Issues with model saving and post_pred_gen

[Hamdi Habacha]

Hi all, 

 I hope you're all doing well

As many of you, I had so much trouble installing HDDM, and I tried every solution I could find in the internet.

I managed to install it after dozens of tries, now I am using it both on a docker image "hcp4715/hddm" and on anaconda.

However,  I  am encountering problems in which I remain stuck until now, without being able to take advantage of all hddm features.

With Docker image:
1-  I want to save my model in a local directory ("D:\RESEARCH\DDM\Models") in order to be able to load it and re-use it for stats and revisions... but I've tried for hours (a lotsof solutions on the ineternet) without sucess.
2- When I run "ppc_data = hddm.utils.post_pred_gen(model)", it displays "Start generating posterior prediction..." and it stucks in this command and never finishes generating !

With Ananconda:

When I run "ppc_data = hddm.utils.post_pred_gen(model2)" I get the following error:

--------------------------------------------------------------------------- TypeError Traceback (most recent call last) <ipython-input-5-c088c9f1ed31> in <module> ----> 1 ppc_data = hddm.utils.post_pred_gen(model2) C:\ProgramData\anaconda3\envs\hddmEnv\lib\site-packages\kabuki\analyze.py in post_pred_gen(model, groupby, samples, append_data, add_model_parameters, progress_bar) 378 data=data, 379 append_data=append_data, --> 380 add_model_parameters=add_model_parameters, 381 ) 382 results[new_name] = pd.concat( C:\ProgramData\anaconda3\envs\hddmEnv\lib\site-packages\kabuki\analyze.py in _post_pred_generate(bottom_node, samples, data, append_data, add_model_parameters) 295 _parents_to_random_posterior_sample(bottom_node) 296 # Generate data from bottom node --> 297 sampled_data = bottom_node.random(add_model_parameters=add_model_parameters) 298 if append_data and data is not None: 299 sampled_data = sampled_data.join(data.reset_index(), lsuffix="_sampled") TypeError: random() got an unexpected keyword argument 'add_model_parameters'---------------------------------------------------------------------------

These are the versions of hddm and kabuki that I I managed to install to make hddm work on anaconda environement:

hddm 0.8.0 kabuki 0.6.5.

Has anyone encountered the same issues? 

And could you please suggest a solution? 

I've been trying to take advantage of hddm's features for 2 weeks without success!

Thank you all in advance,

Best wishes,

Hamdi

[Hamdi Habacha]

Update : 

Saving my model in docker is not an issue anymore,

Once I have "ppc_data = hddm.utils.post_pred_gen(model)" working and I can save it's result, I would be infinitely grateful !

Thank you

Hamdi