SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

small fix and updates to readme

Browse Source
This commit is contained in:
thomwolf
2019-06-19 09:38:38 +02:00
parent f7e2ac01ea
commit 68ab9599ce
5 changed files with 53 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files

31
examples/bertology.py
View File

@@ -8,7 +8,7 @@ import torch
import torch.nn.functional as F
import numpy as np
from pytorch_pretrained_bert import BertModel, BertTokenizer
from pytorch_pretrained_bert import BertForSequenceClassification, BertTokenizer
logging.basicConfig(format = '%(asctime)s - %(levelname)s - %(name)s - %(message)s',
datefmt = '%m/%d/%Y %H:%M:%S',
@@ -17,24 +17,33 @@ logger = logging.getLogger(__name__)
def run_model():
parser = argparse.ArgumentParser()
parser.add_argument('--model_name_or_path', type=str, default='bert-base-uncased',
help='pretrained model name or path to local checkpoint')
parser.add_argument('--model_name_or_path', type=str, default='bert-base-uncased', help='pretrained model name or path to local checkpoint')
parser.add_argument("--seed", type=int, default=42)
parser.add_argument("--local_rank", type=int, default=-1, help="local_rank for distributed training on gpus")
parser.add_argument("--no_cuda", action='store_true', help="Whether not to use CUDA when available")
args = parser.parse_args()
print(args)
if args.batch_size == -1:
args.batch_size = 1
assert args.nsamples % args.batch_size == 0
np.random.seed(args.seed)
torch.random.manual_seed(args.seed)
torch.cuda.manual_seed(args.seed)
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
if args.local_rank == -1 or args.no_cuda:
args.device = torch.device("cuda" if torch.cuda.is_available() and not args.no_cuda else "cpu")
n_gpu = torch.cuda.device_count()
else:
torch.cuda.set_device(args.local_rank)
args.device = torch.device("cuda", args.local_rank)
n_gpu = 1
# Initializes the distributed backend which will take care of sychronizing nodes/GPUs
torch.distributed.init_process_group(backend='nccl')
logging.basicConfig(level = logging.INFO if args.local_rank in [-1, 0] else logging.WARN)
logger.info("device: {} n_gpu: {}, distributed training: {}, 16-bits training: {}".format(
args.device, n_gpu, bool(args.local_rank != -1), args.fp16))
tokenizer = BertTokenizer.from_pretrained(args.model_name_or_path)
model = BertModel.from_pretrained(args.model_name_or_path)
model.to(device)
model = BertForSequenceClassification.from_pretrained(args.model_name_or_path)
model.to(args.device)
model.eval()

6
examples/run_classifier.py
View File

@@ -187,7 +187,7 @@ def main():
if os.path.exists(args.output_dir) and os.listdir(args.output_dir) and args.do_train and not args.overwrite_output_dir:
raise ValueError("Output directory ({}) already exists and is not empty.".format(args.output_dir))
if not os.path.exists(args.output_dir):
if not os.path.exists(args.output_dir) and args.local_rank in [-1, 0]:
os.makedirs(args.output_dir)
task_name = args.task_name.lower()
@@ -361,6 +361,10 @@ def main():
# Load a trained model and vocabulary that you have fine-tuned
model = BertForSequenceClassification.from_pretrained(args.output_dir, num_labels=num_labels)
tokenizer = BertTokenizer.from_pretrained(args.output_dir, do_lower_case=args.do_lower_case)
# Good practice: save your training arguments together with the trained model
output_args_file = os.path.join(args.output_dir, 'training_args.bin')
torch.save(args, output_args_file)
else:
model = BertForSequenceClassification.from_pretrained(args.bert_model)

4
examples/run_squad.py
View File

@@ -331,6 +331,10 @@ def main():
# Load a trained model and vocabulary that you have fine-tuned
model = BertForQuestionAnswering.from_pretrained(args.output_dir)
tokenizer = BertTokenizer.from_pretrained(args.output_dir, do_lower_case=args.do_lower_case)
# Good practice: save your training arguments together with the trained model
output_args_file = os.path.join(args.output_dir, 'training_args.bin')
torch.save(args, output_args_file)
else:
model = BertForQuestionAnswering.from_pretrained(args.bert_model)