Skip to content

Commit

Permalink
update concat results function to use upath
Browse files Browse the repository at this point in the history
  • Loading branch information
@egmcbride
egmcbride committed Oct 30, 2024
1 parent c7a0392 commit a230bcb
Showing 1 changed file with 51 additions and 27 deletions.
78 changes: 51 additions & 27 deletions src/dynamic_routing_analysis/decoding_utils.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -1326,7 +1326,7 @@ def concat_decoder_results(files,savepath=None,return_table=True,single_session=
linear_shift_df.to_csv(upath.UPath(savepath / 'all_linear_shift_decoding_results.csv'))

print('saved decoder results table to:',savepath)

except Exception as e:
print(e)
print('error saving linear shift df')
Expand Down Expand Up @@ -2010,36 +2010,60 @@ def concat_trialwise_decoder_results(files,savepath=None,return_table=False,n_un
if return_table:
return decoder_confidence_versus_response_type,decoder_confidence_dprime_by_block,decoder_confidence_by_switch,decoder_confidence_versus_trials_since_rewarded_target,decoder_confidence_before_after_target



def concat_decoder_summary_tables(dir,savepath):

#create summary folder if does not exist
if not os.path.exists(savepath):
os.makedirs(savepath)

decoder_results_summary_files=glob.glob(os.path.join(dir,'*decoding_results.csv'))

decoder_confidence_versus_response_type_files=glob.glob(os.path.join(dir,'*decoder_confidence_versus_response_type*.pkl'))
decoder_confidence_dprime_by_block=glob.glob(os.path.join(dir,'*decoder_confidence_dprime_by_block*.pkl'))
decoder_confidence_by_switch=glob.glob(os.path.join(dir,'*decoder_confidence_by_switch*.pkl'))
decoder_confidence_versus_trials_since_rewarded_target=glob.glob(os.path.join(dir,'*decoder_confidence_versus_trials_since_rewarded_target*.pkl'))
decoder_confidence_before_after_target=glob.glob(os.path.join(dir,'*decoder_confidence_before_after_target*.pkl'))
if not upath.UPath(savepath).is_dir():
upath.UPath(savepath).mkdir(parents=True)

#load summary tables & concat
decoder_results_summary_files=[]
for p in dir.glob('*decoding_results.csv'):
decoder_results_summary_files.append(pd.read_csv(p))
decoder_results_summary=pd.concat(decoder_results_summary_files)
decoder_results_summary.to_csv(upath.UPath(savepath) / 'decoder_results_summary.csv')

decoder_results_summary=pd.concat([pd.read_csv(f) for f in decoder_results_summary_files],ignore_index=True)

decoder_confidence_versus_response_type=pd.concat([pd.read_pickle(f) for f in decoder_confidence_versus_response_type_files],ignore_index=True)
decoder_confidence_dprime_by_block=pd.concat([pd.read_pickle(f) for f in decoder_confidence_dprime_by_block],ignore_index=True)
decoder_confidence_by_switch=pd.concat([pd.read_pickle(f) for f in decoder_confidence_by_switch],ignore_index=True)
decoder_confidence_versus_trials_since_rewarded_target=pd.concat([pd.read_pickle(f) for f in decoder_confidence_versus_trials_since_rewarded_target],ignore_index=True)
decoder_confidence_before_after_target=pd.concat([pd.read_pickle(f) for f in decoder_confidence_before_after_target],ignore_index=True)

#save summary tables
#find different n_units
n_units=[]
for col in decoder_results_summary.filter(like='true_accuracy_').columns.values:
if len(col.split('_'))==3:
temp_n_units=col.split('_')[2]
try:
n_units.append(int(temp_n_units))
except:
n_units.append(temp_n_units)
else:
n_units.append(None)

decoder_results_summary.to_csv(os.path.join(dir,'summary','decoder_results_summary.csv'))
for nu in n_units:
decoder_confidence_versus_response_type_files=[]
for p in dir.glob('*decoder_confidence_versus_response_type_'+str(nu)+'_units.pkl'):
decoder_confidence_versus_response_type_files.append(pd.read_pickle(p))
decoder_confidence_versus_response_type=pd.concat(decoder_confidence_versus_response_type_files)
decoder_confidence_versus_response_type.to_pickle(upath.UPath(savepath) / ('decoder_confidence_versus_response_type_'+str(nu)+'_units.pkl'))

decoder_confidence_dprime_by_block_files=[]
for p in dir.glob('*decoder_confidence_dprime_by_block_'+str(nu)+'_units.pkl'):
decoder_confidence_dprime_by_block_files.append(pd.read_pickle(p))
decoder_confidence_dprime_by_block=pd.concat(decoder_confidence_dprime_by_block_files)
decoder_confidence_dprime_by_block.to_pickle(upath.UPath(savepath) / ('decoder_confidence_dprime_by_block_'+str(nu)+'_units.pkl'))

decoder_confidence_by_switch_files=[]
for p in dir.glob('*decoder_confidence_by_switch_'+str(nu)+'_units.pkl'):
decoder_confidence_by_switch_files.append(pd.read_pickle(p))
decoder_confidence_by_switch=pd.concat(decoder_confidence_by_switch_files)
decoder_confidence_by_switch.to_pickle(upath.UPath(savepath) / ('decoder_confidence_by_switch_'+str(nu)+'_units.pkl'))

decoder_confidence_versus_trials_since_rewarded_target_files=[]
for p in dir.glob('*decoder_confidence_versus_trials_since_rewarded_target_'+str(nu)+'_units.pkl'):
decoder_confidence_versus_trials_since_rewarded_target_files.append(pd.read_pickle(p))
decoder_confidence_versus_trials_since_rewarded_target=pd.concat(decoder_confidence_versus_trials_since_rewarded_target_files)
decoder_confidence_versus_trials_since_rewarded_target.to_pickle(upath.UPath(savepath) / ('decoder_confidence_versus_trials_since_rewarded_target_'+str(nu)+'_units.pkl'))

decoder_confidence_before_after_target_files=[]
for p in dir.glob('*decoder_confidence_before_after_target_'+str(nu)+'_units.pkl'):
decoder_confidence_before_after_target_files.append(pd.read_pickle(p))
decoder_confidence_before_after_target=pd.concat(decoder_confidence_before_after_target_files)
decoder_confidence_before_after_target.to_pickle(upath.UPath(savepath) / ('decoder_confidence_before_after_target_'+str(nu)+'_units.pkl'))

decoder_confidence_versus_response_type.to_pickle(os.path.join(dir,'summary','decoder_confidence_versus_response_type.pkl'))
decoder_confidence_dprime_by_block.to_pickle(os.path.join(dir,'summary','decoder_confidence_dprime_by_block.pkl'))
decoder_confidence_by_switch.to_pickle(os.path.join(dir,'summary','decoder_confidence_by_switch.pkl'))
decoder_confidence_versus_trials_since_rewarded_target.to_pickle(os.path.join(dir,'summary','decoder_confidence_versus_trials_since_rewarded_target.pkl'))
decoder_confidence_before_after_target.to_pickle(os.path.join(dir,'summary','decoder_confidence_before_after_target.pkl'))

0 comments on commit a230bcb

Please sign in to comment.