-
-
Notifications
You must be signed in to change notification settings - Fork 1.8k
Closed
Description
Ubuntu 16.04 DLC= 2.0.4.1 with virtualenv python3
Hi guys! After using the pretrained MPII model I wanted to use the ImageNet pretrained resnet 101 again. After downloading them via download.sh and declaring 101 in config.yaml I got following error when running deeplabcut.create_new_dataset(config):
In [20]: deeplabcut.create_training_dataset(config)
---------------------------------------------------------------------------
FileNotFoundError Traceback (most recent call last)
<ipython-input-20-ade4483e7b7f> in <module>()
----> 1 deeplabcut.create_training_dataset(config)
/home/sebo/DLC/lib/python3.6/site-packages/deeplabcut/generate_training_dataset/trainingsetmanipulation.py in create_training_dataset(config, num_shuffles, Shuffles, windows2linux)
1012 """
1013 start = os.getcwd()
-> 1014 os.chdir(str(Path(resnet_path).parents[0]))
1015 print("Downloading the pretrained model (ResNets)....")
1016 subprocess.call("download.sh", shell=True)
FileNotFoundError: [Errno 2] No such file or directory: '/home/sebo/DLC/lib/python3.6/site-packages/deeplabcut/Pose_Estimation_Tensorflow/models/pretrained'
I checked trainingsetmanipulation.py and Pose_Estimation_Tensorflow needs to be written in lowercase letters in line 1004. After reinstalling deeplabcut, it found and loaded resnet-101 :)
AlexEMG
Metadata
Metadata
Assignees
Labels
No labels