MVIG-SJTU / AlphaPose

Real-Time and Accurate Full-Body Multi-Person Pose Estimation&Tracking System
http://mvig.org/research/alphapose.html
Other
8.06k stars 1.98k forks source link

No module named 'detector' #1125

Open melgabay opened 1 year ago

melgabay commented 1 year ago

Please it's an urgent request. When I'm running on Colab import os os.chdir('/content/AlphaPose') ! ls ! python3 /content/AlphaPose/scripts/demo_inference.py --cfg configs/coco/resnet/256x192_res50_lr1e-3_1x.yaml --checkpoint pretrained_models/fast_res50_256x192.pth --indir "/content/drive/MyDrive/Drive/final project/new final project/enfant" --save_img --outdir "/content/drive/MyDrive/Drive/final project/new final project/enfant"

I got this error: Traceback (most recent call last): File "/content/AlphaPose/scripts/demo_inference.py", line 13, in from detector.apis import get_detector ModuleNotFoundError: No module named 'detector'

Could someone help me asap, please?? (the path is good and google drive amount is as expected) Thank in advance

nistring commented 1 year ago

As the error message says, the 'detector' module wasn't not detected. I think you change your current directory to '/content/AlphaPose' For example, instead of using os.chdir, do ! cd /content/Alphapose and run python3 scripts/demo_inference.py

melgabay commented 1 year ago

It didnt helped. I just used the colab's link that is just under Quick Start and I got this error (I changed by the os.chdir by ! cd /content/Alphapose as you advised @nistring

Capture d’écran 2023-04-02 à 21 01 18

Also for some weird reason, when I ran this collab on another account, it may work but not every time. Do someone knows why?

RohaanA commented 1 year ago

Hey, were you able to fix this issue?

JIshanRandika commented 4 months ago

Did you find the solution?

melgabay commented 4 months ago

@JIshanRandika @RohaanA You need to run in on GPU (EC2 instance is the best one)

sanjeevini178 commented 2 months ago

sometimes, it works alright if you reupload the model (.pth file) alone (assuming you are running this on Colab/online. Upload the model file manually to pretrained_models directory and run the command again.