Comments (9)
where did you down load the model~ we provide the models trained with udp here
from udp-pose.
I had download your model and updated the w32_256x192_adam_lr1e-3.yaml from Hrnet
but there is error
python inference.py --cfg ../experiments/coco/hrnet/w32_256x192_adam_lr1e-3.yaml --videoFile ../../../Alexander\ Zverev\ vs\ Dominic\ Thiem\ -\ US\ Open\ 2020\ Final.mp4 --writeBoxFrames --outputDir output TEST.MODEL_FILE UDP-hrnet_w32_256x192.pth
=> loading model from UDP-hrnet_w32_256x192.pth
Traceback (most recent call last):
File "inference.py", line 341, in
main()
File "inference.py", line 226, in main
pose_model.load_state_dict(torch.load(cfg.TEST.MODEL_FILE), strict=False)
File "/home/azuryl/anaconda3/envs/udppose/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1044, in load_state_dict
raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
RuntimeError: Error(s) in loading state_dict for PoseHighResolutionNet:
size mismatch for final_layer.weight: copying a param with shape torch.Size([51, 128, 1, 1]) from checkpoint, the shape in current model is torch.Size([17, 128, 1, 1]).
size mismatch for final_layer.bias: copying a param with shape torch.Size([51]) from checkpoint, the shape in current model is torch.Size([17]).
from udp-pose.
can you run the origin hrnet repo with the pretrained models provided by the origin hrnet repo?
from udp-pose.
If you can't, you should question on the origin hrnet repo for support.
If you can, you won't have problem in using this repo with the pretrained models provided in this repo.
from udp-pose.
try not to use the origin hrnet repo alone with the pretrained models in this repo, so do the inverse
from udp-pose.
your repo should use https://github.com/leoxiaobin/deep-high-resolution-net.pytorch/blob/master/experiments/coco/hrnet/w32_256x192_adam_lr1e-3.yaml
should not use w32_256x192_adam_lr1e-3_offset_ofm.yaml
https://github.com/HuangJunJie2017/UDP-Pose/tree/master/deep-high-resolution-net.pytorch/experiments/coco/hrnet
from udp-pose.
emmm, but the others have no problem about using this.
make sure that you use this config: experiments/coco/hrnet/w32_256x192_adam_lr1e-3_offset_ofm.yaml
make sure that you use this checkpoint: UDP-hrnet_w32_256x192.pth which should be downloaded from here: https://pan.baidu.com/s/1mPuVj8piYzgWjoRgyd0Cwg (dsa9)
run this under https://github.com/HuangJunJie2017/UDP-Pose/tree/master/deep-high-resolution-net.pytorch ?
python tools/test.py --cfg experiments/coco/hrnet/w32_256x192_adam_lr1e-3_offset_ofm.yaml TEST.MODEL_FILE models/pytorch/pose_coco/UDP-hrnet_w32_256x192.pth TEST.USE_GT_BBOX False
from udp-pose.
thank you
from udp-pose.
since your hrnet fold's name is same as orginal repo but in fact your hrnet's fold just can run your model but orginal model .I had run youe model success
"emmm, but the others have no problem about using this."
from udp-pose.
Related Issues (20)
- Saved Model for HRNet-W48-256x192 HOT 2
- ERROR: Could not find a version that satisfies the requirement opencv-python==3.4.1.15 HOT 2
- missing keypoint-aware occlusion augmentation HOT 1
- Question about the pre-process HOT 2
- what is difference pred and preds
- why preds need X scale HOT 1
- why need norm = np.ones((pred.shape[0], 2)) * np.array([h, w]) / 10 in accuracy HOT 4
- why your json's bboxes have so many no human's joints HOT 1
- Confuse about merge origin hrnet into mmpose HOT 9
- About the multi-scale testing HOT 1
- About KPD HOT 5
- Good job,but a little flaw? HOT 1
- AID
- Results on COCO val2017
- Download the pre-trained model
- test accuracy too low use offset
- pose_hrnet_w48* means using additional data from AI challenger for training?有 AI challenger 和coco数据集合并完成的数据集提供吗? HOT 1
- Can you give some explanation on apply affine transform, i.e, the use of get_warpmatrix and warpAffine?
- About evaluation index
- can you release the udp code for simple-R50 and pretrained weight?
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from udp-pose.