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

adding proxies options for the from_pretrained methods

Browse Source
This commit is contained in:
thomwolf
2019-08-20 16:59:11 +02:00
parent 6d0aa73981
commit 43489756ad
4 changed files with 36 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files

29
pytorch_transformers/file_utils.py
View File

@@ -17,8 +17,9 @@ from hashlib import sha256
from io import open
import boto3
import requests
from botocore.config import Config
from botocore.exceptions import ClientError
import requests
from tqdm import tqdm
try:
@@ -93,7 +94,7 @@ def filename_to_url(filename, cache_dir=None):
return url, etag
def cached_path(url_or_filename, cache_dir=None, force_download=False):
def cached_path(url_or_filename, cache_dir=None, force_download=False, proxies=None):
"""
Given something that might be a URL (or might be a local path),
determine which. If it's a URL, download the file and cache it, and
@@ -114,7 +115,7 @@ def cached_path(url_or_filename, cache_dir=None, force_download=False):
if parsed.scheme in ('http', 'https', 's3'):
# URL, so get it from the cache (downloading if necessary)
return get_from_cache(url_or_filename, cache_dir=cache_dir, force_download=force_download)
return get_from_cache(url_or_filename, cache_dir=cache_dir, force_download=force_download, proxies=proxies)
elif os.path.exists(url_or_filename):
# File, and it exists.
return url_or_filename
@@ -159,24 +160,24 @@ def s3_request(func):
@s3_request
def s3_etag(url):
def s3_etag(url, proxies=None):
"""Check ETag on S3 object."""
s3_resource = boto3.resource("s3")
s3_resource = boto3.resource("s3", config=Config(proxies=proxies))
bucket_name, s3_path = split_s3_path(url)
s3_object = s3_resource.Object(bucket_name, s3_path)
return s3_object.e_tag
@s3_request
def s3_get(url, temp_file):
def s3_get(url, temp_file, proxies=None):
"""Pull a file directly from S3."""
s3_resource = boto3.resource("s3")
s3_resource = boto3.resource("s3", config=Config(proxies=proxies))
bucket_name, s3_path = split_s3_path(url)
s3_resource.Bucket(bucket_name).download_fileobj(s3_path, temp_file)
def http_get(url, temp_file):
req = requests.get(url, stream=True)
def http_get(url, temp_file, proxies=None):
req = requests.get(url, stream=True, proxies=proxies)
content_length = req.headers.get('Content-Length')
total = int(content_length) if content_length is not None else None
progress = tqdm(unit="B", total=total)
@@ -187,7 +188,7 @@ def http_get(url, temp_file):
progress.close()
def get_from_cache(url, cache_dir=None, force_download=False):
def get_from_cache(url, cache_dir=None, force_download=False, proxies=None):
"""
Given a URL, look for the corresponding dataset in the local cache.
If it's not there, download it. Then return the path to the cached file.
@@ -204,10 +205,10 @@ def get_from_cache(url, cache_dir=None, force_download=False):
# Get eTag to add to filename, if it exists.
if url.startswith("s3://"):
etag = s3_etag(url)
etag = s3_etag(url, proxies=proxies)
else:
try:
response = requests.head(url, allow_redirects=True)
response = requests.head(url, allow_redirects=True, proxies=proxies)
if response.status_code != 200:
etag = None
else:
@@ -238,9 +239,9 @@ def get_from_cache(url, cache_dir=None, force_download=False):
# GET file object
if url.startswith("s3://"):
s3_get(url, temp_file)
s3_get(url, temp_file, proxies=proxies)
else:
http_get(url, temp_file)
http_get(url, temp_file, proxies=proxies)
# we are copying the file before closing it, so flush to avoid truncation
temp_file.flush()