757 add exaone path 2.0 #759
Conversation
Add EXAONE Path 2.0 Signed-off-by: Jongseong Jang <[email protected]>
Signed-off-by: Jongseong Jang <[email protected]>
for more information, see https://pre-commit.ci
There was a problem hiding this comment.
Pull Request Overview
This PR adds the latest version of the EXAONE Path model (EXAONE Path 2.0) to the Hugging Face models directory. The model is a pathology foundation model with 144 million parameters designed for whole-slide image (WSI) analysis and biomarker prediction.
- Adds complete EXAONE Path 2.0 model implementation with three-stage Vision Transformer architecture
- Includes comprehensive utilities for WSI processing, tissue segmentation, and tensor operations
- Provides sample data, documentation, and licensing for the new pathology model
Reviewed Changes
Copilot reviewed 11 out of 14 changed files in this pull request and generated 15 comments.
Show a summary per file
| File | Description |
|---|---|
| hf_models/README.md | Updates model listing table to include EXAONE Path 2.0 entry |
| hf_models/EXAONE-Path-2.0/exaonepath.py | Main model implementation with three-stage inference pipeline |
| hf_models/EXAONE-Path-2.0/networks/vit.py | Vision Transformer architectures for the three processing stages |
| hf_models/EXAONE-Path-2.0/utils/wsi_utils.py | Utilities for whole-slide image processing and tissue segmentation |
| hf_models/EXAONE-Path-2.0/utils/tensor_utils.py | Tensor manipulation utilities for multi-stage processing |
| hf_models/EXAONE-Path-2.0/requirements.txt | Dependencies specification including CUDA-enabled PyTorch |
| hf_models/EXAONE-Path-2.0/config.json | Model configuration parameters |
| hf_models/EXAONE-Path-2.0/README.md | Comprehensive documentation and usage instructions |
| hf_models/EXAONE-Path-2.0/LICENSE | Custom license agreement for the model |
| hf_models/EXAONE-Path-2.0/pytorch_model.bin | Pre-trained model weights (LFS tracked) |
| hf_models/EXAONE-Path-2.0/samples/sample.svs | Sample whole-slide image file (LFS tracked) |
Comments suppressed due to low confidence (2)
| import typing as t | ||
| from concurrent.futures import ThreadPoolExecutor | ||
| from pathlib import Path | ||
| from tracemalloc import start |
There was a problem hiding this comment.
The import 'start' from tracemalloc is not used anywhere in the code. This appears to be an unused import that should be removed.
| from tracemalloc import start | |
| # Line removed as it is unused |
Copilot uses AI. Check for mistakes.
|
|
||
| try: | ||
| from skimage import img_as_ubyte # type: ignore | ||
| except: |
There was a problem hiding this comment.
Bare except clause should specify the exception type. Consider using 'except ImportError:' to catch only the specific import error.
| except: | |
| except ImportError: |
Copilot uses AI. Check for mistakes.
| boxes = [cv2.boundingRect(c) for c in large_contours] | ||
|
|
||
| return ( | ||
| [cv2.boundingRect(c) for c in large_contours] |
There was a problem hiding this comment.
This list comprehension duplicates the same computation from line 198. Consider storing the result in the 'boxes' variable and reusing it.
| [cv2.boundingRect(c) for c in large_contours] | |
| boxes |
Copilot uses AI. Check for mistakes.
| 255, | ||
| dtype=wsi_arr.dtype, | ||
| ) | ||
| mask_out = np.zeros((packed_size[1], packed_size[0]), dtype=np.bool) |
There was a problem hiding this comment.
Using deprecated 'np.bool' dtype. Should use 'np.bool_' or 'bool' instead as np.bool has been deprecated.
| mask_out = np.zeros((packed_size[1], packed_size[0]), dtype=np.bool) | |
| mask_out = np.zeros((packed_size[1], packed_size[0]), dtype=bool) |
Copilot uses AI. Check for mistakes.
| mask_out[pos[1] : pos[1] + box[3], pos[0] : pos[0] + box[2]] = mask[ | ||
| box[1] : box[1] + box[3], box[0] : box[0] + box[2] | ||
| ] |
There was a problem hiding this comment.
[nitpick] This array slicing operation spans multiple lines in a confusing way. Consider restructuring for better readability.
| mask_out[pos[1] : pos[1] + box[3], pos[0] : pos[0] + box[2]] = mask[ | |
| box[1] : box[1] + box[3], box[0] : box[0] + box[2] | |
| ] | |
| mask_out[ | |
| pos[1] : pos[1] + box[3], pos[0] : pos[0] + box[2] | |
| ] = mask[box[1] : box[1] + box[3], box[0] : box[0] + box[2]] |
Copilot uses AI. Check for mistakes.
| white = [255, 255, 255] # 흰색 (B, G, R) | ||
|
|
||
| # cv2.copyMakeBorder 함수를 사용해 흰색 띠를 추가 | ||
| # 두께 30픽셀의 위쪽 흰색 띠 그리기 | ||
| cv2.rectangle(img, (0, 0), (width, thickness), white, -1) | ||
|
|
||
| # 두께 30픽셀의 아래쪽 흰색 띠 그리기 | ||
| cv2.rectangle(img, (0, height - thickness), (width, height), white, -1) | ||
|
|
||
| # 두께 30픽셀의 왼쪽 흰색 띠 그리기 | ||
| cv2.rectangle(img, (0, 0), (thickness, height), white, -1) | ||
|
|
||
| # 두께 30픽셀의 오른쪽 흰색 띠 그리기 |
There was a problem hiding this comment.
Comment is in Korean. Consider translating to English for consistency in an international codebase.
| white = [255, 255, 255] # 흰색 (B, G, R) | |
| # cv2.copyMakeBorder 함수를 사용해 흰색 띠를 추가 | |
| # 두께 30픽셀의 위쪽 흰색 띠 그리기 | |
| cv2.rectangle(img, (0, 0), (width, thickness), white, -1) | |
| # 두께 30픽셀의 아래쪽 흰색 띠 그리기 | |
| cv2.rectangle(img, (0, height - thickness), (width, height), white, -1) | |
| # 두께 30픽셀의 왼쪽 흰색 띠 그리기 | |
| cv2.rectangle(img, (0, 0), (thickness, height), white, -1) | |
| # 두께 30픽셀의 오른쪽 흰색 띠 그리기 | |
| white = [255, 255, 255] # White color (B, G, R) | |
| # Add white bands using the cv2.copyMakeBorder function | |
| # Draw a white band of 30-pixel thickness at the top | |
| cv2.rectangle(img, (0, 0), (width, thickness), white, -1) | |
| # Draw a white band of 30-pixel thickness at the bottom | |
| cv2.rectangle(img, (0, height - thickness), (width, height), white, -1) | |
| # Draw a white band of 30-pixel thickness on the left | |
| cv2.rectangle(img, (0, 0), (thickness, height), white, -1) | |
| # Draw a white band of 30-pixel thickness on the right |
Copilot uses AI. Check for mistakes.
| white = [255, 255, 255] # 흰색 (B, G, R) | ||
|
|
||
| # cv2.copyMakeBorder 함수를 사용해 흰색 띠를 추가 | ||
| # 두께 30픽셀의 위쪽 흰색 띠 그리기 | ||
| cv2.rectangle(img, (0, 0), (width, thickness), white, -1) | ||
|
|
||
| # 두께 30픽셀의 아래쪽 흰색 띠 그리기 | ||
| cv2.rectangle(img, (0, height - thickness), (width, height), white, -1) | ||
|
|
||
| # 두께 30픽셀의 왼쪽 흰색 띠 그리기 | ||
| cv2.rectangle(img, (0, 0), (thickness, height), white, -1) | ||
|
|
||
| # 두께 30픽셀의 오른쪽 흰색 띠 그리기 |
There was a problem hiding this comment.
Comment is in Korean. Consider translating to English for consistency in an international codebase.
| white = [255, 255, 255] # 흰색 (B, G, R) | |
| # cv2.copyMakeBorder 함수를 사용해 흰색 띠를 추가 | |
| # 두께 30픽셀의 위쪽 흰색 띠 그리기 | |
| cv2.rectangle(img, (0, 0), (width, thickness), white, -1) | |
| # 두께 30픽셀의 아래쪽 흰색 띠 그리기 | |
| cv2.rectangle(img, (0, height - thickness), (width, height), white, -1) | |
| # 두께 30픽셀의 왼쪽 흰색 띠 그리기 | |
| cv2.rectangle(img, (0, 0), (thickness, height), white, -1) | |
| # 두께 30픽셀의 오른쪽 흰색 띠 그리기 | |
| white = [255, 255, 255] # White color (B, G, R) | |
| # Add white bands using the cv2.copyMakeBorder function | |
| # Draw a white band of 30-pixel thickness at the top | |
| cv2.rectangle(img, (0, 0), (width, thickness), white, -1) | |
| # Draw a white band of 30-pixel thickness at the bottom | |
| cv2.rectangle(img, (0, height - thickness), (width, height), white, -1) | |
| # Draw a white band of 30-pixel thickness on the left | |
| cv2.rectangle(img, (0, 0), (thickness, height), white, -1) | |
| # Draw a white band of 30-pixel thickness on the right |
Copilot uses AI. Check for mistakes.
| white = [255, 255, 255] # 흰색 (B, G, R) | ||
|
|
||
| # cv2.copyMakeBorder 함수를 사용해 흰색 띠를 추가 | ||
| # 두께 30픽셀의 위쪽 흰색 띠 그리기 | ||
| cv2.rectangle(img, (0, 0), (width, thickness), white, -1) | ||
|
|
||
| # 두께 30픽셀의 아래쪽 흰색 띠 그리기 | ||
| cv2.rectangle(img, (0, height - thickness), (width, height), white, -1) | ||
|
|
||
| # 두께 30픽셀의 왼쪽 흰색 띠 그리기 | ||
| cv2.rectangle(img, (0, 0), (thickness, height), white, -1) | ||
|
|
||
| # 두께 30픽셀의 오른쪽 흰색 띠 그리기 |
There was a problem hiding this comment.
Comment is in Korean. Consider translating to English for consistency in an international codebase.
| white = [255, 255, 255] # 흰색 (B, G, R) | |
| # cv2.copyMakeBorder 함수를 사용해 흰색 띠를 추가 | |
| # 두께 30픽셀의 위쪽 흰색 띠 그리기 | |
| cv2.rectangle(img, (0, 0), (width, thickness), white, -1) | |
| # 두께 30픽셀의 아래쪽 흰색 띠 그리기 | |
| cv2.rectangle(img, (0, height - thickness), (width, height), white, -1) | |
| # 두께 30픽셀의 왼쪽 흰색 띠 그리기 | |
| cv2.rectangle(img, (0, 0), (thickness, height), white, -1) | |
| # 두께 30픽셀의 오른쪽 흰색 띠 그리기 | |
| white = [255, 255, 255] # White color (B, G, R) | |
| # Add white bands using the cv2.copyMakeBorder function | |
| # Draw a white band of 30-pixel thickness at the top | |
| cv2.rectangle(img, (0, 0), (width, thickness), white, -1) | |
| # Draw a white band of 30-pixel thickness at the bottom | |
| cv2.rectangle(img, (0, height - thickness), (width, height), white, -1) | |
| # Draw a white band of 30-pixel thickness on the left | |
| cv2.rectangle(img, (0, 0), (thickness, height), white, -1) | |
| # Draw a white band of 30-pixel thickness on the right |
Copilot uses AI. Check for mistakes.
| white = [255, 255, 255] # 흰색 (B, G, R) | ||
|
|
||
| # cv2.copyMakeBorder 함수를 사용해 흰색 띠를 추가 | ||
| # 두께 30픽셀의 위쪽 흰색 띠 그리기 | ||
| cv2.rectangle(img, (0, 0), (width, thickness), white, -1) | ||
|
|
||
| # 두께 30픽셀의 아래쪽 흰색 띠 그리기 | ||
| cv2.rectangle(img, (0, height - thickness), (width, height), white, -1) | ||
|
|
||
| # 두께 30픽셀의 왼쪽 흰색 띠 그리기 | ||
| cv2.rectangle(img, (0, 0), (thickness, height), white, -1) | ||
|
|
||
| # 두께 30픽셀의 오른쪽 흰색 띠 그리기 |
There was a problem hiding this comment.
Comment is in Korean. Consider translating to English for consistency in an international codebase.
| white = [255, 255, 255] # 흰색 (B, G, R) | |
| # cv2.copyMakeBorder 함수를 사용해 흰색 띠를 추가 | |
| # 두께 30픽셀의 위쪽 흰색 띠 그리기 | |
| cv2.rectangle(img, (0, 0), (width, thickness), white, -1) | |
| # 두께 30픽셀의 아래쪽 흰색 띠 그리기 | |
| cv2.rectangle(img, (0, height - thickness), (width, height), white, -1) | |
| # 두께 30픽셀의 왼쪽 흰색 띠 그리기 | |
| cv2.rectangle(img, (0, 0), (thickness, height), white, -1) | |
| # 두께 30픽셀의 오른쪽 흰색 띠 그리기 | |
| white = [255, 255, 255] # White color (B, G, R) | |
| # Add white bands using the cv2.copyMakeBorder function | |
| # Draw a white band at the top with a thickness of 30 pixels | |
| cv2.rectangle(img, (0, 0), (width, thickness), white, -1) | |
| # Draw a white band at the bottom with a thickness of 30 pixels | |
| cv2.rectangle(img, (0, height - thickness), (width, height), white, -1) | |
| # Draw a white band on the left with a thickness of 30 pixels | |
| cv2.rectangle(img, (0, 0), (thickness, height), white, -1) | |
| # Draw a white band on the right with a thickness of 30 pixels |
Copilot uses AI. Check for mistakes.
| if i < len(hole_contours) and hole_contours[i]: | ||
| for hole in hole_contours[i]: # type: ignore | ||
| cv2.fillPoly(mask, [hole], 0) # type: ignore | ||
| mask = mask.astype(np.bool) |
There was a problem hiding this comment.
Using deprecated 'np.bool' dtype. Should use 'np.bool_' or 'bool' instead as np.bool has been deprecated.
| mask = mask.astype(np.bool) | |
| mask = mask.astype(bool) |
Copilot uses AI. Check for mistakes.
|
Hi @cjs6211 thank you for the contribution. We do require that models in our zoo use the bundle format. You have a lot of your own code here so I wouldn't suggest reformatting everything to use MONAI components, but there does need to be at least an inference config file which can call into code you have defined. This bundle here is an example of an inference script with few components, and just calls into a defined method to implement an inference process. You can follow that pattern here but please do look at other bundles and see what you can do to make your code into a bundle. We do also require that this model be in the @aylward could you please double check that this model's license is acceptable to us, I think it's fine from our previous discussions on licenses. Thanks! |
Hi, @ericspod. We have already uploaded the EXAONE Path 2.0 model to Hugging Face at https://huggingface.co/LGAI-EXAONE/EXAONE-Path-2.0 (different from the past model), and we would like this version to be referenced in hf_models following the same pattern as the existing EXAONE Path model. However, since we were unsure of the correct process to request this kind of update, we submitted the current pull request. Please let us know exact way to refer to EXAONE Path 2.0. Thank you again! |
Hi @cjs6211 I would suggest removing the files that you've proposed here and instead create a directory in |
Fixes #757 .
Description
This PR adds the latest version of the EXAONE Path model, EXAONE Path 2.0.
The model is based on Hugging Face and is hosted at: https://huggingface.co/LGAI-EXAONE/EXAONE-Path-2.0.
Status
Ready
Please ensure all the checkboxes:
./runtests.sh --codeformat.versionandchangeloginmetadata.jsonif changing an existing bundle.CONTRIBUTING.md).monai,pytorchandnumpyare correct inmetadata.json.eval_metricsof the provided weights and TorchScript modules.large_file.yml./home/your_name/for"bundle_root").