Wanli
commited on
Commit
·
c920270
1
Parent(s):
b7e21c2
update accuracy evaluation scripts (#184)
Browse files* update accuracy evaluation scripts
* remove labels of image classification
- tools/eval/README.md +2 -2
- tools/eval/eval.py +15 -11
tools/eval/README.md
CHANGED
@@ -5,7 +5,7 @@ Make sure you have the following packages installed:
|
|
5 |
```shell
|
6 |
pip install tqdm
|
7 |
pip install scikit-learn
|
8 |
-
pip install scipy
|
9 |
```
|
10 |
|
11 |
Generally speaking, evaluation can be done with the following command:
|
@@ -27,7 +27,7 @@ Supported datasets:
|
|
27 |
|
28 |
### Prepare data
|
29 |
|
30 |
-
Please visit https://image-net.org/ to download the ImageNet dataset and [the labels from caffe](http://dl.caffe.berkeleyvision.org/caffe_ilsvrc12.tar.gz). Organize files as follow:
|
31 |
|
32 |
```shell
|
33 |
$ tree -L 2 /path/to/imagenet
|
|
|
5 |
```shell
|
6 |
pip install tqdm
|
7 |
pip install scikit-learn
|
8 |
+
pip install scipy==1.8.1
|
9 |
```
|
10 |
|
11 |
Generally speaking, evaluation can be done with the following command:
|
|
|
27 |
|
28 |
### Prepare data
|
29 |
|
30 |
+
Please visit https://image-net.org/ to download the ImageNet dataset (only need images in `ILSVRC/Data/CLS-LOC/val`) and [the labels from caffe](http://dl.caffe.berkeleyvision.org/caffe_ilsvrc12.tar.gz). Organize files as follow:
|
31 |
|
32 |
```shell
|
33 |
$ tree -L 2 /path/to/imagenet
|
tools/eval/eval.py
CHANGED
@@ -22,24 +22,24 @@ args = parser.parse_args()
|
|
22 |
|
23 |
models = dict(
|
24 |
mobilenetv1=dict(
|
25 |
-
name="
|
26 |
topic="image_classification",
|
27 |
modelPath=os.path.join(root_dir, "models/image_classification_mobilenet/image_classification_mobilenetv1_2022apr.onnx"),
|
28 |
topK=5),
|
29 |
mobilenetv1_q=dict(
|
30 |
-
name="
|
31 |
topic="image_classification",
|
32 |
-
modelPath=os.path.join(root_dir, "models/image_classification_mobilenet/
|
33 |
topK=5),
|
34 |
mobilenetv2=dict(
|
35 |
-
name="
|
36 |
topic="image_classification",
|
37 |
modelPath=os.path.join(root_dir, "models/image_classification_mobilenet/image_classification_mobilenetv2_2022apr.onnx"),
|
38 |
topK=5),
|
39 |
mobilenetv2_q=dict(
|
40 |
-
name="
|
41 |
topic="image_classification",
|
42 |
-
modelPath=os.path.join(root_dir, "models/image_classification_mobilenet/
|
43 |
topK=5),
|
44 |
ppresnet=dict(
|
45 |
name="PPResNet",
|
@@ -49,7 +49,7 @@ models = dict(
|
|
49 |
ppresnet_q=dict(
|
50 |
name="PPResNet",
|
51 |
topic="image_classification",
|
52 |
-
modelPath=os.path.join(root_dir, "models/image_classification_ppresnet/
|
53 |
topK=5),
|
54 |
yunet=dict(
|
55 |
name="YuNet",
|
@@ -72,19 +72,23 @@ models = dict(
|
|
72 |
sface_q=dict(
|
73 |
name="SFace",
|
74 |
topic="face_recognition",
|
75 |
-
modelPath=os.path.join(root_dir, "models/face_recognition_sface/
|
76 |
-
|
77 |
name="CRNN",
|
78 |
topic="text_recognition",
|
79 |
modelPath=os.path.join(root_dir, "models/text_recognition_crnn/text_recognition_CRNN_EN_2021sep.onnx")),
|
|
|
|
|
|
|
|
|
80 |
pphumanseg=dict(
|
81 |
name="PPHumanSeg",
|
82 |
topic="human_segmentation",
|
83 |
-
modelPath=os.path.join(root_dir, "models/human_segmentation_pphumanseg/
|
84 |
pphumanseg_q=dict(
|
85 |
name="PPHumanSeg",
|
86 |
topic="human_segmentation",
|
87 |
-
modelPath=os.path.join(root_dir, "models/human_segmentation_pphumanseg/
|
88 |
)
|
89 |
|
90 |
datasets = dict(
|
|
|
22 |
|
23 |
models = dict(
|
24 |
mobilenetv1=dict(
|
25 |
+
name="MobileNet",
|
26 |
topic="image_classification",
|
27 |
modelPath=os.path.join(root_dir, "models/image_classification_mobilenet/image_classification_mobilenetv1_2022apr.onnx"),
|
28 |
topK=5),
|
29 |
mobilenetv1_q=dict(
|
30 |
+
name="MobileNet",
|
31 |
topic="image_classification",
|
32 |
+
modelPath=os.path.join(root_dir, "models/image_classification_mobilenet/image_classification_mobilenetv1_2022apr_int8.onnx"),
|
33 |
topK=5),
|
34 |
mobilenetv2=dict(
|
35 |
+
name="MobileNet",
|
36 |
topic="image_classification",
|
37 |
modelPath=os.path.join(root_dir, "models/image_classification_mobilenet/image_classification_mobilenetv2_2022apr.onnx"),
|
38 |
topK=5),
|
39 |
mobilenetv2_q=dict(
|
40 |
+
name="MobileNet",
|
41 |
topic="image_classification",
|
42 |
+
modelPath=os.path.join(root_dir, "models/image_classification_mobilenet/image_classification_mobilenetv2_2022apr_int8.onnx"),
|
43 |
topK=5),
|
44 |
ppresnet=dict(
|
45 |
name="PPResNet",
|
|
|
49 |
ppresnet_q=dict(
|
50 |
name="PPResNet",
|
51 |
topic="image_classification",
|
52 |
+
modelPath=os.path.join(root_dir, "models/image_classification_ppresnet/image_classification_ppresnet50_2022jan_int8.onnx"),
|
53 |
topK=5),
|
54 |
yunet=dict(
|
55 |
name="YuNet",
|
|
|
72 |
sface_q=dict(
|
73 |
name="SFace",
|
74 |
topic="face_recognition",
|
75 |
+
modelPath=os.path.join(root_dir, "models/face_recognition_sface/face_recognition_sface_2021dec_int8.onnx")),
|
76 |
+
crnn_en=dict(
|
77 |
name="CRNN",
|
78 |
topic="text_recognition",
|
79 |
modelPath=os.path.join(root_dir, "models/text_recognition_crnn/text_recognition_CRNN_EN_2021sep.onnx")),
|
80 |
+
crnn_en_q=dict(
|
81 |
+
name="CRNN",
|
82 |
+
topic="text_recognition",
|
83 |
+
modelPath=os.path.join(root_dir, "models/text_recognition_crnn/text_recognition_CRNN_EN_2022oct_int8.onnx")),
|
84 |
pphumanseg=dict(
|
85 |
name="PPHumanSeg",
|
86 |
topic="human_segmentation",
|
87 |
+
modelPath=os.path.join(root_dir, "models/human_segmentation_pphumanseg/human_segmentation_pphumanseg_2023mar.onnx")),
|
88 |
pphumanseg_q=dict(
|
89 |
name="PPHumanSeg",
|
90 |
topic="human_segmentation",
|
91 |
+
modelPath=os.path.join(root_dir, "models/human_segmentation_pphumanseg/human_segmentation_pphumanseg_2023mar_int8.onnx")),
|
92 |
)
|
93 |
|
94 |
datasets = dict(
|