Update app.py
Browse files
app.py
CHANGED
@@ -84,10 +84,16 @@ def set_example_url(example: list) -> dict:
|
|
84 |
title = """<h1 id="title">Face Mask Detection with YOLOS</h1>"""
|
85 |
|
86 |
description = """
|
87 |
-
The model used in this space is the fine-tuned version of the COCO trained [hustlv/yolos-small](https://huggingface.co/hustlv/yolos-small). This fine-tuned model was trained for 200 epochs on the [face-mask-dataset]() from Kaggle which consisted of 853 images.
|
88 |
|
89 |
-
|
90 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
91 |
"""
|
92 |
|
93 |
models = ["nickmuchi/yolos-small-finetuned-masks"]
|
|
|
84 |
title = """<h1 id="title">Face Mask Detection with YOLOS</h1>"""
|
85 |
|
86 |
description = """
|
|
|
87 |
|
88 |
+
YOLOS is a Vision Transformer (ViT) trained using the DETR loss. Despite its simplicity, a base-sized YOLOS model is able to achieve 42 AP on COCO validation 2017 (similar to DETR and more complex frameworks such as Faster R-CNN).
|
89 |
+
|
90 |
+
The YOLOS model was fine-tuned on COCO 2017 object detection (118k annotated images). It was introduced in the paper [You Only Look at One Sequence: Rethinking Transformer in Vision through Object Detection](https://arxiv.org/abs/2106.00666) by Fang et al. and first released in [this repository](https://github.com/hustvl/YOLOS).
|
91 |
+
|
92 |
+
This model was further fine-tuned on the [face mask dataset]("https://www.kaggle.com/datasets/andrewmvd/face-mask-detection") from Kaggle. The dataset consists of 853 images of people with annotations categorised as "with mask","without mask" and "mask not worn correctly". The model was trained for 200 epochs on a single GPU.
|
93 |
+
|
94 |
+
Links to HuggingFace Models:
|
95 |
+
- [nickmuchi/yolos-small-finetuned-masks](https://huggingface.co/nickmuchi/yolos-small-finetuned-masks)
|
96 |
+
- [hustlv/yolos-small](https://huggingface.co/hustlv/yolos-small)
|
97 |
"""
|
98 |
|
99 |
models = ["nickmuchi/yolos-small-finetuned-masks"]
|