huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-11 04:14:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

Hotfixing idefics base64 parsing.

Browse Source
This commit is contained in:
Nicolas Patry 2023-10-05 11:04:42 +00:00
parent 3c373dcc53
commit 4930a9d8b7
2 changed files with 13 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
server/text_generation_server/models/custom_modeling/idefics_image_processing.py
View File

@ -201,8 +201,13 @@ class IdeficsImageProcessor(BaseImageProcessor):
response = requests.get(image_url_or_urls, stream=True, headers=headers, timeout=(1, 5)) response = requests.get(image_url_or_urls, stream=True, headers=headers, timeout=(1, 5))
response.raise_for_status() response.raise_for_status()
content = response.content content = response.content
else: elif image.startswith("data:"):
# https://stackoverflow.com/questions/17090571/is-there-a-way-to-set-background-image-as-a-base64-encoded-image
# data:image/png;base64,xxx
image = image.split(",")[-1]
content = base64.b64decode(image) content = base64.b64decode(image)
else:
raise ValueError(f"Unrecognized image {image}")
try: try:
image = Image.open(BytesIO(content)) image = Image.open(BytesIO(content))

8
server/text_generation_server/models/custom_modeling/idefics_processing.py
View File

@ -112,6 +112,11 @@ def is_url(string):
result = urlparse(string) result = urlparse(string)
return all([result.scheme, result.netloc]) return all([result.scheme, result.netloc])
def is_image(string):
"""Checks if the passed string contains a valid url and nothing else. e.g. if space is included it's immediately
invalidated the url"""
return is_url(string) or string.startswith("data:")
class IdeficsProcessor(ProcessorMixin): class IdeficsProcessor(ProcessorMixin):
r""" r"""
@ -314,7 +319,7 @@ class IdeficsProcessor(ProcessorMixin):
if isinstance(item, str): if isinstance(item, str):
item = item.strip(" ") item = item.strip(" ")
if is_url(item): if is_image(item):
image = self.image_processor.fetch_images(item) image = self.image_processor.fetch_images(item)
full_text += image_tokens(last_was_image) full_text += image_tokens(last_was_image)
image_objects.append(image) image_objects.append(image)
@ -339,6 +344,7 @@ class IdeficsProcessor(ProcessorMixin):
image_objects = self.image_processor(image_objects, transform=transform) image_objects = self.image_processor(image_objects, transform=transform)
text_encoding = self.tokenizer( text_encoding = self.tokenizer(
text=full_text, text=full_text,
add_special_tokens=False, add_special_tokens=False,