What is your clinica version？ about clinicadl HOT 15 CLOSED

What is your pytorch version？

from clinicadl.

Hi,
thanks for your interest in clinicadl!

I just updated the main README.md file with the current versions for dependencies (clinica >=0.3.2 and Pytorch >= 1.1.0).

Did you follow the instructions described there?
In resume:

conda create --name clinicadl_env python=3.6 jupyter
conda activate clinicadl_env
conda install -c aramislab -c conda-forge clinica

git clone [email protected]:aramis-lab/AD-DL.git
cd AD-DL
pip install -r requirements.txt
conda install -c pytorch pytorch torchvision

then install clinicadl inside your active environment

cd clinicadl
pip install -e .

I hope that this can works. Otherwise, send me the output message of the error.

from clinicadl.

File "/home/AD-DL/clinicadl/clinicadl/preprocessing/T1_linear.py", line 25, in preprocessing_t1w
from clinica.utils.inputs import check_bids_folder
ImportError: No module named clinica.utils.inputs

from clinicadl.

What is the structure of these two directories？

from clinicadl.

File "/home/AD-DL/clinicadl/clinicadl/preprocessing/T1_linear.py", line 25, in preprocessing_t1w
from clinica.utils.inputs import check_bids_folder
ImportError: No module named clinica.utils.inputs

Could you please show me the output of these commands

conda activate clinicadl_env
conda list | grep "clinica"

?

from clinicadl.

File "/home/AD-DL/clinicadl/clinicadl/preprocessing/T1_linear.py", line 25, in preprocessing_t1w
from clinica.utils.inputs import check_bids_folder
ImportError: No module named clinica.utils.inputs

Could you please show me the output of these commands

conda activate clinicadl_env
conda list | grep "clinica"

?

from clinicadl.

What is the structure of these two directories？

BIDS_DIR is the input dir. BIDS (Brain Imaging Data Structure) is the standard adopted for the organisation of the datasets. You can look for a detailed description in the official link or here.

CAPS_DIR for the preprocessing is the output dir. The caps structure is detailed here.

from clinicadl.

File "/home/AD-DL/clinicadl/clinicadl/preprocessing/T1_linear.py", line 25, in preprocessing_t1w
from clinica.utils.inputs import check_bids_folder
ImportError: No module named clinica.utils.inputs

Could you please show me the output of these commands

conda activate clinicadl_env
conda list | grep "clinica"

?

It seems that you haven't installed clinicadl (inside the clinicadl_env environment). What's the exact command that generates the error ?

from clinicadl.

File "/home/AD-DL/clinicadl/clinicadl/preprocessing/T1_linear.py", line 25, in preprocessing_t1w
from clinica.utils.inputs import check_bids_folder
ImportError: No module named clinica.utils.inputs

Could you please show me the output of these commands

conda activate clinicadl_env
conda list | grep "clinica"

?

It seems that you haven't installed clinicadl (inside the clinicadl_env environment). What's the exact command that generates the error ?

from clinicadl.

I think that there is a problem with your $PYTHONPATH, the path where Python search for installed libraries.

I replicated the installation steps in a fresh new installation (an Ubuntu machine) and normally the output of the command conda list | grep "clinica" should show these lines:

# packages in environment at /Users/mdiaz/conda/envs/clinicadl_env:
clinica                   0.3.2                    py36_0    aramislab
clinicadl                 0.0.1                     dev_0    <develop>

Proposed you to delete your conda environment and create a new one, following the installation instructions in strict order.

from clinicadl.

I think that there is a problem with your $PYTHONPATH, the path where Python search for installed libraries.

I replicated the installation steps in a fresh new installation (an Ubuntu machine) and normally the output of the command conda list | grep "clinica" should show these lines:

# packages in environment at /Users/mdiaz/conda/envs/clinicadl_env:
clinica                   0.3.2                    py36_0    aramislab
clinicadl                 0.0.1                     dev_0    <develop>

Proposed you to delete your conda environment and create a new one, following the installation instructions in strict order.

Thank you very much. I'll try

from clinicadl.

Hi，The data preprocessor can run normally, but there will be errors in the process。
File "/home/geng/anaconda3/envs/clinicadl_env/lib/python3.6/site-packages/nipype/pipeline/plugins/tools.py", line 82, in report_nodes_not_run
raise RuntimeError(('Workflow did not execute cleanly. '
RuntimeError: Workflow did not execute cleanly. Check log for details

from clinicadl.

This is an error message

Traceback (most recent call last):
File "/home/geng/anaconda3/envs/clinicadl_env/bin/clinicadl", line 11, in
load_entry_point('clinicadl', 'console_scripts', 'clinicadl')()
File "/home/geng/PycharmProjects/AD-DL/clinicadl/clinicadl/main.py", line 23, in main
args.func(args)
File "/home/geng/PycharmProjects/AD-DL/clinicadl/clinicadl/cli.py", line 11, in preprocessing_t1w_func
wf.run(plugin='MultiProc', plugin_args={'n_procs': args.nproc})
File "/home/geng/anaconda3/envs/clinicadl_env/lib/python3.6/site-packages/nipype/pipeline/engine/workflows.py", line 595, in run
runner.run(execgraph, updatehash=updatehash, config=self.config)
File "/home/geng/anaconda3/envs/clinicadl_env/lib/python3.6/site-packages/nipype/pipeline/plugins/base.py", line 192, in run
report_nodes_not_run(notrun)
File "/home/geng/anaconda3/envs/clinicadl_env/lib/python3.6/site-packages/nipype/pipeline/plugins/tools.py", line 82, in report_nodes_not_run
raise RuntimeError(('Workflow did not execute cleanly. '
RuntimeError: Workflow did not execute cleanly. Check log for details

from clinicadl.

Can you inspect the crash file (it is normally generated in the folder where you launch the application clinicadl) ? You can do this using this command: nipypecli crash crashfile-date-time.pklz
Notice that to execute preprocessing you need to have ANTs previously installed.

from clinicadl.

Dear user,

Are you still facing this issue? You opened another issue (issue #6) so I was wondering if we could close this issue.

Best,
Alexandre

from clinicadl.