Spaces:
Running
Running
Update
Browse files- .pre-commit-config.yaml +59 -35
- .vscode/settings.json +30 -0
- app.py +34 -37
- model.py +44 -59
- palette.py +10 -7
- style.css +6 -2
.pre-commit-config.yaml
CHANGED
@@ -1,37 +1,61 @@
|
|
1 |
exclude: ^patch
|
2 |
repos:
|
3 |
-
- repo: https://github.com/pre-commit/pre-commit-hooks
|
4 |
-
|
5 |
-
|
6 |
-
|
7 |
-
|
8 |
-
|
9 |
-
|
10 |
-
|
11 |
-
|
12 |
-
|
13 |
-
|
14 |
-
|
15 |
-
|
16 |
-
|
17 |
-
-
|
18 |
-
|
19 |
-
|
20 |
-
|
21 |
-
|
22 |
-
|
23 |
-
|
24 |
-
|
25 |
-
|
26 |
-
|
27 |
-
- repo: https://github.com/pre-commit/mirrors-mypy
|
28 |
-
|
29 |
-
|
30 |
-
|
31 |
-
|
32 |
-
|
33 |
-
|
34 |
-
|
35 |
-
|
36 |
-
|
37 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
exclude: ^patch
|
2 |
repos:
|
3 |
+
- repo: https://github.com/pre-commit/pre-commit-hooks
|
4 |
+
rev: v4.6.0
|
5 |
+
hooks:
|
6 |
+
- id: check-executables-have-shebangs
|
7 |
+
- id: check-json
|
8 |
+
- id: check-merge-conflict
|
9 |
+
- id: check-shebang-scripts-are-executable
|
10 |
+
- id: check-toml
|
11 |
+
- id: check-yaml
|
12 |
+
- id: end-of-file-fixer
|
13 |
+
- id: mixed-line-ending
|
14 |
+
args: ["--fix=lf"]
|
15 |
+
- id: requirements-txt-fixer
|
16 |
+
- id: trailing-whitespace
|
17 |
+
- repo: https://github.com/myint/docformatter
|
18 |
+
rev: v1.7.5
|
19 |
+
hooks:
|
20 |
+
- id: docformatter
|
21 |
+
args: ["--in-place"]
|
22 |
+
- repo: https://github.com/pycqa/isort
|
23 |
+
rev: 5.13.2
|
24 |
+
hooks:
|
25 |
+
- id: isort
|
26 |
+
args: ["--profile", "black"]
|
27 |
+
- repo: https://github.com/pre-commit/mirrors-mypy
|
28 |
+
rev: v1.10.0
|
29 |
+
hooks:
|
30 |
+
- id: mypy
|
31 |
+
args: ["--ignore-missing-imports"]
|
32 |
+
additional_dependencies:
|
33 |
+
[
|
34 |
+
"types-python-slugify",
|
35 |
+
"types-requests",
|
36 |
+
"types-PyYAML",
|
37 |
+
"types-pytz",
|
38 |
+
]
|
39 |
+
- repo: https://github.com/psf/black
|
40 |
+
rev: 24.4.2
|
41 |
+
hooks:
|
42 |
+
- id: black
|
43 |
+
language_version: python3.10
|
44 |
+
args: ["--line-length", "119"]
|
45 |
+
- repo: https://github.com/kynan/nbstripout
|
46 |
+
rev: 0.7.1
|
47 |
+
hooks:
|
48 |
+
- id: nbstripout
|
49 |
+
args:
|
50 |
+
[
|
51 |
+
"--extra-keys",
|
52 |
+
"metadata.interpreter metadata.kernelspec cell.metadata.pycharm",
|
53 |
+
]
|
54 |
+
- repo: https://github.com/nbQA-dev/nbQA
|
55 |
+
rev: 1.8.5
|
56 |
+
hooks:
|
57 |
+
- id: nbqa-black
|
58 |
+
- id: nbqa-pyupgrade
|
59 |
+
args: ["--py37-plus"]
|
60 |
+
- id: nbqa-isort
|
61 |
+
args: ["--float-to-top"]
|
.vscode/settings.json
ADDED
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"editor.formatOnSave": true,
|
3 |
+
"files.insertFinalNewline": false,
|
4 |
+
"[python]": {
|
5 |
+
"editor.defaultFormatter": "ms-python.black-formatter",
|
6 |
+
"editor.formatOnType": true,
|
7 |
+
"editor.codeActionsOnSave": {
|
8 |
+
"source.organizeImports": "explicit"
|
9 |
+
}
|
10 |
+
},
|
11 |
+
"[jupyter]": {
|
12 |
+
"files.insertFinalNewline": false
|
13 |
+
},
|
14 |
+
"black-formatter.args": [
|
15 |
+
"--line-length=119"
|
16 |
+
],
|
17 |
+
"isort.args": ["--profile", "black"],
|
18 |
+
"flake8.args": [
|
19 |
+
"--max-line-length=119"
|
20 |
+
],
|
21 |
+
"ruff.lint.args": [
|
22 |
+
"--line-length=119"
|
23 |
+
],
|
24 |
+
"notebook.output.scrolling": true,
|
25 |
+
"notebook.formatOnCellExecution": true,
|
26 |
+
"notebook.formatOnSave.enabled": true,
|
27 |
+
"notebook.codeActionsOnSave": {
|
28 |
+
"source.organizeImports": "explicit"
|
29 |
+
}
|
30 |
+
}
|
app.py
CHANGED
@@ -8,60 +8,57 @@ import gradio as gr
|
|
8 |
|
9 |
from model import Model
|
10 |
|
11 |
-
DESCRIPTION =
|
12 |
|
13 |
model = Model()
|
14 |
|
15 |
-
with gr.Blocks(css=
|
16 |
gr.Markdown(DESCRIPTION)
|
17 |
|
18 |
with gr.Row():
|
19 |
with gr.Column():
|
20 |
with gr.Row():
|
21 |
-
input_image = gr.Image(label=
|
22 |
with gr.Row():
|
23 |
-
detector_name = gr.Dropdown(
|
24 |
-
|
25 |
-
|
26 |
with gr.Row():
|
27 |
-
detect_button = gr.Button(
|
28 |
detection_results = gr.Variable()
|
29 |
with gr.Column():
|
30 |
with gr.Row():
|
31 |
-
detection_visualization = gr.Image(label=
|
32 |
-
type='numpy')
|
33 |
with gr.Row():
|
34 |
visualization_score_threshold = gr.Slider(
|
35 |
-
label=
|
36 |
-
|
37 |
-
maximum=1,
|
38 |
-
step=0.05,
|
39 |
-
value=0.3)
|
40 |
with gr.Row():
|
41 |
-
redraw_button = gr.Button(
|
42 |
|
43 |
with gr.Row():
|
44 |
-
paths = sorted(pathlib.Path(
|
45 |
-
gr.Examples(examples=[[path.as_posix()] for path in paths],
|
46 |
-
inputs=input_image)
|
47 |
|
48 |
-
detector_name.change(fn=model.set_model_name,
|
49 |
-
|
50 |
-
|
51 |
-
|
52 |
-
|
53 |
-
|
54 |
-
|
55 |
-
|
56 |
-
|
57 |
-
|
58 |
-
|
59 |
-
|
60 |
-
redraw_button.click(
|
61 |
-
|
62 |
-
|
63 |
-
|
64 |
-
|
65 |
-
|
66 |
-
|
|
|
|
|
67 |
demo.queue(max_size=10).launch()
|
|
|
8 |
|
9 |
from model import Model
|
10 |
|
11 |
+
DESCRIPTION = "# [CBNetV2](https://github.com/VDIGPKU/CBNetV2)"
|
12 |
|
13 |
model = Model()
|
14 |
|
15 |
+
with gr.Blocks(css="style.css") as demo:
|
16 |
gr.Markdown(DESCRIPTION)
|
17 |
|
18 |
with gr.Row():
|
19 |
with gr.Column():
|
20 |
with gr.Row():
|
21 |
+
input_image = gr.Image(label="Input Image", type="numpy")
|
22 |
with gr.Row():
|
23 |
+
detector_name = gr.Dropdown(
|
24 |
+
label="Detector", choices=list(model.models.keys()), value=model.model_name
|
25 |
+
)
|
26 |
with gr.Row():
|
27 |
+
detect_button = gr.Button("Detect")
|
28 |
detection_results = gr.Variable()
|
29 |
with gr.Column():
|
30 |
with gr.Row():
|
31 |
+
detection_visualization = gr.Image(label="Detection Result", type="numpy")
|
|
|
32 |
with gr.Row():
|
33 |
visualization_score_threshold = gr.Slider(
|
34 |
+
label="Visualization Score Threshold", minimum=0, maximum=1, step=0.05, value=0.3
|
35 |
+
)
|
|
|
|
|
|
|
36 |
with gr.Row():
|
37 |
+
redraw_button = gr.Button("Redraw")
|
38 |
|
39 |
with gr.Row():
|
40 |
+
paths = sorted(pathlib.Path("images").rglob("*.jpg"))
|
41 |
+
gr.Examples(examples=[[path.as_posix()] for path in paths], inputs=input_image)
|
|
|
42 |
|
43 |
+
detector_name.change(fn=model.set_model_name, inputs=[detector_name], outputs=None)
|
44 |
+
detect_button.click(
|
45 |
+
fn=model.detect_and_visualize,
|
46 |
+
inputs=[
|
47 |
+
input_image,
|
48 |
+
visualization_score_threshold,
|
49 |
+
],
|
50 |
+
outputs=[
|
51 |
+
detection_results,
|
52 |
+
detection_visualization,
|
53 |
+
],
|
54 |
+
)
|
55 |
+
redraw_button.click(
|
56 |
+
fn=model.visualize_detection_results,
|
57 |
+
inputs=[
|
58 |
+
input_image,
|
59 |
+
detection_results,
|
60 |
+
visualization_score_threshold,
|
61 |
+
],
|
62 |
+
outputs=[detection_visualization],
|
63 |
+
)
|
64 |
demo.queue(max_size=10).launch()
|
model.py
CHANGED
@@ -6,26 +6,26 @@ import shlex
|
|
6 |
import subprocess
|
7 |
import sys
|
8 |
|
9 |
-
if os.getenv(
|
10 |
import mim
|
11 |
|
12 |
-
mim.uninstall(
|
13 |
-
mim.install(
|
14 |
|
15 |
-
subprocess.run(shlex.split(
|
16 |
-
subprocess.run(shlex.split(
|
17 |
-
subprocess.run(shlex.split(
|
18 |
|
19 |
-
with open(
|
20 |
-
subprocess.run(shlex.split(
|
21 |
-
subprocess.run(
|
22 |
|
23 |
import numpy as np
|
24 |
import torch
|
25 |
import torch.nn as nn
|
26 |
|
27 |
app_dir = pathlib.Path(__file__).parent
|
28 |
-
submodule_dir = app_dir /
|
29 |
sys.path.insert(0, submodule_dir.as_posix())
|
30 |
|
31 |
from mmdet.apis import inference_detector, init_detector
|
@@ -33,24 +33,19 @@ from mmdet.apis import inference_detector, init_detector
|
|
33 |
|
34 |
class Model:
|
35 |
def __init__(self):
|
36 |
-
self.device = torch.device(
|
37 |
-
'cuda:0' if torch.cuda.is_available() else 'cpu')
|
38 |
self.models = self._load_models()
|
39 |
-
self.model_name =
|
40 |
|
41 |
def _load_models(self) -> dict[str, nn.Module]:
|
42 |
model_dict = {
|
43 |
-
|
44 |
-
|
45 |
-
|
46 |
-
'model':
|
47 |
-
'https://github.com/CBNetwork/storage/releases/download/v1.0.0/faster_rcnn_cbv2d1_r50_fpn_1x_coco.pth.zip',
|
48 |
},
|
49 |
-
|
50 |
-
|
51 |
-
|
52 |
-
'model':
|
53 |
-
'https://github.com/CBNetwork/storage/releases/download/v1.0.0/mask_rcnn_cbv2_swin_tiny_patch4_window7_mstrain_480-800_adamw_3x_coco.pth.zip',
|
54 |
},
|
55 |
# 'Cascade Mask R-CNN (DB-Swin-S)': {
|
56 |
# 'config':
|
@@ -58,34 +53,28 @@ class Model:
|
|
58 |
# 'model':
|
59 |
# 'https://github.com/CBNetwork/storage/releases/download/v1.0.0/cascade_mask_rcnn_cbv2_swin_small_patch4_window7_mstrain_400-1400_adamw_3x_coco.pth.zip',
|
60 |
# },
|
61 |
-
|
62 |
-
|
63 |
-
|
64 |
-
'model':
|
65 |
-
'https://github.com/CBNetwork/storage/releases/download/v1.0.0/htc_cbv2_swin_base22k_patch4_window7_mstrain_400-1400_giou_4conv1f_adamw_20e_coco.pth.zip',
|
66 |
},
|
67 |
-
|
68 |
-
|
69 |
-
|
70 |
-
'model':
|
71 |
-
'https://github.com/CBNetwork/storage/releases/download/v1.0.0/htc_cbv2_swin_large22k_patch4_window7_mstrain_400-1400_giou_4conv1f_adamw_1x_coco.pth.zip',
|
72 |
},
|
73 |
-
|
74 |
-
|
75 |
-
|
76 |
-
'model':
|
77 |
-
'https://github.com/CBNetwork/storage/releases/download/v1.0.0/htc_cbv2_swin_large22k_patch4_window7_mstrain_400-1400_giou_4conv1f_adamw_1x_coco.pth.zip',
|
78 |
},
|
79 |
}
|
80 |
|
81 |
-
weight_dir = pathlib.Path(
|
82 |
weight_dir.mkdir(exist_ok=True)
|
83 |
|
84 |
def _download(model_name: str, out_dir: pathlib.Path) -> None:
|
85 |
import zipfile
|
86 |
|
87 |
-
model_url = model_dict[model_name][
|
88 |
-
zip_name = model_url.split(
|
89 |
|
90 |
out_path = out_dir / zip_name
|
91 |
if out_path.exists():
|
@@ -96,17 +85,15 @@ class Model:
|
|
96 |
f.extractall(out_dir)
|
97 |
|
98 |
def _get_model_path(model_name: str) -> str:
|
99 |
-
model_url = model_dict[model_name][
|
100 |
-
model_name = model_url.split(
|
101 |
return (weight_dir / model_name).as_posix()
|
102 |
|
103 |
for model_name in model_dict:
|
104 |
_download(model_name, weight_dir)
|
105 |
|
106 |
models = {
|
107 |
-
key: init_detector(dic[
|
108 |
-
_get_model_path(key),
|
109 |
-
device=self.device)
|
110 |
for key, dic in model_dict.items()
|
111 |
}
|
112 |
return models
|
@@ -114,9 +101,7 @@ class Model:
|
|
114 |
def set_model_name(self, name: str) -> None:
|
115 |
self.model_name = name
|
116 |
|
117 |
-
def detect_and_visualize(
|
118 |
-
self, image: np.ndarray,
|
119 |
-
score_threshold: float) -> tuple[list[np.ndarray], np.ndarray]:
|
120 |
out = self.detect(image)
|
121 |
vis = self.visualize_detection_results(image, out, score_threshold)
|
122 |
return out, vis
|
@@ -128,16 +113,16 @@ class Model:
|
|
128 |
return out
|
129 |
|
130 |
def visualize_detection_results(
|
131 |
-
|
132 |
-
|
133 |
-
detection_results: list[np.ndarray],
|
134 |
-
score_threshold: float = 0.3) -> np.ndarray:
|
135 |
image = image[:, :, ::-1] # RGB -> BGR
|
136 |
model = self.models[self.model_name]
|
137 |
-
vis = model.show_result(
|
138 |
-
|
139 |
-
|
140 |
-
|
141 |
-
|
142 |
-
|
|
|
|
|
143 |
return vis[:, :, ::-1] # BGR -> RGB
|
|
|
6 |
import subprocess
|
7 |
import sys
|
8 |
|
9 |
+
if os.getenv("SYSTEM") == "spaces":
|
10 |
import mim
|
11 |
|
12 |
+
mim.uninstall("mmcv-full", confirm_yes=True)
|
13 |
+
mim.install("mmcv-full==1.5.0", is_yes=True)
|
14 |
|
15 |
+
subprocess.run(shlex.split("pip uninstall -y opencv-python"))
|
16 |
+
subprocess.run(shlex.split("pip uninstall -y opencv-python-headless"))
|
17 |
+
subprocess.run(shlex.split("pip install opencv-python-headless==4.8.0.74"))
|
18 |
|
19 |
+
with open("patch") as f:
|
20 |
+
subprocess.run(shlex.split("patch -p1"), cwd="CBNetV2", stdin=f)
|
21 |
+
subprocess.run("mv palette.py CBNetV2/mmdet/core/visualization/".split())
|
22 |
|
23 |
import numpy as np
|
24 |
import torch
|
25 |
import torch.nn as nn
|
26 |
|
27 |
app_dir = pathlib.Path(__file__).parent
|
28 |
+
submodule_dir = app_dir / "CBNetV2/"
|
29 |
sys.path.insert(0, submodule_dir.as_posix())
|
30 |
|
31 |
from mmdet.apis import inference_detector, init_detector
|
|
|
33 |
|
34 |
class Model:
|
35 |
def __init__(self):
|
36 |
+
self.device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
|
|
|
37 |
self.models = self._load_models()
|
38 |
+
self.model_name = "Improved HTC (DB-Swin-B)"
|
39 |
|
40 |
def _load_models(self) -> dict[str, nn.Module]:
|
41 |
model_dict = {
|
42 |
+
"Faster R-CNN (DB-ResNet50)": {
|
43 |
+
"config": "CBNetV2/configs/cbnet/faster_rcnn_cbv2d1_r50_fpn_1x_coco.py",
|
44 |
+
"model": "https://github.com/CBNetwork/storage/releases/download/v1.0.0/faster_rcnn_cbv2d1_r50_fpn_1x_coco.pth.zip",
|
|
|
|
|
45 |
},
|
46 |
+
"Mask R-CNN (DB-Swin-T)": {
|
47 |
+
"config": "CBNetV2/configs/cbnet/mask_rcnn_cbv2_swin_tiny_patch4_window7_mstrain_480-800_adamw_3x_coco.py",
|
48 |
+
"model": "https://github.com/CBNetwork/storage/releases/download/v1.0.0/mask_rcnn_cbv2_swin_tiny_patch4_window7_mstrain_480-800_adamw_3x_coco.pth.zip",
|
|
|
|
|
49 |
},
|
50 |
# 'Cascade Mask R-CNN (DB-Swin-S)': {
|
51 |
# 'config':
|
|
|
53 |
# 'model':
|
54 |
# 'https://github.com/CBNetwork/storage/releases/download/v1.0.0/cascade_mask_rcnn_cbv2_swin_small_patch4_window7_mstrain_400-1400_adamw_3x_coco.pth.zip',
|
55 |
# },
|
56 |
+
"Improved HTC (DB-Swin-B)": {
|
57 |
+
"config": "CBNetV2/configs/cbnet/htc_cbv2_swin_base_patch4_window7_mstrain_400-1400_giou_4conv1f_adamw_20e_coco.py",
|
58 |
+
"model": "https://github.com/CBNetwork/storage/releases/download/v1.0.0/htc_cbv2_swin_base22k_patch4_window7_mstrain_400-1400_giou_4conv1f_adamw_20e_coco.pth.zip",
|
|
|
|
|
59 |
},
|
60 |
+
"Improved HTC (DB-Swin-L)": {
|
61 |
+
"config": "CBNetV2/configs/cbnet/htc_cbv2_swin_large_patch4_window7_mstrain_400-1400_giou_4conv1f_adamw_1x_coco.py",
|
62 |
+
"model": "https://github.com/CBNetwork/storage/releases/download/v1.0.0/htc_cbv2_swin_large22k_patch4_window7_mstrain_400-1400_giou_4conv1f_adamw_1x_coco.pth.zip",
|
|
|
|
|
63 |
},
|
64 |
+
"Improved HTC (DB-Swin-L (TTA))": {
|
65 |
+
"config": "CBNetV2/configs/cbnet/htc_cbv2_swin_large_patch4_window7_mstrain_400-1400_giou_4conv1f_adamw_1x_coco.py",
|
66 |
+
"model": "https://github.com/CBNetwork/storage/releases/download/v1.0.0/htc_cbv2_swin_large22k_patch4_window7_mstrain_400-1400_giou_4conv1f_adamw_1x_coco.pth.zip",
|
|
|
|
|
67 |
},
|
68 |
}
|
69 |
|
70 |
+
weight_dir = pathlib.Path("weights")
|
71 |
weight_dir.mkdir(exist_ok=True)
|
72 |
|
73 |
def _download(model_name: str, out_dir: pathlib.Path) -> None:
|
74 |
import zipfile
|
75 |
|
76 |
+
model_url = model_dict[model_name]["model"]
|
77 |
+
zip_name = model_url.split("/")[-1]
|
78 |
|
79 |
out_path = out_dir / zip_name
|
80 |
if out_path.exists():
|
|
|
85 |
f.extractall(out_dir)
|
86 |
|
87 |
def _get_model_path(model_name: str) -> str:
|
88 |
+
model_url = model_dict[model_name]["model"]
|
89 |
+
model_name = model_url.split("/")[-1][:-4]
|
90 |
return (weight_dir / model_name).as_posix()
|
91 |
|
92 |
for model_name in model_dict:
|
93 |
_download(model_name, weight_dir)
|
94 |
|
95 |
models = {
|
96 |
+
key: init_detector(dic["config"], _get_model_path(key), device=self.device)
|
|
|
|
|
97 |
for key, dic in model_dict.items()
|
98 |
}
|
99 |
return models
|
|
|
101 |
def set_model_name(self, name: str) -> None:
|
102 |
self.model_name = name
|
103 |
|
104 |
+
def detect_and_visualize(self, image: np.ndarray, score_threshold: float) -> tuple[list[np.ndarray], np.ndarray]:
|
|
|
|
|
105 |
out = self.detect(image)
|
106 |
vis = self.visualize_detection_results(image, out, score_threshold)
|
107 |
return out, vis
|
|
|
113 |
return out
|
114 |
|
115 |
def visualize_detection_results(
|
116 |
+
self, image: np.ndarray, detection_results: list[np.ndarray], score_threshold: float = 0.3
|
117 |
+
) -> np.ndarray:
|
|
|
|
|
118 |
image = image[:, :, ::-1] # RGB -> BGR
|
119 |
model = self.models[self.model_name]
|
120 |
+
vis = model.show_result(
|
121 |
+
image,
|
122 |
+
detection_results,
|
123 |
+
score_thr=score_threshold,
|
124 |
+
bbox_color=None,
|
125 |
+
text_color=(200, 200, 200),
|
126 |
+
mask_color=None,
|
127 |
+
)
|
128 |
return vis[:, :, ::-1] # BGR -> RGB
|
palette.py
CHANGED
@@ -208,6 +208,7 @@ Copyright 2018-2023 OpenMMLab. All rights reserved.
|
|
208 |
limitations under the License.
|
209 |
```
|
210 |
"""
|
|
|
211 |
# Copyright (c) OpenMMLab. All rights reserved.
|
212 |
import mmcv
|
213 |
import numpy as np
|
@@ -245,29 +246,31 @@ def get_palette(palette, num_classes):
|
|
245 |
dataset_palette = palette
|
246 |
elif isinstance(palette, tuple):
|
247 |
dataset_palette = [palette] * num_classes
|
248 |
-
elif palette ==
|
249 |
state = np.random.get_state()
|
250 |
# random color
|
251 |
np.random.seed(42)
|
252 |
palette = np.random.randint(0, 256, size=(num_classes, 3))
|
253 |
np.random.set_state(state)
|
254 |
dataset_palette = [tuple(c) for c in palette]
|
255 |
-
elif palette ==
|
256 |
from mmdet.datasets import CocoDataset, CocoPanopticDataset
|
|
|
257 |
dataset_palette = CocoDataset.PALETTE
|
258 |
if len(dataset_palette) < num_classes:
|
259 |
dataset_palette = CocoPanopticDataset.PALETTE
|
260 |
-
elif palette ==
|
261 |
from mmdet.datasets import CityscapesDataset
|
|
|
262 |
dataset_palette = CityscapesDataset.PALETTE
|
263 |
-
elif palette ==
|
264 |
from mmdet.datasets import VOCDataset
|
|
|
265 |
dataset_palette = VOCDataset.PALETTE
|
266 |
elif mmcv.is_str(palette):
|
267 |
dataset_palette = [mmcv.color_val(palette)[::-1]] * num_classes
|
268 |
else:
|
269 |
-
raise TypeError(f
|
270 |
|
271 |
-
assert len(dataset_palette) >= num_classes,
|
272 |
-
'The length of palette should not be less than `num_classes`.'
|
273 |
return dataset_palette
|
|
|
208 |
limitations under the License.
|
209 |
```
|
210 |
"""
|
211 |
+
|
212 |
# Copyright (c) OpenMMLab. All rights reserved.
|
213 |
import mmcv
|
214 |
import numpy as np
|
|
|
246 |
dataset_palette = palette
|
247 |
elif isinstance(palette, tuple):
|
248 |
dataset_palette = [palette] * num_classes
|
249 |
+
elif palette == "random" or palette is None:
|
250 |
state = np.random.get_state()
|
251 |
# random color
|
252 |
np.random.seed(42)
|
253 |
palette = np.random.randint(0, 256, size=(num_classes, 3))
|
254 |
np.random.set_state(state)
|
255 |
dataset_palette = [tuple(c) for c in palette]
|
256 |
+
elif palette == "coco":
|
257 |
from mmdet.datasets import CocoDataset, CocoPanopticDataset
|
258 |
+
|
259 |
dataset_palette = CocoDataset.PALETTE
|
260 |
if len(dataset_palette) < num_classes:
|
261 |
dataset_palette = CocoPanopticDataset.PALETTE
|
262 |
+
elif palette == "citys":
|
263 |
from mmdet.datasets import CityscapesDataset
|
264 |
+
|
265 |
dataset_palette = CityscapesDataset.PALETTE
|
266 |
+
elif palette == "voc":
|
267 |
from mmdet.datasets import VOCDataset
|
268 |
+
|
269 |
dataset_palette = VOCDataset.PALETTE
|
270 |
elif mmcv.is_str(palette):
|
271 |
dataset_palette = [mmcv.color_val(palette)[::-1]] * num_classes
|
272 |
else:
|
273 |
+
raise TypeError(f"Invalid type for palette: {type(palette)}")
|
274 |
|
275 |
+
assert len(dataset_palette) >= num_classes, "The length of palette should not be less than `num_classes`."
|
|
|
276 |
return dataset_palette
|
style.css
CHANGED
@@ -1,7 +1,11 @@
|
|
1 |
h1 {
|
2 |
text-align: center;
|
3 |
-
}
|
4 |
-
img#visitor-badge {
|
5 |
display: block;
|
|
|
|
|
|
|
6 |
margin: auto;
|
|
|
|
|
|
|
7 |
}
|
|
|
1 |
h1 {
|
2 |
text-align: center;
|
|
|
|
|
3 |
display: block;
|
4 |
+
}
|
5 |
+
|
6 |
+
#duplicate-button {
|
7 |
margin: auto;
|
8 |
+
color: #fff;
|
9 |
+
background: #1565c0;
|
10 |
+
border-radius: 100vh;
|
11 |
}
|