Annas branch

behav/loading.py

@@ -63,11 +63,11 @@ def load_data_pandas(subjects, data_folder, force_boolean=['reward']):
                 dt_maker = _make_dt_maker(year)
                 df['date'] = df.apply(dt_maker, axis=1)
                 df.set_index('date', inplace=True)
-                df['type_'] = df['old_type'].map(lambda(x): ['correction','normal'][x])
-                df['response'] = df['old_response'].map(lambda(x): ['none', 'L', 'R'][x])
-                df['correct'] = df['old_correct'].map(lambda(x): [False, True, float('nan')][x])
-                df['reward'] = df.apply(lambda(x): x['reinforcement'] == 1 and x['correct'] == True, axis=1)
-                df['class_'] = df['old_class'].map(lambda(x): ['none', 'L', 'R'][x])
+                df['type_'] = df['old_type'].map(lambda x: ['correction','normal'][x])
+                df['response'] = df['old_response'].map(lambda x: ['none', 'L', 'R'][x])
+                df['correct'] = df['old_correct'].map(lambda x: [False, True, float('nan')][x])
+                df['reward'] = df.apply(lambda x: x['reinforcement'] == 1 and x['correct'] == True, axis=1)
+                df['class_'] = df['old_class'].map(lambda x: ['none', 'L', 'R'][x])
                 df['data_file'] = data_f
                 df_set.append(df)
 
@@ -95,11 +95,11 @@ def load_data_pandas(subjects, data_folder, force_boolean=['reward']):
                 dt_maker = _make_dt_maker(year)
                 df['date'] = df.apply(dt_maker, axis=1)
                 df.set_index('date', inplace=True)
-                df['type_'] = df['old_type'].map(lambda(x): ['correction','normal'][x])
-                df['response'] = df['old_response'].map(lambda(x): ['none', 'C'][x])
-                df['correct'] = df['old_correct'].map(lambda(x): [False, True, float('nan')][x])
-                df['reward'] = df.apply(lambda(x): x['reinforcement'] == 1 and x['correct'] == True, axis=1)
-                df['class_'] = df['old_class'].map(lambda(x): ['none', 'GO', 'NOGO'][x])
+                df['type_'] = df['old_type'].map(lambda x: ['correction','normal'][x])
+                df['response'] = df['old_response'].map(lambda x: ['none', 'C'][x])
+                df['correct'] = df['old_correct'].map(lambda x: [False, True, float('nan')][x])
+                df['reward'] = df.apply(lambda x: x['reinforcement'] == 1 and x['correct'] == True, axis=1)
+                df['class_'] = df['old_class'].map(lambda x: ['none', 'GO', 'NOGO'][x])
                 df['data_file'] = data_f
                 df_set.append(df)
 
@@ -122,12 +122,12 @@ def _parse(datestr, timestr):
                                  )
                 df.rename(columns=col_map, inplace=True)
                 df.set_index('date',inplace=True)
-                df['type_'] = df['Correction'].map(lambda(x): {0:'normal',1:'correction',243:'error',-1:None}[x])
-                df['correct'] = df['ResponseAccuracy'].map(lambda(x): [False, True, float('nan')][x])
-                df['reward'] = df.apply(lambda(x): x['Reinforced'] == 1 and x['correct'] == True, axis=1)
-                df['punish'] = df.apply(lambda(x): x['Reinforced'] == 1 and x['correct'] == False, axis=1)
-                df['class_'] = df['StimClass'].map(lambda(x): {0:'none',1:'L',2:'R',243:'error',-1:None}[x])
-                df['response'] = df['ResponseSelection'].map(lambda(x): ['none', 'L', 'R'][x])
+                df['type_'] = df['Correction'].map(lambda x: {0:'normal',1:'correction',243:'error',-1:None}[x])
+                df['correct'] = df['ResponseAccuracy'].map(lambda x: [False, True, float('nan')][x])
+                df['reward'] = df.apply(lambda x: x['Reinforced'] == 1 and x['correct'] == True, axis=1)
+                df['punish'] = df.apply(lambda x: x['Reinforced'] == 1 and x['correct'] == False, axis=1)
+                df['class_'] = df['StimClass'].map(lambda x: {0:'none',1:'L',2:'R',243:'error',-1:None}[x])
+                df['response'] = df['ResponseSelection'].map(lambda x: ['none', 'L', 'R'][x])
                 df['data_file'] = data_f
 
                 is_behave = df['BehavioralRecording'] > 0
@@ -141,12 +141,12 @@ def _parse(datestr, timestr):
         if df_set:
             behav_data[subj] = pd.concat(df_set).sort_index()
         else:
-            print 'data not found for %s' % (subj)
+            print("data not found for %s" % (subj))
     if force_boolean:
         for subj in subjects:
             if subj in behav_data:
                 for forced in force_boolean:
-                    behav_data[subj][forced] = behav_data[subj][forced].map(lambda(x): x in [True, 'True', 'true', 1, '1'])
+                    behav_data[subj][forced] = behav_data[subj][forced].map(lambda x: x in [True, 'True', 'true', 1, '1'])
     return behav_data
 
 def _make_dt_maker(year):

behav/plotting.py

@@ -2,7 +2,7 @@
 import matplotlib.pyplot as plt
 import seaborn as sns
 import pandas as pd
-import utils
+import behav.utils as utils
 import scipy as sp
 from scipy import ndimage
 
@@ -214,7 +214,7 @@ def plot_accuracy_bias(subj, df, x_axis='time', smoothing='exponential', trial_l
     else:
         raise Exception('invalid value for x_axis')
 
-    datas = (df['correct'], df['response']=='L', df['response']=='R')
+    datas = (df['correct'], df['response'].isin(['L','left']), df['response'].isin(['R','right']))
     plot_smoothed_mask = (plt_correct_smoothed, plt_L_response_smoothed, plt_R_response_smoothed)
     plot_shaded_mask = (plt_correct_shade, plt_L_response_shade, plt_R_response_shade)
     plot_line_mask = (plt_correct_line, plt_L_response_line, plt_R_response_line)

behav/utils.py

@@ -113,7 +113,7 @@ def filter_normal_trials(df):
     '''
     filters dataframe, df, to only include normal (non-correction) trials that got a response.
     '''
-    return df[(df.response<>'none')&(df.type_=='normal')]
+    return df[(df.response!='none')&(df.type_=='normal')]
 
 def filter_recent_days(df, num_days):
     '''