Update README.md

README.md

@@ -1,45 +1,64 @@
 ---
 license: mit
 tags:
-  - super-resolution
-  - ddim
-  - weather
-  - wind
+  - downscaling
+  - ERA5 - COSMO-REA6
+  - reanalysis data
+  - wind velocities
+  - diffusion
+  - superresolution
 library_name: diffusers
 model_type: ddim
 datasets:
   - your-dataset-name
 ---
-# DDIM-DSC (4× Super-Resolution for Wind Data)
----
 
-## 🧠 Model Architecture
+# DDIM-DSC: 4× Downscaling of Wind Velocities
+
+**DDIM-DSC** is a custom-trained [Denoising Diffusion Implicit Model (DDIM)](https://github.com/huggingface/diffusers) designed for the **superresolution of wind velocity fields** from coarse- to high-resolution using reanalysis data.
+
+It performs **4× spatial downscaling** on 2-channel wind fields (u and v components), using **ERA5** as low-resolution input and **COSMO-REA6** as the high-resolution target.
 
-- **Base**: DDIM
-- **Input channels**: 10 
-- **Output channels**: 10
-- **Scale factor**: 4×
 ---
 
+## 📊 Data
+
+- **Input**: ERA5 100 m wind components (u, v), 2 channels  
+- **Target**: COSMO-REA6 100 m wind components (u, v), 2 channels  
+- **Sequence length**: 3 (with temporal context across 3 timesteps)  
+- **Total input channels**: 8 (2 channels × 3 timesteps + 2 static channels)
+
 ---
-## 🚀 How to Use
+
+## 🧠 Model Architecture
+
+- **Model type**: DDIM (using `diffusers`)
+- **Scheduler**: DDIMScheduler  
+- **Conditioning**: Concatenated temporal sequences  
+- **Latent noise sampling**: 10 per input  
+- **Scale factor**: 4×  
+- **Input channels**: 8  
+- **Output channels**: 2  
+- **Note**: The low-resolution input must be **resized to high-resolution shape using bilinear interpolation** before being passed into the model.
+
+## 🚀 Usage
+
 ```python
 import torch
-from diffusers import DiffusionPipeline, DDIMScheduler
-from diffusers import DDIMScheduler
+from diffusers import DiffusionPipeline
 
-# 
+# load the custom DDIM pipeline
 pipe = DiffusionPipeline.from_pretrained(
     "lschmidt/ddim-dsc",
-    custom_pipeline="cond_ddim_pipeline", 
+    custom_pipeline="cond_ddim_pipeline",
     trust_remote_code=True
 )
 
-#
-batch_size = 1
-in_channels = 6
-sequence_len = 3
+# create a sample low-resolution input --> shape: (sequence_length, channels, height, width)
+lres_image = torch.randn((3, 2, 32, 32)).to(pipe.device)
+
+# interpolate to match high-resolution
+
 
-# 
-lres_image = torch.randn((sequence_len, in_channels, sample_size, sample_size)).to(pipe.device)
-pipe(image=lres_image)
+# run inference
+outputs = pipe(image=lres_image)