File size: 1,368 Bytes
f375838
 
 
 
 
 
 
9a09293
f375838
 
 
 
 
 
2ce5d17
 
 
97d7046
f375838
 
9a09293
f375838
 
 
5f47025
 
 
 
 
 
 
 
f375838
 
 
 
 
2ce5d17
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
# Gradio app for YOLOv3

import numpy as np
import gradio as gr
from pytorch_grad_cam import GradCAM
from pytorch_grad_cam.utils.image import show_cam_on_image
from pytorch_grad_cam.utils.model_targets import ClassifierOutputTarget
from display import inference, draw_predictions


gr.Interface(
    inference,
    inputs=[
        gr.Image(label="Input Image"),
        gr.Slider(0, 1, value=0.50, label="IOU Threshold"),
        gr.Slider(0, 1, value=0.50, label="Threshold"),
        gr.Checkbox(label="Show GradCam Image"),
        gr.Slider(0, 1, value=0.5, label="Opacity of GradCAM"),        
    ],
    outputs=gr.Gallery(rows=2, columns=1),
    title = "Object Detection : YoloV3 on PASCAL VOC Dataset From Scratch (with GradCAM)" 
    
    ,examples=[
       ["Examples/000001.jpg", 0.75, 0.75, True, 0.5],
       ["Examples/000002.jpg", 0.75, 0.75, True, 0.5],
       ["Examples/000003.jpg", 0.75, 0.75, True, 0.5],
       ["Examples/000004.jpg", 0.75, 0.75, True, 0.5],
       ["Examples/000005.jpg", 0.75, 0.75, True, 0.5],
       ["Examples/000006.jpg", 0.75, 0.75, True, 0.5],
       ["Examples/000007.jpg", 0.75, 0.75, True, 0.5],
       ["Examples/000008.jpg", 0.75, 0.75, True, 0.5],
       ["Examples/000009.jpg", 0.75, 0.75, True, 0.5],
       ["Examples/000010.jpg", 0.75, 0.75, True, 0.5]
       ]
      
    ,
    layout="horizontal"
).launch()