Update README.md

README.md

@@ -73,43 +73,55 @@ tags:
 pretty_name: mmls-v2
 size_categories:
 - 1K<n<10K
+
+license: cc-by-4.0
+
+citation: >
+  @inproceedings{paheding2026mmlsv2,
+    title={Mmlsv2: A multimodal dataset for martian landslide detection in remote sensing imagery},
+    author={Paheding, S. and Reyes-Angulo, A. A. and Ramos, L. T. and Sappa, A. D. and KS, S. K. and Oommen, T.},
+    booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
+    pages={10329--10338},
+    year={2026}
+  }
 ---
 
-# MMLS v2 (Multimodal Martian Landslide Dataset)
+# MMLS v2: Multimodal Martian Landslide Dataset
+
+This repository hosts a Hugging Face compatible, Parquet-formatted version of the **MMLS v2** dataset. It contains 128x128 multi-channel images for Martian landslide segmentation, featuring 7 data channels and 1 segmentation mask.
+
+> **Disclaimer:** I am not the original creator of this dataset. I am hosting this pre-processed version to allow researchers to easily stream the multi-channel arrays directly into PyTorch training pipelines without dealing with manual TIFF decoding. All credit for data collection and the original paper belongs to the authors of MMLSv2. Please see the Citation section below.
 
-This dataset contains 128x128 multi-channel images for Martian landslide segmentation. 
-It features 7 data channels (RGB, DEM, Thermal Inertia, Grayscale) and 1 segmentation mask.
+## Dataset Structure
 
-## How to use in PyTorch
+Each sample in the dataset is a dictionary containing pre-formatted, multi-dimensional arrays. To maintain 3D spatial consistency across all modalities, single-channel arrays include an explicit channel dimension of `1`.
 
-Because this dataset contains stacked multidimensional arrays, you should use the following wrapper to load it directly into your PyTorch training loops:
+* **`rgb`**: `(3, 128, 128)` — Float32
+* **`dem`**: `(1, 128, 128)` — Float32
+* **`thermal_inertial`**: `(1, 128, 128)` — Float32
+* **`grayscale`**: `(1, 128, 128)` — Float32
+* **`label`**: `(1, 128, 128)` — Float32 (Segmentation Mask)
+
+## Quickstart (PyTorch)
+
+You can load this dataset directly into PyTorch tensors using the standard Hugging Face `datasets` library. No manual data conversion is required.
 
 ```python
-import torch
-from torch.utils.data import DataLoader, Dataset
 from datasets import load_dataset
 
+# 1. Load a specific split from the Hub (train, val, or test)
+dataset = load_dataset("sattwik21/mmls-v2", split="train")
+
+# 2. Automatically format all arrays as native PyTorch Tensors
+dataset = dataset.with_format("torch")
 
-class MMLSHubDataset(Dataset):
-    def __init__(self, split="train"):
-        # load_dataset pulls from the hub; .with_format("torch") auto-converts arrays to Tensors
-        self.hf_dataset = load_dataset("sattwik21/mmls-v2", split=split).with_format("torch")
-        
-    def __len__(self):
-        return len(self.hf_dataset)
-        
-    def __getitem__(self, idx):
-        sample = self.hf_dataset[idx]
-        
-        # Returns a dictionary (or replace with your custom Tensorclass)
-        return {
-            "rgb": sample["rgb"],
-            "dem": sample["dem"],
-            "thermal": sample["thermal_inertial"],
-            "grayscale": sample["grayscale"],
-            "label": sample["label"]
-        }
+# 3. Pull a sample to verify
+sample = dataset[0]
 
-# Usage:
-train_dataset = MMLSHubDataset(split="train")
-train_loader = DataLoader(train_dataset, batch_size=32, shuffle=True)
+# 4. Verify tensor dimensions
+print("RGB shape:", sample["rgb"].shape)                 # Expected: torch.Size([3, 128, 128])
+print("DEM shape:", sample["dem"].shape)                 # Expected: torch.Size([1, 128, 128])
+print("Thermal shape:", sample["thermal_inertial"].shape)# Expected: torch.Size([1, 128, 128])
+print("Grayscale shape:", sample["grayscale"].shape)     # Expected: torch.Size([1, 128, 128])
+print("Label shape:", sample["label"].shape)             # Expected: torch.Size([1, 128, 128])
+```