Upload README.md with huggingface_hub

README.md

@@ -4,8 +4,126 @@ tags:
 - lighter
 - model_hub_mixin
 - pytorch_model_hub_mixin
+language: en
+license: apache-2.0
+arxiv: 2501.09001
 ---
 
-This model has been pushed to the Hub using the [PytorchModelHubMixin](https://huggingface.co/docs/huggingface_hub/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin) integration:
-- Library: https://github.com/project-lighter/lighter-zoo
-- Docs: https://project-lighter.github.io/lighter/
+# CT-FM SegResNet
+
+This model is a SegResNet containing the weights of the pre-trained CT-FM, using contrastive self-supervised learning on a huge dataset of 148,000 CT scans from the Imaging Data Commons. 
+
+
+## Running instructions
+
+
+# CT-FM SegResNet Fine-tuning
+ 
+This notebook demonstrates how to:
+1. Load a SSL pre-trained model into a SegResNet
+2. Recommended preprocessing and postprocessing steps that were used during pre-training
+3. Finetuning instructions overview
+
+## Setup
+Install requirements and import necessary packages
+
+
+```python
+# Install lighter_zoo package
+%pip install lighter_zoo -U -qq
+```
+
+    Note: you may need to restart the kernel to use updated packages.
+
+
+
+```python
+# Imports
+import torch
+from lighter_zoo import SegResNet
+from monai.transforms import (
+    Compose, LoadImage, EnsureType, Orientation,
+    ScaleIntensityRange, CropForeground, Invert,
+    Activations, AsDiscrete, KeepLargestConnectedComponent,
+    SaveImage
+)
+from monai.inferers import SlidingWindowInferer
+```
+
+## Load Model
+Download and initialize the pre-trained model from HuggingFace Hub
+
+
+```python
+# Load pre-trained model
+model = SegResNet.from_pretrained(
+    "project-lighter/ct_fm_segresnet"
+)
+```
+
+## Setup Processing Pipelines
+Define preprocessing and postprocessing transforms
+
+
+```python
+# Preprocessing pipeline
+preprocess = Compose([
+    LoadImage(ensure_channel_first=True),  # Load image and ensure channel dimension
+    EnsureType(),                         # Ensure correct data type
+    Orientation(axcodes="SPL"),           # Standardize orientation
+    # Scale intensity to [0,1] range, clipping outliers
+    ScaleIntensityRange(
+        a_min=-1024,    # Min HU value
+        a_max=2048,     # Max HU value
+        b_min=0,        # Target min
+        b_max=1,        # Target max
+        clip=True       # Clip values outside range
+    ),
+    CropForeground()    # Remove background to reduce computation
+])
+```
+
+## Run Inference
+Process an input CT scan and extract features
+
+
+```python
+# Configure sliding window inference
+inferer = SlidingWindowInferer(
+    roi_size=[96, 160, 160],  # Size of patches to process
+    sw_batch_size=2,          # Number of windows to process in parallel
+    overlap=0.625,            # Overlap between windows (reduces boundary artifacts)
+    mode="gaussian"           # Gaussian weighting for overlap regions
+)
+
+# Input path
+input_path = "/home/suraj/Repositories/semantic-search-app/assets/scans/s0114.nii.gz"
+
+# Preprocess input
+input_tensor = preprocess(input_path)
+
+# Run inference
+with torch.no_grad():
+    model = model.to("cuda")
+    input_tensor = input_tensor.to("cuda")
+    output = inferer(input_tensor.unsqueeze(dim=0), model)[0]
+    output = output.to("cpu")
+
+
+print(output.shape)
+```
+
+    torch.Size([2, 227, 181, 258])
+
+
+## Fine-tuning instructions
+
+The above model does not have a trained decoder which means the predictions you will get are nonsensical. 
+
+You can however use the pre-trained encoder and the model architecture to finetune on your own datasets - especially if they are small sized. 
+
+A very simple way to fit this into your pipelines is to take the model loaded above using model = SegResNet.from_pretrained('project-lighter/ct_fm_segresnet') and replace the model in your training pipeline with this. 
+
+Using Auto3DSeg with our model is the recommended approach, follow the instructions here: https://project-lighter.github.io/CT-FM/replication-guide/downstream/#tumor-segmentation-with-auto3dseg
+
+