Set Space sdk_version to 4.24.0 for IDM-VTON app compatibility
Browse files
README.md
CHANGED
|
@@ -1,46 +1,46 @@
|
|
| 1 |
-
---
|
| 2 |
-
title: StyleSense AI VTON + AR
|
| 3 |
-
emoji: 👗
|
| 4 |
-
colorFrom: pink
|
| 5 |
-
colorTo: blue
|
| 6 |
-
sdk: gradio
|
| 7 |
-
sdk_version: "
|
| 8 |
-
python_version: "3.10"
|
| 9 |
-
app_file: app.py
|
| 10 |
-
pinned: false
|
| 11 |
-
---
|
| 12 |
-
|
| 13 |
-
# Hugging Face Space Starter Assets
|
| 14 |
-
|
| 15 |
-
This folder contains the AR overlay modules and dependency pins to apply on top of your duplicated IDM-VTON Space.
|
| 16 |
-
|
| 17 |
-
## Included
|
| 18 |
-
|
| 19 |
-
- pipeline/pose_estimator.py
|
| 20 |
-
- pipeline/cloth_warper.py
|
| 21 |
-
- requirements.txt
|
| 22 |
-
- packages.txt
|
| 23 |
-
|
| 24 |
-
## How to apply
|
| 25 |
-
|
| 26 |
-
Important: This package is a merge patch, not a full Space repository.
|
| 27 |
-
Do not replace the whole duplicated Space with only these files.
|
| 28 |
-
Keep the original `app.py` from the duplicated IDM-VTON Space (or provide your own `app.py` that calls `launch()`).
|
| 29 |
-
If `app.py` is missing or exits immediately, Space can fail with messages like "signal is aborted without reason" or exit code 0.
|
| 30 |
-
|
| 31 |
-
1. Duplicate yisol/IDM-VTON into your own Space.
|
| 32 |
-
2. Copy pipeline/\*.py into the Space pipeline folder.
|
| 33 |
-
3. Merge AR endpoint function into app.py and expose api_name as ar_overlay.
|
| 34 |
-
4. Update Space requirements.txt and packages.txt.
|
| 35 |
-
- Do not pin gradio in requirements.txt when using Gradio SDK Spaces unless README metadata sdk_version is pinned to the same version.
|
| 36 |
-
- Let the Space builder install gradio/spaces to avoid resolver conflicts.
|
| 37 |
-
5. Push to Hugging Face Space and wait for build.
|
| 38 |
-
|
| 39 |
-
## Endpoint contract
|
| 40 |
-
|
| 41 |
-
- POST /api/predict for AI try-on
|
| 42 |
-
- POST /api/ar_overlay for AR overlay
|
| 43 |
-
|
| 44 |
-
## Recommendation
|
| 45 |
-
|
| 46 |
-
Keep baseline and medium fine-tuned checkpoints as switchable options during rollout.
|
|
|
|
| 1 |
+
---
|
| 2 |
+
title: StyleSense AI VTON + AR
|
| 3 |
+
emoji: 👗
|
| 4 |
+
colorFrom: pink
|
| 5 |
+
colorTo: blue
|
| 6 |
+
sdk: gradio
|
| 7 |
+
sdk_version: "4.24.0"
|
| 8 |
+
python_version: "3.10"
|
| 9 |
+
app_file: app.py
|
| 10 |
+
pinned: false
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Hugging Face Space Starter Assets
|
| 14 |
+
|
| 15 |
+
This folder contains the AR overlay modules and dependency pins to apply on top of your duplicated IDM-VTON Space.
|
| 16 |
+
|
| 17 |
+
## Included
|
| 18 |
+
|
| 19 |
+
- pipeline/pose_estimator.py
|
| 20 |
+
- pipeline/cloth_warper.py
|
| 21 |
+
- requirements.txt
|
| 22 |
+
- packages.txt
|
| 23 |
+
|
| 24 |
+
## How to apply
|
| 25 |
+
|
| 26 |
+
Important: This package is a merge patch, not a full Space repository.
|
| 27 |
+
Do not replace the whole duplicated Space with only these files.
|
| 28 |
+
Keep the original `app.py` from the duplicated IDM-VTON Space (or provide your own `app.py` that calls `launch()`).
|
| 29 |
+
If `app.py` is missing or exits immediately, Space can fail with messages like "signal is aborted without reason" or exit code 0.
|
| 30 |
+
|
| 31 |
+
1. Duplicate yisol/IDM-VTON into your own Space.
|
| 32 |
+
2. Copy pipeline/\*.py into the Space pipeline folder.
|
| 33 |
+
3. Merge AR endpoint function into app.py and expose api_name as ar_overlay.
|
| 34 |
+
4. Update Space requirements.txt and packages.txt.
|
| 35 |
+
- Do not pin gradio in requirements.txt when using Gradio SDK Spaces unless README metadata sdk_version is pinned to the same version.
|
| 36 |
+
- Let the Space builder install gradio/spaces to avoid resolver conflicts.
|
| 37 |
+
5. Push to Hugging Face Space and wait for build.
|
| 38 |
+
|
| 39 |
+
## Endpoint contract
|
| 40 |
+
|
| 41 |
+
- POST /api/predict for AI try-on
|
| 42 |
+
- POST /api/ar_overlay for AR overlay
|
| 43 |
+
|
| 44 |
+
## Recommendation
|
| 45 |
+
|
| 46 |
+
Keep baseline and medium fine-tuned checkpoints as switchable options during rollout.
|