Added info and removed git dependencies (conch, musk) to prepare pypi upload

Author: PierreMarza

README.md

@@ -1,7 +1,7 @@
 
 # Tile-level Histopathology image Understanding benchmark
 
-<img src="docs/banner.png" />
+<img src="https://raw.githubusercontent.com/MICS-Lab/thunder/main/docs/banner.png" />
 
 <div align="center">
 
@@ -19,7 +19,7 @@ We introduce **THUNDER**, a comprehensive benchmark designed to rigorously compa
 
 We propose a benchmark to compare and study foundation models across three axes: (i) downstream task performance, (ii) feature space comparisons, and (iii) uncertainty and robustness. Our current version integrates 23 foundation models, vision-only, vision-language, trained on pathology or natural images, on 16 datasets covering different magnifications and organs. THUNDER also supports the use of new user-defined models for direct comparisons.
 
-<img src="docs/overview.png" />
+<img src="https://raw.githubusercontent.com/MICS-Lab/thunder/main/docs/overview.png" />
 
 
 ## Usage
@@ -73,3 +73,10 @@ export THUNDER_BASE_DATA_FOLDER="/path/to/your/data/folder"
 ```
 
 Replace `/path/to/your/data/folder` with your desired storage directory.
+
+If you want to use the CONCH and MUSK models, you should install them as follows:
+
+```console
+pip install git+https://github.com/Mahmoodlab/CONCH.git # CONCH
+pip install git+https://github.com/lilab-stanford/MUSK.git # MUSK
+```

docs/getting_started.md

@@ -21,6 +21,13 @@ export THUNDER_BASE_DATA_FOLDER=/path/to/thunder_base_data_folder
     Without this environment variable, the package will not work. The folder should be empty and the package will create the necessary subfolders. We will save datasets, foundation models, pre-computed embeddings and output files from ran tasks. Importantly, you will be able to find the output file for a task at the following location:
     `$THUNDER_BASE_DATA_FOLDER/outputs/res/<dataset>/<model>/<task>/<adaptation_type>/outputs.json`
 
+If you want to use the CONCH and MUSK models, you should install them as follows:
+
+```console
+pip install git+https://github.com/Mahmoodlab/CONCH.git # CONCH
+pip install git+https://github.com/lilab-stanford/MUSK.git # MUSK
+```
+
 ## CLI Usage
 
 You can run the following command to see all available options,

setup.py

@@ -4,6 +4,10 @@
     name="thundr",
     version="0.1.0",
     description="THUNDER: Tile-level Histopathology image UNDERstanding benchmark",
+    long_description=open("README.md").read(),
+    long_description_content_type="text/markdown",
+    url="https://github.com/MICS-Lab/thunder",
+    license="CC-BY-4.0",
     package_dir={"": "src"},
     packages=find_packages(where="src"),
     install_requires=[
@@ -39,8 +43,6 @@
         "wilds==2.0.0",
         "kornia==0.8.0",
         "typer==0.15.3",
-        "conch @ git+https://github.com/Mahmoodlab/CONCH.git",
-        "musk @ git+https://github.com/lilab-stanford/MUSK.git",
     ],
     extras_require={
         "dev": ["pytest"],

src/thunder/models/pretrained_models.py

@@ -354,7 +354,12 @@ def get_from_open_clip(ckpt_path: str):
 
     :param ckpt_path: path to the stored checkpoint.
     """
-    from conch.open_clip_custom import create_model_from_pretrained
+    try:
+        from conch.open_clip_custom import create_model_from_pretrained
+    except ImportError:
+        raise ImportError(
+            "In order to use CONCH, please run the following: 'pip install git+https://github.com/Mahmoodlab/CONCH.git'"
+        )
 
     model, transform = create_model_from_pretrained("conch_ViT-B-16", ckpt_path)
 
@@ -417,7 +422,12 @@ def get_musk(ckpt_path: str):
 
     :ckpt_path: path to the stored checkpoint.
     """
-    from musk import utils, modeling
+    try:
+        from musk import utils, modeling
+    except ImportError:
+        raise ImportError(
+            "In order to use MUSK, please run the following: 'pip install git+https://github.com/lilab-stanford/MUSK.git'"
+        )
     from timm.data.constants import IMAGENET_INCEPTION_MEAN, IMAGENET_INCEPTION_STD
     from torchvision import transforms