search

EthoML / VAME

Variational Animal Motion Embedding - A tool for time series embedding and clustering
https://ethoml.github.io/VAME/
GNU General Public License v3.0
2 stars 0 forks source link

Community analysis breaking for cohort=True and show_umap=True #32

Open vinicvaz opened 3 weeks ago

vinicvaz commented 3 weeks ago

When running vame.community(cohort=True, show_umap=True) the function breaks.

The problem is happening in this line: https://github.com/EthoML/VAME/blob/59d5d730d9c8f9ca048022bd3224ddfc58891863/src/vame/analysis/community_analysis.py#L478

The problem is that in line 577 we pass a list of files to the umap_embedding function, but it expects a single file and not an array of files paths. How should we handle that? Should we create an umap embedding for each file when cohort=True and then agg them?

luiztauffer commented 3 weeks ago

@vinicvaz can you please paste the error track?

vinicvaz commented 3 weeks ago 
--> [479](https://file+.vscode-resource.vscode-cdn.net/home/vinicius/Documents/work/catalyst/VAME/examples/~/Documents/work/catalyst/VAME/src/vame/analysis/community_analysis.py:479) folder = os.path.join(cfg['project_path'],"results",file,model_name, parametrization +'-'+str(n_cluster),"")
    [480](https://file+.vscode-resource.vscode-cdn.net/home/vinicius/Documents/work/catalyst/VAME/examples/~/Documents/work/catalyst/VAME/src/vame/analysis/community_analysis.py:480) latent_vector = np.load(os.path.join(folder,'latent_vector_'+file+'.npy'))
    [482](https://file+.vscode-resource.vscode-cdn.net/home/vinicius/Documents/work/catalyst/VAME/examples/~/Documents/work/catalyst/VAME/src/vame/analysis/community_analysis.py:482) num_points = cfg['num_points']

File <frozen posixpath>:90, in join(a, *p)

File <frozen genericpath>:152, in _check_arg_types(funcname, *args)

TypeError: join() argument must be str, bytes, or os.PathLike object, not 'list'