are sent as arguments when the script
# is executed. Here we set up an argument parser to easily access the parameters.
parser = argparse.ArgumentParser()
# Training Parameters
parser.add_argument('--batch-size', type=int, default=512, metavar='N',
help='input batch size for training (default: 512)')
parser.add_argument('--epochs', type=int, default=10, metavar='N',
help='number of epochs to train (default: 10)')
parser.add_argument('--seed', type=int, default=1, metavar='S',
help='random seed (default: 1)')
# Model Parameters
parser.add_argument('--embedding_dim', type=int, default=32, metavar='N',
help='size of the word embeddings (default: 32)')
parser.add_argument('--hidden_dim', type=int, default=100, metavar='N',
help='size of the hidden dimension (default: 100)')
parser.add_argument('--vocab_size', type=int, default=5000, metavar='N',
help='size of the vocabulary (default: 5000)')
# SageMaker Parameters
parser.add_argument('--hosts', type=list, default=json.loads(os.environ['SM_HOSTS']))
parser.add_argument('--current-host', type=str, default=os.environ['SM_CURRENT_HOST'])
parser.add_argument('--model-dir', type=str, default=os.environ['SM_MODEL_DIR'])
parser.add_argument('--data-dir', type=str, default=os.environ['SM_CHANNEL_TRAINING'])
parser.add_argument('--num-gpus', type=int, default=os.environ['SM_NUM_GPUS'])
args = parser.parse_args()
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
print("Using device {}.".format(device))
torch.manual_seed(args.seed)
# Load the training data.
train_loader = _get_train_data_loader(args.batch_size, args.data_dir)
# Build the model.
model = LSTMClassifier(args.embedding_dim, args.hidden_dim, args.vocab_size).to(device)
with open(os.path.join(args.data_dir, "word_dict.pkl"), "rb") as f:
model.word_dict = pickle.load(f)
print("Model loaded with embedding_dim {}, hidden_dim {}, vocab_size {}.".format(
args.embedding_dim, args.hidden_dim, args.vocab_size
))
# Train the model.
optimizer = optim.Adam(model.parameters())
loss_fn = torch.nn.BCELoss()
train(model, train_loader, args.epochs, optimizer, loss_fn, device)
# Save the parameters used to construct the model
model_info_path = os.path.join(args.model_dir, 'model_info.pth')
with open(model_info_path, 'wb') as f:
model_info = {
'embedding_dim': args.embedding_dim,
'hidden_dim': args.hidden_dim,
'vocab_size': args.vocab_size,
}
torch.save(model_info, f)
# Save the word_dict
word_dict_path = os.path.join(args.model_dir, 'word_dict.pkl')
with open(word_dict_path, 'wb') as f:
pickle.dump(model.word_dict, f)
# Save the model parameters
model_path = os.path.join(args.model_dir, 'model.pth')
with open(model_path, 'wb') as f:
torch.save(model.cpu().state_dict(), f)
