Load the model about clustimage HOT 3 CLOSED

Without specifying a path, it stores the pickle file in the working directory. I'm not sure why it does not load correctly. However, I changed the saving directory to the default systems path. Maybe that helps. For your second questions, the models were not stored but now they are!

Update to the latest version.

pip install -U clustimage

This should work now:

from clustimage import Clustimage
# Init
cl = Clustimage(method='pca')
# load example with flowers
pathnames = cl.import_example(data='flowers')
# Cluster flowers
cl.fit_transform(pathnames)

cl.save(overwrite=True)
cl.load()

results_find = cl.find(pathnames[0:5], k=10, alpha=0.05)
cl.plot_find()

# Results are stored in:
print(cl.results)

# Parameters are stored in:
print(cl.params)
# Pickle file is saved to:
cl.params['filepath']

from clustimage.

I added the label output in the results of the find functionality.
update with: pip install -U clustimage

Check the documentation over here:
https://erdogant.github.io/clustimage/pages/html/core_functions.html#find

from clustimage.

Thank you so much, now the saving and uploading of the file works perfectly well, I appreciate your support!

on a second note, what I am trying to achieve is to run the clustering model on dataset nb1 (let's say cats and dogs), save it, upload it and run it again on a second dataset (let's say cats, dogs and frogs).

I want it to be able to recognize the labels of the two first classes (which it had clustered before) so I only have to manually label the new class.

I found that the function "find() (clustimage.clustimage.Clustimage method)" might be helpful but I still don't know how to recover the labels!

I would appreciate your help!

from clustimage.