fix build - remove docker + simplify deps

Dockerfile

@@ -1,31 +0,0 @@
-# Dockerfile for Hugging Face Space with CUDA-enabled PyTorch
-# Uses an official PyTorch CUDA runtime image so torch + CUDA are already installed
-FROM pytorch/pytorch:2.5.1-cuda121-cudnn8-runtime
-
-WORKDIR /app
-
-# Copy project
-COPY . /app
-
-# Install system packages required by the repo and git-lfs
-RUN apt-get update && apt-get install -y \
-    git \
-    git-lfs \
-    ffmpeg \
-    libsm6 \
-    libxext6 \
-    libgl1 \
-    && rm -rf /var/lib/apt/lists/* \
-    && git lfs install
-
-# Upgrade pip
-RUN python -m pip install --upgrade pip
-
-# Install Python requirements but skip torch/torchvision/torchaudio (provided by base image)
-RUN sed -e '/^torch\b/d' -e '/^torchvision\b/d' -e '/^torchaudio\b/d' requirements.txt > /tmp/requirements_no_torch.txt \
-    && python -m pip install --no-cache-dir -r /tmp/requirements_no_torch.txt \
-    && python -m pip install --no-cache-dir gradio==6.13.0
-
-# Expose port and run the Gradio app
-EXPOSE 7860
-CMD ["python", "app.py"]

README.md

@@ -1,149 +1 @@
----
-title: Virtual Try-On
-emoji: 👕
-colorFrom: blue
-colorTo: pink
-sdk: gradio
-app_file: app.py
-pinned: false
-python_version: 3.10
----
-
-# FASHN VTON v1.5: Efficient Maskless Virtual Try-On in Pixel Space
-
-<div align="center">
-  <a href="https://fashn.ai/research/vton-1-5"><img src='https://img.shields.io/badge/Project-Page-1A1A1A?style=flat' alt='Project Page'></a>&ensp;
-  <a href='https://huggingface.co/fashn-ai/fashn-vton-1.5'><img src='https://img.shields.io/badge/Hugging%20Face-Model-FFD21E?style=flat&logo=HuggingFace&logoColor=FFD21E' alt='Hugging Face Model'></a>&ensp;
-  <a href="https://huggingface.co/spaces/fashn-ai/fashn-vton-1.5"><img src='https://img.shields.io/badge/Hugging%20Face-Spaces-FFD21E?style=flat&logo=HuggingFace&logoColor=FFD21E' alt='Hugging Face Spaces'></a>&ensp;
-  <a href=""><img src='https://img.shields.io/badge/arXiv-Coming%20Soon-b31b1b?style=flat&logo=arXiv&logoColor=b31b1b' alt='arXiv'></a>&ensp;
-  <a href="LICENSE"><img src='https://img.shields.io/badge/License-Apache--2.0-gray?style=flat' alt='License'></a>
-</div>
-
-by [FASHN AI](https://fashn.ai)
-
-Virtual try-on model that generates photorealistic images directly in pixel space without requiring segmentation masks.
-
-<p align="center">
-  <img src="https://static.fashn.ai/repositories/fashn-vton-v15/results/hero_collage.webp" alt="FASHN VTON v1.5 examples" width="900">
-</p>
-
-This repo contains minimal inference code to run virtual try-on with the FASHN VTON v1.5 model weights. Given a person image and a garment image, the model generates a photorealistic image of the person wearing the garment. Supports both model photos and flat-lay product shots as garment inputs.
-
----
-
-## Local Installation
-
-We recommend using a virtual environment:
-
-```bash
-git clone https://github.com/fashn-AI/fashn-vton-1.5.git
-cd fashn-vton-1.5
-python -m venv .venv && source .venv/bin/activate
-pip install -e .
-```
-
-**Note:** Installation includes `onnxruntime-gpu` for GPU-accelerated pose detection. Ensure CUDA is properly configured on your system. For CPU-only environments, replace with the CPU version:
-
-```bash
-pip uninstall onnxruntime-gpu && pip install onnxruntime
-```
-
----
-
-## Model Weights
-
-Download the required model weights (~2 GB total):
-
-```bash
-python scripts/download_weights.py --weights-dir ./weights
-```
-
-This downloads:
-- `model.safetensors` — TryOnModel weights from [HuggingFace](https://huggingface.co/fashn-ai/fashn-vton-1.5)
-- `dwpose/` — DWPose ONNX models for pose detection
-
-**Note:** The human parser weights (~244 MB) are automatically downloaded on first use to the HuggingFace cache folder. Set `HF_HOME` to customize the location.
-
----
-
-## Usage
-
-```python
-from fashn_vton import TryOnPipeline
-from PIL import Image
-
-# Initialize pipeline (automatically uses GPU if available)
-pipeline = TryOnPipeline(weights_dir="./weights")
-
-# Load images
-person = Image.open("examples/data/model.webp").convert("RGB")
-garment = Image.open("examples/data/garment.webp").convert("RGB")
-
-# Run inference
-result = pipeline(
-    person_image=person,
-    garment_image=garment,
-    category="tops",  # "tops" | "bottoms" | "one-pieces"
-)
-
-# Save output
-result.images[0].save("output.png")
-```
-
-### CLI
-
-```bash
-python examples/basic_inference.py \
-    --weights-dir ./weights \
-    --person-image examples/data/model.webp \
-    --garment-image examples/data/garment.webp \
-    --category tops
-```
-
-**Note:** The pipeline automatically uses GPU if available. The try-on model weights are stored in bfloat16 and will run in bf16 precision on Ampere+ GPUs (RTX 30xx/40xx, A100, H100). On older hardware or CPU, weights are converted to float32.
-
-See [`examples/basic_inference.py`](examples/basic_inference.py) for additional options.
-
----
-
-## Categories
-
-| Category | Description |
-|----------|-------------|
-| `tops` | Upper body: t-shirts, blouses, jackets |
-| `bottoms` | Lower body: pants, skirts, shorts |
-| `one-pieces` | Full body: dresses, jumpsuits |
-
----
-
-## API
-
-FASHN provides a suite of [fashion AI APIs](https://fashn.ai/products/api) including virtual try-on, model generation, image-to-video, and more. See the [docs](https://docs.fashn.ai/) to get started.
-
----
-
-## Citation
-
-If you use FASHN VTON v1.5 in your research, please cite:
-
-```bibtex
-@article{bochman2026fashnvton,
-  title={FASHN VTON v1.5: Efficient Maskless Virtual Try-On in Pixel Space},
-  author={Bochman, Dan and Bochman, Aya},
-  journal={arXiv preprint},
-  year={2026},
-  note={Paper coming soon}
-}
-```
-
----
-
-## License
-
-Apache-2.0. See [LICENSE](LICENSE) for details.
-
-**Third-party components:**
-- [DWPose](https://github.com/IDEA-Research/DWPose) (Apache-2.0)
-- [YOLOX](https://github.com/Megvii-BaseDetection/YOLOX) (Apache-2.0)
-- [fashn-human-parser](https://github.com/fashn-AI/fashn-human-parser) ([License](https://github.com/fashn-AI/fashn-human-parser?tab=readme-ov-file#license))
-
+python_version: 3.10

requirements.txt

@@ -1,78 +1,11 @@
-annotated-doc==0.0.4
-annotated-types==0.7.0
-anyio==4.13.0
-brotli==1.2.0
-certifi==2026.4.22
-click==8.3.3
-colorama==0.4.6
-coloredlogs==15.0.1
-contourpy==1.3.2
-cycler==0.12.1
-einops==0.8.2
-exceptiongroup==1.3.1
-fashn-human-parser==0.1.1
--e git+https://github.com/fashn-AI/fashn-vton-1.5.git@7c0f10af3f91ad4048fe9729c470a13ef905d25a#egg=fashn_vton
-fastapi==0.136.1
-filelock==3.25.2
-flatbuffers==25.12.19
-fonttools==4.62.1
-fsspec==2026.2.0
-gradio==6.13.0
-gradio_client==2.5.0
-groovy==0.1.2
-h11==0.16.0
-hf-gradio==0.4.1
-hf-xet==1.4.3
-httpcore==1.0.9
-httpx==0.28.1
-huggingface_hub==1.11.0
-humanfriendly==10.0
-idna==3.13
-Jinja2==3.1.6
-kiwisolver==1.5.0
-markdown-it-py==4.0.0
-MarkupSafe==3.0.3
-matplotlib==3.10.9
-mdurl==0.1.2
-mpmath==1.3.0
-networkx==3.4.2
-numpy==2.2.6
-onnxruntime==1.20.0
-opencv-python==4.13.0.92
-orjson==3.11.8
-packaging==26.0
-pandas==2.3.3
-pillow==12.1.1
-protobuf==7.34.1
-pydantic==2.13.3
-pydantic_core==2.46.3
-pydub==0.25.1
-Pygments==2.20.0
-pyparsing==3.3.2
-pyreadline3==3.5.4
-python-dateutil==2.9.0.post0
-python-multipart==0.0.26
-pytz==2026.1.post1
-PyYAML==6.0.3
-regex==2026.4.4
-rich==15.0.0
-safehttpx==0.1.7
-safetensors==0.7.0
-semantic-version==2.10.0
-shellingham==1.5.4
-six==1.17.0
-starlette==1.0.0
-sympy==1.13.1
-tokenizers==0.22.2
-tomlkit==0.14.0
 torch==2.2.2
 torchvision
-torchaudio==2.5.1+cu121
-torchvision==0.20.1+cu121
-tqdm==4.67.3
-transformers==5.6.2
-typer==0.24.2
-typing-inspection==0.4.2
-typing_extensions==4.15.0
-tzdata==2026.1
-uvicorn==0.46.0
+diffusers
+transformers
+accelerate
+safetensors
+pillow
+opencv-python
+gradio
+onnxruntime
+git+https://github.com/fashn-AI/fashn-vton-1.5.git