upload inference code

README.md

@@ -12,10 +12,68 @@ This is an official implementation of paper 'Improving Diffusion Models for Auth
 
 
 - [x] demo model
-- [ ] inference code
+- [x] inference code
 - [ ] training code
 
 
+
+## Requirements
+
+```
+git clone https://github.com/yisol/IDM-VTON.git
+cd IDM-VTON
+
+conda env create -f environment.yaml
+conda activate idm
+```
+
+## Data preparation
+You can download VITON-HD dataset from [VITON-HD](https://github.com/shadow2496/VITON-HD).
+After download VITON-HD dataset, move vitonhd_test_tagged.json into the test folder.
+Structure of the Dataset directory should be as follows.
+
+```
+
+train
+|-- ...
+
+test
+|-- image
+|-- image-densepose
+|-- agnostic-mask
+|-- cloth
+|-- vitonhd_test_tagged.json
+
+```
+
+
+
+
+
+## Inference
+
+Inference with python file with argument.
+
+```
+accelerate launch inference.py \
+    --width 768 --height 1024 --num_inference_steps 30 \
+    --output_dir "result" \
+    --unpaired \
+    --data_dir "DATA_DIR" \
+    --seed 42 \
+    --test_batch_size 2 \
+    --guidance_scale 2.0
+```
+
+You can simply run with the script file.
+
+```
+sh inference.sh
+```
+
+
+
 ## Acknowledgements
 
-For the demo, GPUs are supported from [zerogpu](https://huggingface.co/zero-gpu-explorers), and auto masking generation codes are based on [OOTDiffusion](https://github.com/levihsu/OOTDiffusion)
+For the demo, GPUs are supported from [zerogpu](https://huggingface.co/zero-gpu-explorers), and auto masking generation codes are based on [OOTDiffusion](https://github.com/levihsu/OOTDiffusion).
+Parts of the code were based on [IP-Adapter](https://github.com/tencent-ailab/IP-Adapter).

environment.yaml

@@ -0,0 +1,23 @@
+name: idm
+channels:
+  - pytorch
+  - nvidia
+  - defaults
+dependencies:
+  - python=3.10.0=h12debd9_5
+  - pytorch=2.0.1=py3.10_cuda11.8_cudnn8.7.0_0
+  - pytorch-cuda=11.8=h7e8668a_5
+  - torchaudio=2.0.2=py310_cu118
+  - torchtriton=2.0.0=py310
+  - torchvision=0.15.2=py310_cu118
+  - pip=23.3.1=py310h06a4308_0
+
+  - pip:
+      - accelerate==0.25.0
+      - torchmetrics==1.2.1
+      - tqdm==4.66.1
+      - transformers==4.36.2
+      - diffusers==0.25.0
+      - einops==0.7.0
+      - bitsandbytes==0.39.0
+      - scipy==1.11.1