sahilchachra/video-analytics-dashboard

Video Analytics dashboard built using YoloV5 and Streamlit

👋  Hey there! I'm Sahil Chachra

👨🏻‍💻  About Me

💻  AI Engineer at SparkCognition.
💻  Former Deep Learning Engineer at Data Annotation Studio, Smart Mobility, TCS.
💻  Former Intern at Regional Remote Sensing Centre, ISRO.
🎓  Completed my Bachelors of Technology in Computer Science & Engineering in 2021.
🌱  Working in the field of Deep learning and Computer vision.
💬  Feel free to reach out to me for projects, or just for some interesting discussion.

Languages and Tools

🚀  OpenSource Contribution

• Ultralytics/Yolov3 - Fix ONNX inference code
• Ultralytics/Yolov5 - Fix FP32 TensorRT model export

📕  Medium Articles

• Comparative study — Using EfficientNetB0 to EfficientNetB7 as encoder in UNet
• Vanilla UNet vs UNet with EfficientNetB0 as Encoder
• Setting up Nvidia’s Jetson Nano : From Jetpack to YoloV5
• Run YoloV5s with TensorRT and DeepStream on Nvidia Jetson Nano
• Paper summary — Decoupled Weight Decay Regularization
• Paper Summary — What is being transferred in transfer learning?
• Paper Summary — torch.manual_seed(3407) is all you need
• Paper Summary — MetaFormer is Actually What You Need for Vision
• Dive into basics of GPU, CUDA & Accelerated programming using Numba in Python
• Video Analytics Dashboard for YoloV5 and DeepSort
• Intel OpenVINO in brief
• All you want, to get started with GStreamer in Python
• Paper Summary — Searching for TrioNet
• Paper Summary — Efficient Long Range Attention Network (ELAN)
• Paper Summary — RepVGG : Making VGG-style ConvNets Great Again
• Paper Summary — ShuffleNetv2 : Practical Guidelines for Efficient CNN Architecture Design
• Paper Summary — Deep Leakage from Gradients

⚙️  GitHub Analytics

🖥️ Dev Environment

🤝🏻  Connect with Me

video-analytics-dashboard's Issues

ImportError: cannot import name 'plot_one_box' from 'utils.plots'

ImportError: cannot import name 'plot_one_box' from 'utils.plots' ，What is this situation

AttributeError: 'Upsample' object has no attribute 'recompute_scale_factor'

So Slow

I checked speed on Strong_sort git and your app on video streaming. it is sooo slow on everything: video, stream webcam, but STRONG_sort is the way faster

AttributeError: 'Upsample' object has no attribute 'recompute_scale_factor'

Windows PowerShell
Copyright (C) Microsoft Corporation. All rights reserved.

Try the new cross-platform PowerShell https://aka.ms/pscore6

PS C:\Users\intel\Desktop\Video-Analytics-Dashboard-main> streamlit run app.py

You can now view your Streamlit app in your browser.

Local URL: http://localhost:8501
Network URL: http://192.168.29.114:8501

2022-08-31 18:58:38.650 Loading weights from deep_sort_pytorch/deep_sort/deep/checkpoint/ckpt.t7... Done!
Folder exists, overwriting...
2022-08-31 18:58:38.674 YOLOv5 2022-8-4 torch 1.12.1+cpu CPU

2022-08-31 18:58:38.686 YOLOv5 2022-8-4 torch 1.12.1+cpu CPU

2022-08-31 18:58:38.792 Fusing layers...
2022-08-31 18:58:39.092 Model Summary: 213 layers, 7225885 parameters, 0 gradients
2022-08-31 18:58:39.093 WARNING: --img-size (640, 640) must be multiple of max stride 32, updating to [640, 640]
2022-08-31 18:58:39.598 Uncaught app exception
Traceback (most recent call last):
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\streamlit\runtime\scriptrunner\script_runner.py", line 556, in run_script
exec(code, module.dict)
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main\app.py", line 232, in
main()
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main\app.py", line 97, in main
detect(source=video.name, stframe=stframe, kpi1_text=kpi1_text, kpi2_text=kpi2_text, kpi3_text=kpi3_text, js1_text=js1_text, js2_text=js2_text, js3_text=js3_text, conf_thres=float(conf_thres), nosave=nosave, display_labels=display_labels, conf_thres_drift = float(conf_thres_drift), save_poor_frame_= save_poor_frame__, inf_ov_1_text=inf_ov_1_text, inf_ov_2_text=inf_ov_2_text, inf_ov_3_text=inf_ov_3_text, inf_ov_4_text=inf_ov_4_text, fps_warn=fps_warn, fps_drop_warn_thresh File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\autograd\grad_mode.py", line 27, in decorate_context
return func(*args, **kwargs)
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main\deep_list.py", line 170, in detect
pred = model(im, augment=augment, visualize=visualize)
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\nn\modules\module.py", line 1130, in _call_impl
return forward_call(*input, **kwargs)
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main\yolov5\models\common.py", line 402, in forward
y = self.model(im) if self.jit else self.model(im, augment=augment, visualize=visualize)
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\nn\modules\module.py", line 1130, in _call_impl
return forward_call(*input, **kwargs)
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main./yolov5\models\yolo.py", line 126, in forward
return self._forward_once(x, profile, visualize) # single-scale inference, train
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main./yolov5\models\yolo.py", line 149, in _forward_once
x = m(x) # run
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\nn\modules\module.py", line 1130, in _call_impl
return forward_call(*input, **kwargs)
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\nn\modules\upsampling.py", line 154, in forward
recompute_scale_factor=self.recompute_scale_factor)
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\nn\modules\module.py", line 1207, in getattr
raise AttributeError("'{}' object has no attribute '{}'".format(
AttributeError: 'Upsample' object has no attribute 'recompute_scale_factor'
Stopping...
PS C:\Users\intel\Desktop\Video-Analytics-Dashboard-main> streamlit run app.py

You can now view your Streamlit app in your browser.

Local URL: http://localhost:8501
Network URL: http://192.168.29.114:8501

2022-08-31 19:01:35.927 Loading weights from deep_sort_pytorch/deep_sort/deep/checkpoint/ckpt.t7... Done!
Folder exists, overwriting...
2022-08-31 19:01:35.949 YOLOv5 2022-8-4 torch 1.12.1+cpu CPU

2022-08-31 19:01:35.962 YOLOv5 2022-8-4 torch 1.12.1+cpu CPU

2022-08-31 19:01:36.090 Fusing layers...
2022-08-31 19:01:36.410 Model Summary: 213 layers, 7225885 parameters, 0 gradients
2022-08-31 19:01:36.412 WARNING: --img-size (640, 640) must be multiple of max stride 32, updating to [640, 640]
2022-08-31 19:01:36.500 Uncaught app exception
Traceback (most recent call last):
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\streamlit\runtime\scriptrunner\script_runner.py", line 556, in run_script
exec(code, module.dict)
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main\app.py", line 232, in
main()
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main\app.py", line 162, in main
detect(source='1', stframe=stframe, kpi1_text=kpi1_text, kpi2_text=kpi2_text, kpi3_text=kpi3_text, js1_text=js1_text, js2_text=js2_text, js3_text=js3_text, conf_thres=float(conf_thres), nosave=nosave, display_labels=display_labels, conf_thres_drift = float(conf_thres_drift), save_poor_frame_= save_poor_frame__, inf_ov_1_text=inf_ov_1_text, inf_ov_2_text=inf_ov_2_text, inf_ov_3_text=inf_ov_3_text, inf_ov_4_text=inf_ov_4_text, fps_warn=fps_warn, fps_drop_warn_thresh = float(fps_drop_warn_thresh))
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\autograd\grad_mode.py", line 27, in decorate_context
return func(*args, **kwargs)
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main\deep_list.py", line 136, in detect
dataset = LoadStreams(source, img_size=imgsz, stride=stride, auto=pt)
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main\yolov5\utils\datasets.py", line 309, in init
assert cap.isOpened(), f'{st}Failed to open {s}'
AssertionError: 1/1: 1... Failed to open 1
2022-08-31 19:08:09.912 Loading weights from deep_sort_pytorch/deep_sort/deep/checkpoint/ckpt.t7... Done!
Folder exists, overwriting...
2022-08-31 19:08:09.946 YOLOv5 2022-8-4 torch 1.12.1+cpu CPU

2022-08-31 19:08:09.964 YOLOv5 2022-8-4 torch 1.12.1+cpu CPU

2022-08-31 19:08:10.071 Fusing layers...
2022-08-31 19:08:10.591 Model Summary: 213 layers, 7225885 parameters, 0 gradients
2022-08-31 19:08:10.594 WARNING: --img-size (640, 640) must be multiple of max stride 32, updating to [640, 640]
2022-08-31 19:08:11.045 Uncaught app exception
Traceback (most recent call last):
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\streamlit\runtime\scriptrunner\script_runner.py", line 556, in run_script
exec(code, module.dict)
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main\app.py", line 232, in
main()
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main\app.py", line 97, in main
detect(source=video.name, stframe=stframe, kpi1_text=kpi1_text, kpi2_text=kpi2_text, kpi3_text=kpi3_text, js1_text=js1_text, js2_text=js2_text, js3_text=js3_text, conf_thres=float(conf_thres), nosave=nosave, display_labels=display_labels, conf_thres_drift = float(conf_thres_drift), save_poor_frame_= save_poor_frame__, inf_ov_1_text=inf_ov_1_text, inf_ov_2_text=inf_ov_2_text, inf_ov_3_text=inf_ov_3_text, inf_ov_4_text=inf_ov_4_text, fps_warn=fps_warn, fps_drop_warn_thresh = float(fps_drop_warn_thresh))
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\autograd\grad_mode.py", line 27, in decorate_context
return func(*args, **kwargs)
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main\deep_list.py", line 170, in detect
pred = model(im, augment=augment, visualize=visualize)
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\nn\modules\module.py", line 1130, in _call_impl
return forward_call(*input, **kwargs)
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main\yolov5\models\common.py", line 402, in forward
y = self.model(im) if self.jit else self.model(im, augment=augment, visualize=visualize)
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\nn\modules\module.py", line 1130, in _call_impl
return forward_call(*input, **kwargs)
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main./yolov5\models\yolo.py", line 126, in forward
return self._forward_once(x, profile, visualize) # single-scale inference, train
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main./yolov5\models\yolo.py", line 149, in _forward_once
x = m(x) # run
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\nn\modules\module.py", line 1130, in _call_impl
return forward_call(*input, **kwargs)
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\nn\modules\upsampling.py", line 154, in forward
recompute_scale_factor=self.recompute_scale_factor)
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\nn\modules\module.py", line 1207, in getattr
raise AttributeError("'{}' object has no attribute '{}'".format(
AttributeError: 'Upsample' object has no attribute 'recompute_scale_factor'
2022-08-31 19:12:05.900 Loading weights from deep_sort_pytorch/deep_sort/deep/checkpoint/ckpt.t7... Done!
Folder exists, overwriting...
2022-08-31 19:12:05.940 YOLOv5 2022-8-4 torch 1.12.1+cpu CPU

2022-08-31 19:12:05.970 YOLOv5 2022-8-4 torch 1.12.1+cpu CPU

2022-08-31 19:12:06.074 Fusing layers...
2022-08-31 19:12:06.440 Model Summary: 213 layers, 7225885 parameters, 0 gradients
2022-08-31 19:12:06.444 WARNING: --img-size (640, 640) must be multiple of max stride 32, updating to [640, 640]
2022-08-31 19:13:37.880 Uncaught app exception
Traceback (most recent call last):
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\streamlit\runtime\scriptrunner\script_runner.py", line 556, in run_script
exec(code, module.dict)
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main\app.py", line 232, in
main()
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main\app.py", line 226, in main
detect(source=rtsp_input, stframe=stframe, kpi1_text=kpi1_text, kpi2_text=kpi2_text, kpi3_text=kpi3_text, js1_text=js1_text, js2_text=js2_text, js3_text=js3_text, conf_thres=float(conf_thres), nosave=nosave, display_labels=display_labels, conf_thres_drift = float(conf_thres_drift), save_poor_frame_= save_poor_frame__, inf_ov_1_text=inf_ov_1_text, inf_ov_2_text=inf_ov_2_text, inf_ov_3_text=inf_ov_3_text, inf_ov_4_text=inf_ov_4_text, fps_warn=fps_warn, fps_drop_warn_thresh = float(fps_drop_warn_thresh))
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\autograd\grad_mode.py", line 27, in decorate_context
return func(*args, **kwargs)
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main\deep_list.py", line 136, in detect
dataset = LoadStreams(source, img_size=imgsz, stride=stride, auto=pt)
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main\yolov5\utils\datasets.py", line 309, in init
assert cap.isOpened(), f'{st}Failed to open {s}'
AssertionError: 1/1: rtsp://192.168.0.1... Failed to open rtsp://192.168.0.1
2022-08-31 19:13:43.080 Loading weights from deep_sort_pytorch/deep_sort/deep/checkpoint/ckpt.t7... Done!
Folder exists, overwriting...
2022-08-31 19:13:43.109 YOLOv5 2022-8-4 torch 1.12.1+cpu CPU

2022-08-31 19:13:43.122 YOLOv5 2022-8-4 torch 1.12.1+cpu CPU

2022-08-31 19:13:43.233 Fusing layers...
2022-08-31 19:13:43.638 Model Summary: 213 layers, 7225885 parameters, 0 gradients
2022-08-31 19:13:43.640 WARNING: --img-size (640, 640) must be multiple of max stride 32, updating to [640, 640]
2022-08-31 19:13:44.094 Uncaught app exception
Traceback (most recent call last):
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\streamlit\runtime\scriptrunner\script_runner.py", line 556, in run_script
exec(code, module.dict)
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main\app.py", line 232, in
main()
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main\app.py", line 97, in main
detect(source=video.name, stframe=stframe, kpi1_text=kpi1_text, kpi2_text=kpi2_text, kpi3_text=kpi3_text, js1_text=js1_text, js2_text=js2_text, js3_text=js3_text, conf_thres=float(conf_thres), nosave=nosave, display_labels=display_labels, conf_thres_drift = float(conf_thres_drift), save_poor_frame_= save_poor_frame__, inf_ov_1_text=inf_ov_1_text, inf_ov_2_text=inf_ov_2_text, inf_ov_3_text=inf_ov_3_text, inf_ov_4_text=inf_ov_4_text, fps_warn=fps_warn, fps_drop_warn_thresh = float(fps_drop_warn_thresh))
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\autograd\grad_mode.py", line 27, in decorate_context
return func(*args, **kwargs)
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main\deep_list.py", line 170, in detect
pred = model(im, augment=augment, visualize=visualize)
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\nn\modules\module.py", line 1130, in _call_impl
return forward_call(*input, **kwargs)
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main\yolov5\models\common.py", line 402, in forward
y = self.model(im) if self.jit else self.model(im, augment=augment, visualize=visualize)
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\nn\modules\module.py", line 1130, in _call_impl
return forward_call(*input, **kwargs)
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main./yolov5\models\yolo.py", line 126, in forward
return self._forward_once(x, profile, visualize) # single-scale inference, train
File "C:\Users\intel\Desktop\Video-Analytics-Dashboard-main./yolov5\models\yolo.py", line 149, in _forward_once
x = m(x) # run
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\nn\modules\module.py", line 1130, in _call_impl
return forward_call(*input, **kwargs)
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\nn\modules\upsampling.py", line 154, in forward
recompute_scale_factor=self.recompute_scale_factor)
File "C:\Users\intel\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\nn\modules\module.py", line 1207, in getattr
raise AttributeError("'{}' object has no attribute '{}'".format(
AttributeError: 'Upsample' object has no attribute 'recompute_scale_factor'

Disable SORT in some RTMP

I have 3 RTMP link

• water level
• traffic 1
• traffic 2

I want to use deep sort only for traffic 1 and 2, while at water level I don't need it. How to achieve this? (I can duplicate the detect() but I don't know at which part to disable the Deep Sort).

ImportError: cannot import name 'plot_one_box' from 'yolov5.utils.plots'

This is occuring even after installing accurately. There is no function by the name plot_one_box in the utils folder.
ImportError: cannot import name 'plot_one_box' from 'yolov5.utils.plots'

AttributeError: module 'streamlit' has no attribute 'columns'

Now while trying to track objects I get this error.

multi streaming

How can I deploy multi streaming on streamlit app?
Well, In video upload, I upload cameras.txt where is rtsp:// urls, it works, but it shows only 1 stream in streamlit frontend, So I really don't get how to add another stframe = st.empty() and how to pass it another streaming video

CPU higher than GPU

Why the CPU usage higher than GPU (and fps too low) while already active GPU?
RTX 3070 8GB

AttributeError: 'Upsample' object has no attribute 'recompute_scale_factor'

File "C:\Users\Bob\AppData\Roaming\Python\Python39\site-packages\streamlit\runtime\scriptrunner\script_runner.py", line 565, in run_script
exec(code, module.dict)
File "C:\Users\Bob\Video-Analytics-Dashboard\app.py", line 232, in
main()
File "C:\Users\Bob\Video-Analytics-Dashboard\app.py", line 162, in main
detect(source='0', stframe=stframe, kpi1_text=kpi1_text, kpi2_text=kpi2_text, kpi3_text=kpi3_text, js1_text=js1_text, js2_text=js2_text, js3_text=js3_text, conf_thres=float(conf_thres), nosave=nosave, display_labels=display_labels, conf_thres_drift = float(conf_thres_drift), save_poor_frame_= save_poor_frame__, inf_ov_1_text=inf_ov_1_text, inf_ov_2_text=inf_ov_2_text, inf_ov_3_text=inf_ov_3_text, inf_ov_4_text=inf_ov_4_text, fps_warn=fps_warn, fps_drop_warn_thresh = float(fps_drop_warn_thresh))
File "C:\Users\Bob\AppData\Roaming\Python\Python39\site-packages\torch\autograd\grad_mode.py", line 27, in decorate_context
return func(*args, **kwargs)
File "C:\Users\Bob\Video-Analytics-Dashboard\deep_list.py", line 170, in detect
pred = model(im, augment=augment, visualize=visualize)
File "C:\Users\Bob\AppData\Roaming\Python\Python39\site-packages\torch\nn\modules\module.py", line 1194, in _call_impl
return forward_call(*input, **kwargs)
File "C:\Users\Bob\Video-Analytics-Dashboard\yolov5\models\common.py", line 402, in forward
y = self.model(im) if self.jit else self.model(im, augment=augment, visualize=visualize)
File "C:\Users\Bob\AppData\Roaming\Python\Python39\site-packages\torch\nn\modules\module.py", line 1194, in _call_impl
return forward_call(*input, **kwargs)
File "C:\Users\Bob\Video-Analytics-Dashboard./yolov5\models\yolo.py", line 126, in forward
return self._forward_once(x, profile, visualize) # single-scale inference, train
File "C:\Users\Bob\Video-Analytics-Dashboard./yolov5\models\yolo.py", line 149, in _forward_once
x = m(x) # run
File "C:\Users\Bob\AppData\Roaming\Python\Python39\site-packages\torch\nn\modules\module.py", line 1194, in _call_impl
return forward_call(*input, **kwargs)
File "C:\Users\Bob\AppData\Roaming\Python\Python39\site-packages\torch\nn\modules\upsampling.py", line 157, in forward
recompute_scale_factor=self.recompute_scale_factor)
File "C:\Users\Bob\AppData\Roaming\Python\Python39\site-packages\torch\nn\modules\module.py", line 1269, in getattr
raise AttributeError("'{}' object has no attribute '{}'".format(

Crop 1 image for every new object

Well I'm facing some issues bout cropping, when cropping is set to True, it does its job, but it exports detected objects from every frame, but I need to export 1 cropped image for every new car. for example if the first 3 second of video there is 1 car I need 1 crop, if there appears another car I need second 1 crop. Is there any suggestions how to make it workable , that would be interesting for everybody

.ckpt file

I want to ask about this ckpt file. I didn't see it in the checkpoint folder.
Thank you

checkpoint file is deleted

https://drive.google.com/file/d/1TmZRcYQMemPLGjY62LBXwX8HnI_zxy2W/view?usp=sharing

How can I use a GPU for video inference?

Thank you very much for your post. I would like to ask how to use a GPU instead of a CPU for video inference to improve the speed of video inference analysis. My current environment is CUDA 12.3, torch 2.1.2, torchvision 0.16.2, and the GPU I'm using is an RTX 4060.
I look forward to your reply.

What changes to make to run in using our own yolov5 moels?

Do we have to change the deepsort checkpoint?
We can replace the yolov5s.pt with our custom model. But what else changes do we have to make?

AttributeError: 'Upsample' object has no attribute 'recompute_scale_factor'

and when I'm trying to start tracking on another video it says video.mp4 doesn't exist

Webcam do not release

When i open webcam is show succesfully. However it do not release or totally closed when you choose rtsp or local video. I try to add this code below but its not working

if stop_button:
detect(vid_cap=True)

In deeplist:
def detect(vid_cap = False):
if vid_cap:
vid_cap.release()

Filter with model.class

I want to filter detections with model.class = 1 or 2 but I couldn't get when you load model below I wrote model.class = 1 but didn't work.

AttributeError: 'Upsample' object has no attribute 'recompute_scale_factor'

Trying streamlit but getting this error

is it with streamlit version?

I have 1.9

Azure deploy

Hello, do you have any info what is system requirements for the azure to deploy it with cuda(So I need gpu server)?

and also I've searched it but didnt get any info for my question. is azure supports gpu servers with cuda
if you have infor bout it, please share it

Exception: ERROR: fire_cut2.mp4 does not exist

Traceback:
File "C:\Users\Bob\AppData\Roaming\Python\Python39\site-packages\streamlit\runtime\scriptrunner\script_runner.py", line 565, in run_script
exec(code, module.dict)
File "C:\Users\Bob\Video-Analytics-Dashboard\app.py", line 287, in
main()
File "C:\Users\Bob\Video-Analytics-Dashboard\app.py", line 131, in main
detect(source = video.name, stframe=stframe, kpi1_text=kpi1_text, kpi2_text=kpi2_text, kpi3_text=kpi3_text, js1_text=js1_text, js2_text=js2_text, js3_text=js3_text, conf_thres=float(conf_thres), nosave=nosave, display_labels=display_labels, conf_thres_drift = float(conf_thres_drift), save_poor_frame_= save_poor_frame__, inf_ov_1_text=inf_ov_1_text, inf_ov_2_text=inf_ov_2_text, inf_ov_3_text=inf_ov_3_text, inf_ov_4_text=inf_ov_4_text, fps_warn=fps_warn, fps_drop_warn_thresh = float(fps_drop_warn_thresh))
File "C:\Users\Bob\AppData\Roaming\Python\Python39\site-packages\torch\autograd\grad_mode.py", line 27, in decorate_context
return func(*args, **kwargs)
File "C:\Users\Bob\Video-Analytics-Dashboard\deep_list.py", line 139, in detect
dataset = LoadImages(source, img_size=imgsz, stride=stride, auto=pt)
File "C:\Users\Bob\Video-Analytics-Dashboard\yolov5\utils\datasets.py", line 170, in init
raise Exception(f'ERROR: {p} does not exist')

the file is successfully uploaded but does not start when clicking the start tracking

Nothing on the screen

I have done exactly the same steps mentioned in the readme.
I see a blank screen(dashboard).

How do i execute the yolo weights to see through webcam or any video. Please see the screenshot of my screen.
Nothing opens up.

Ask questions

Hello, what should I do if I want to display data in real time in the form of statistical chart，thanks

Error Module numpy has no attribute float (RTMP source)

After getting rid the compute error, I got another error Module Numpy (source from RTMP):

AttributeError: module 'numpy' has no attribute 'float'
Traceback:
File "C:\Users\mikah\miniconda3\envs\analyticdashboard\lib\site-packages\streamlit\runtime\scriptrunner\script_runner.py", line 565, in _run_script
    exec(code, module.__dict__)
File "D:\00_research\streamlit\Video-Analytics-Dashboard\app.py", line 232, in <module>
    main()
File "D:\00_research\streamlit\Video-Analytics-Dashboard\app.py", line 226, in main
    detect(source=rtsp_input, stframe=stframe, kpi1_text=kpi1_text, kpi2_text=kpi2_text, kpi3_text=kpi3_text, js1_text=js1_text, js2_text=js2_text, js3_text=js3_text, conf_thres=float(conf_thres), nosave=nosave, display_labels=display_labels, conf_thres_drift = float(conf_thres_drift), save_poor_frame__= save_poor_frame__, inf_ov_1_text=inf_ov_1_text, inf_ov_2_text=inf_ov_2_text, inf_ov_3_text=inf_ov_3_text, inf_ov_4_text=inf_ov_4_text, fps_warn=fps_warn, fps_drop_warn_thresh = float(fps_drop_warn_thresh))
File "C:\Users\mikah\miniconda3\envs\analyticdashboard\lib\site-packages\torch\autograd\grad_mode.py", line 27, in decorate_context
    return func(*args, **kwargs)
File "D:\00_research\streamlit\Video-Analytics-Dashboard\deep_list.py", line 235, in detect
    outputs = deepsort.update(xywhs, confss, im0)
File "D:\00_research\streamlit\Video-Analytics-Dashboard\deep_sort_pytorch\deep_sort\deep_sort.py", line 31, in update
    detections = [Detection(bbox_tlwh[i], conf, features[i]) for i, conf in enumerate(
File "D:\00_research\streamlit\Video-Analytics-Dashboard\deep_sort_pytorch\deep_sort\deep_sort.py", line 31, in <listcomp>
    detections = [Detection(bbox_tlwh[i], conf, features[i]) for i, conf in enumerate(
File "D:\00_research\streamlit\Video-Analytics-Dashboard\deep_sort_pytorch\deep_sort\sort\detection.py", line 30, in __init__
    self.tlwh = np.asarray(tlwh, dtype=np.float)
File "C:\Users\mikah\miniconda3\envs\analyticdashboard\lib\site-packages\numpy\__init__.py", line 284, in __getattr__
    raise AttributeError("module {!r} has no attribute "

File missing in deep sort folder

Hello,
While running the code in my anaconda environment i found that in the deep_sort_pytorch/deep_sort/deep/checkpoint
a file named ckpt.t7 is missing which is giving error. Kindly fix it. I have to separately download it from another repo.
Thank you,

MOT-16 MOTA score

Have you evaluated the model yolov5s using MOT-16? What were the results?