Add 421.image caption generator benchmark and added its data in bench… #218
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…marks-data submodule Signed-off-by: Abhishek Kumar <[email protected]>
WalkthroughThe changes introduce multiple new files for an image captioning benchmark within a subproject. This includes configuration files, input generation functionality, and an implementation for generating captions using pre-trained models. Additionally, requirements files are added to specify necessary Python packages. The subproject reference has also been updated to a new commit in the version control system. Changes
Poem
Thank you for using CodeRabbit. We offer it for free to the OSS community and would appreciate your support in helping us grow. If you find it useful, would you consider giving us a shout-out on your favorite social media? TipsChatThere are 3 ways to chat with CodeRabbit:
Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments. CodeRabbit Commands (invoked as PR comments)
Additionally, you can add CodeRabbit Configuration File (
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 3
Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Files selected for processing (5)
- benchmarks-data (1 hunks)
- benchmarks/700.image/701.image-captioning/config.json (1 hunks)
- benchmarks/700.image/701.image-captioning/input.py (1 hunks)
- benchmarks/700.image/701.image-captioning/python/function.py (1 hunks)
- benchmarks/700.image/701.image-captioning/python/requirements.txt (1 hunks)
Files skipped from review due to trivial changes (2)
- benchmarks-data
- benchmarks/700.image/701.image-captioning/config.json
benchmarks/700.image/701.image-captioning/python/requirements.txt
Outdated
Show resolved
Hide resolved
|
Benchmark PR data is here spcl/serverless-benchmarks-data#4 |
Signed-off-by: Abhishek Kumar <[email protected]>
octonawish-akcodes
changed the title
mcopik
requested changes
Aug 27, 2024
benchmarks/400.inference/421.image-captioning/python/function.py
Outdated
Show resolved
Hide resolved
Signed-off-by: Abhishek Kumar <[email protected]>
There was a problem hiding this comment.
Actionable comments posted: 2
Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Files selected for processing (4)
- benchmarks/400.inference/421.image-captioning/config.json (1 hunks)
- benchmarks/400.inference/421.image-captioning/input.py (1 hunks)
- benchmarks/400.inference/421.image-captioning/python/function.py (1 hunks)
- benchmarks/400.inference/421.image-captioning/python/requirements.txt (1 hunks)
Files skipped from review due to trivial changes (1)
- benchmarks/400.inference/421.image-captioning/config.json
Additional context used
Ruff
benchmarks/400.inference/421.image-captioning/python/function.py
34-34: Undefined name
os(F821)
Additional comments not posted (1)
benchmarks/400.inference/421.image-captioning/python/requirements.txt (1)
1-3: Verify Python version compatibility.The listed package versions need to be verified for compatibility with the Python versions used in this project.
Comment on lines
+18
to
+42
| input_files = [] | ||
| for ext in ['*.jpg', '*.jpeg', '*.png']: | ||
| input_files.extend(glob.glob(os.path.join(data_dir, ext))) | ||
|
|
||
| if not input_files: | ||
| raise ValueError("No input files found in the provided directory.") | ||
|
|
||
| for file in input_files: | ||
| img = os.path.relpath(file, data_dir) | ||
| upload_func(0, img, file) | ||
|
|
||
| input_config = { | ||
| 'object': { | ||
| 'key': img, | ||
| 'width': 200, | ||
| 'height': 200 | ||
| }, | ||
| 'bucket': { | ||
| 'bucket': benchmarks_bucket, | ||
| 'input': input_paths[0], | ||
| 'output': output_paths[0] | ||
| } | ||
| } | ||
|
|
||
| return input_config |
There was a problem hiding this comment.
Fix variable scope and enhance error handling.
The variable img is used outside its defining loop, which could lead to unexpected behavior if no files are found. Consider defining img outside the loop or handling this case explicitly.
Apply this diff to address the scope issue and enhance error handling:
-def generate_input(data_dir, size, benchmarks_bucket, input_paths, output_paths, upload_func):
+def generate_input(data_dir, size, benchmarks_bucket, input_paths, output_paths, upload_func):
input_files = []
for ext in ['*.jpg', '*.jpeg', '*.png']:
input_files.extend(glob.glob(os.path.join(data_dir, ext)))
if not input_files:
raise ValueError("No input files found in the provided directory.")
+ img = None # Define img outside the loop to ensure it's available later
for file in input_files:
img = os.path.relpath(file, data_dir)
upload_func(0, img, file)
if img is None:
raise ValueError("No valid image files processed.")Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
Suggested change
| input_files = [] | |
| for ext in ['*.jpg', '*.jpeg', '*.png']: | |
| input_files.extend(glob.glob(os.path.join(data_dir, ext))) | |
| if not input_files: | |
| raise ValueError("No input files found in the provided directory.") | |
| for file in input_files: | |
| img = os.path.relpath(file, data_dir) | |
| upload_func(0, img, file) | |
| input_config = { | |
| 'object': { | |
| 'key': img, | |
| 'width': 200, | |
| 'height': 200 | |
| }, | |
| 'bucket': { | |
| 'bucket': benchmarks_bucket, | |
| 'input': input_paths[0], | |
| 'output': output_paths[0] | |
| } | |
| } | |
| return input_config | |
| def generate_input(data_dir, size, benchmarks_bucket, input_paths, output_paths, upload_func): | |
| input_files = [] | |
| for ext in ['*.jpg', '*.jpeg', '*.png']: | |
| input_files.extend(glob.glob(os.path.join(data_dir, ext))) | |
| if not input_files: | |
| raise ValueError("No input files found in the provided directory.") | |
| img = None # Define img outside the loop to ensure it's available later | |
| for file in input_files: | |
| img = os.path.relpath(file, data_dir) | |
| upload_func(0, img, file) | |
| if img is None: | |
| raise ValueError("No valid image files processed.") |
Comment on lines
+1
to
+53
| import datetime | ||
| import io | ||
| from urllib.parse import unquote_plus | ||
| from PIL import Image | ||
| import torch | ||
| from transformers import VisionEncoderDecoderModel, ViTImageProcessor, AutoTokenizer | ||
| from . import storage | ||
|
|
||
| # Load the pre-trained ViT-GPT2 model | ||
| model = VisionEncoderDecoderModel.from_pretrained("nlpconnect/vit-gpt2-image-captioning") | ||
| image_processor = ViTImageProcessor.from_pretrained("nlpconnect/vit-gpt2-image-captioning") | ||
| tokenizer = AutoTokenizer.from_pretrained("nlpconnect/vit-gpt2-image-captioning") | ||
|
|
||
| model.eval() | ||
|
|
||
| client = storage.storage.get_instance() | ||
|
|
||
| def generate_caption(image_bytes): | ||
| image = Image.open(io.BytesIO(image_bytes)).convert("RGB") | ||
| pixel_values = image_processor(images=image, return_tensors="pt").pixel_values | ||
|
|
||
| with torch.no_grad(): | ||
| generated_ids = model.generate(pixel_values, max_length=16, num_beams=4) | ||
| generated_text = tokenizer.decode(generated_ids[0], skip_special_tokens=True) | ||
|
|
||
| return generated_text | ||
|
|
||
| def handler(event): | ||
| bucket = event.get('bucket').get('bucket') | ||
| input_prefix = event.get('bucket').get('input') | ||
| key = unquote_plus(event.get('object').get('key')) | ||
|
|
||
| download_begin = datetime.datetime.now() | ||
| img = client.download_stream(bucket, os.path.join(input_prefix, key)) | ||
| download_end = datetime.datetime.now() | ||
|
|
||
| process_begin = datetime.datetime.now() | ||
| caption = generate_caption(img) | ||
| process_end = datetime.datetime.now() | ||
|
|
||
| download_time = (download_end - download_begin) / datetime.timedelta(microseconds=1) | ||
| process_time = (process_end - process_begin) / datetime.timedelta(microseconds=1) | ||
|
|
||
| return { | ||
| 'result': { | ||
| 'caption': caption, | ||
| }, | ||
| 'measurement': { | ||
| 'download_time': download_time, | ||
| 'download_size': len(img), | ||
| 'compute_time': process_time | ||
| } | ||
| } |
There was a problem hiding this comment.
Fix undefined name and approve model usage.
The code uses os without importing it, which will cause a runtime error. The implementation of the model loading and image processing is correctly done.
Fix the missing import:
+import osCommittable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
Suggested change
| import datetime | |
| import io | |
| from urllib.parse import unquote_plus | |
| from PIL import Image | |
| import torch | |
| from transformers import VisionEncoderDecoderModel, ViTImageProcessor, AutoTokenizer | |
| from . import storage | |
| # Load the pre-trained ViT-GPT2 model | |
| model = VisionEncoderDecoderModel.from_pretrained("nlpconnect/vit-gpt2-image-captioning") | |
| image_processor = ViTImageProcessor.from_pretrained("nlpconnect/vit-gpt2-image-captioning") | |
| tokenizer = AutoTokenizer.from_pretrained("nlpconnect/vit-gpt2-image-captioning") | |
| model.eval() | |
| client = storage.storage.get_instance() | |
| def generate_caption(image_bytes): | |
| image = Image.open(io.BytesIO(image_bytes)).convert("RGB") | |
| pixel_values = image_processor(images=image, return_tensors="pt").pixel_values | |
| with torch.no_grad(): | |
| generated_ids = model.generate(pixel_values, max_length=16, num_beams=4) | |
| generated_text = tokenizer.decode(generated_ids[0], skip_special_tokens=True) | |
| return generated_text | |
| def handler(event): | |
| bucket = event.get('bucket').get('bucket') | |
| input_prefix = event.get('bucket').get('input') | |
| key = unquote_plus(event.get('object').get('key')) | |
| download_begin = datetime.datetime.now() | |
| img = client.download_stream(bucket, os.path.join(input_prefix, key)) | |
| download_end = datetime.datetime.now() | |
| process_begin = datetime.datetime.now() | |
| caption = generate_caption(img) | |
| process_end = datetime.datetime.now() | |
| download_time = (download_end - download_begin) / datetime.timedelta(microseconds=1) | |
| process_time = (process_end - process_begin) / datetime.timedelta(microseconds=1) | |
| return { | |
| 'result': { | |
| 'caption': caption, | |
| }, | |
| 'measurement': { | |
| 'download_time': download_time, | |
| 'download_size': len(img), | |
| 'compute_time': process_time | |
| } | |
| } | |
| import datetime | |
| import io | |
| from urllib.parse import unquote_plus | |
| from PIL import Image | |
| import torch | |
| from transformers import VisionEncoderDecoderModel, ViTImageProcessor, AutoTokenizer | |
| from . import storage | |
| +import os | |
| # Load the pre-trained ViT-GPT2 model | |
| model = VisionEncoderDecoderModel.from_pretrained("nlpconnect/vit-gpt2-image-captioning") | |
| image_processor = ViTImageProcessor.from_pretrained("nlpconnect/vit-gpt2-image-captioning") | |
| tokenizer = AutoTokenizer.from_pretrained("nlpconnect/vit-gpt2-image-captioning") | |
| model.eval() | |
| client = storage.storage.get_instance() | |
| def generate_caption(image_bytes): | |
| image = Image.open(io.BytesIO(image_bytes)).convert("RGB") | |
| pixel_values = image_processor(images=image, return_tensors="pt").pixel_values | |
| with torch.no_grad(): | |
| generated_ids = model.generate(pixel_values, max_length=16, num_beams=4) | |
| generated_text = tokenizer.decode(generated_ids[0], skip_special_tokens=True) | |
| return generated_text | |
| def handler(event): | |
| bucket = event.get('bucket').get('bucket') | |
| input_prefix = event.get('bucket').get('input') | |
| key = unquote_plus(event.get('object').get('key')) | |
| download_begin = datetime.datetime.now() | |
| img = client.download_stream(bucket, os.path.join(input_prefix, key)) | |
| download_end = datetime.datetime.now() | |
| process_begin = datetime.datetime.now() | |
| caption = generate_caption(img) | |
| process_end = datetime.datetime.now() | |
| download_time = (download_end - download_begin) / datetime.timedelta(microseconds=1) | |
| process_time = (process_end - process_begin) / datetime.timedelta(microseconds=1) | |
| return { | |
| 'result': { | |
| 'caption': caption, | |
| }, | |
| 'measurement': { | |
| 'download_time': download_time, | |
| 'download_size': len(img), | |
| 'compute_time': process_time | |
| } | |
| } |
Tools
Ruff
34-34: Undefined name
os(F821)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This Pull Request introduces a new
Image Caption Generation benchmarkfor SEBS, implemented using Python. This benchmark uses a hugging face modelnlpconnect/vit-gpt2-image-captioning. Link to model https://huggingface.co/nlpconnect/vit-gpt2-image-captioning and its license under Apache 2.0 https://huggingface.co/datasets/choosealicense/licenses/blob/main/markdown/apache-2.0.mdcc @mcopik
Summary by CodeRabbit
New Features
Bug Fixes