Various small changes.

config.yaml

@@ -489,7 +489,7 @@ EMPATICA_ACCELEROMETER:
       WINDOWS:
         COMPUTE: True
         WINDOW_LENGTH: 15 # specify window length in seconds
-        SECOND_ORDER_FEATURES: ['mean', 'median', 'sd', 'nlargest_mean', 'nsmallest_mean', 'count_windows']
+        SECOND_ORDER_FEATURES: ['mean', 'median', 'sd', 'nlargest', 'nsmallest', 'count_windows']
         STANDARDIZE_FEATURES: False
       SRC_SCRIPT: src/features/empatica_accelerometer/cr/main.py
 
@@ -518,7 +518,7 @@ EMPATICA_TEMPERATURE:
       WINDOWS:
         COMPUTE: True
         WINDOW_LENGTH: 300 # specify window length in seconds
-        SECOND_ORDER_FEATURES: ['mean', 'median', 'sd', 'nlargest_mean', 'nsmallest_mean', 'count_windows']
+        SECOND_ORDER_FEATURES: ['mean', 'median', 'sd', 'nlargest', 'nsmallest', 'count_windows']
         STANDARDIZE_FEATURES: False
       SRC_SCRIPT: src/features/empatica_temperature/cr/main.py
 
@@ -536,11 +536,11 @@ EMPATICA_ELECTRODERMAL_ACTIVITY:
                   'sigTonicDifference', 'freqFeats','maxPeakAmplitudeChangeBefore', 'maxPeakAmplitudeChangeAfter', 'avgPeakAmplitudeChangeBefore', 
                   'avgPeakAmplitudeChangeAfter', 'avgPeakChangeRatio', 'maxPeakIncreaseTime', 'maxPeakDecreaseTime', 'maxPeakDuration', 'maxPeakChangeRatio',
                   'avgPeakIncreaseTime', 'avgPeakDecreaseTime', 'avgPeakDuration', 'signalOverallChange', 'changeDuration', 'changeRate', 'significantIncrease', 
-                  'significantDecrease']
+                  'significantDecrease', maxPeakResponseSlopeBefore, maxPeakResponseSlopeAfter]
       WINDOWS:
         COMPUTE: True
         WINDOW_LENGTH: 60 # specify window length in seconds
-        SECOND_ORDER_FEATURES: ['mean', 'median', 'sd', 'nlargest_mean', 'nsmallest_mean', count_windows, eda_num_peaks_non_zero]
+        SECOND_ORDER_FEATURES: ['mean', 'median', 'sd', 'nlargest', 'nsmallest', count_windows, eda_num_peaks_non_zero]
         STANDARDIZE_FEATURES: False
       SRC_SCRIPT: src/features/empatica_electrodermal_activity/cr/main.py
 
@@ -559,7 +559,7 @@ EMPATICA_BLOOD_VOLUME_PULSE:
       WINDOWS:
         COMPUTE: True
         WINDOW_LENGTH: 300 # specify window length in seconds
-        SECOND_ORDER_FEATURES: ['mean', 'median', 'sd', 'nlargest_mean', 'nsmallest_mean', 'count_windows', 'hrv_num_windows_non_nan']
+        SECOND_ORDER_FEATURES: ['mean', 'median', 'sd', 'nlargest', 'nsmallest', 'count_windows', 'hrv_num_windows_non_nan']
         STANDARDIZE_FEATURES: False
       SRC_SCRIPT: src/features/empatica_blood_volume_pulse/cr/main.py
 
@@ -579,7 +579,7 @@ EMPATICA_INTER_BEAT_INTERVAL:
       WINDOWS:
         COMPUTE: True
         WINDOW_LENGTH: 300 # specify window length in seconds
-        SECOND_ORDER_FEATURES: ['mean', 'median', 'sd', 'nlargest_mean', 'nsmallest_mean', 'count_windows', 'hrv_num_windows_non_nan']
+        SECOND_ORDER_FEATURES: ['mean', 'median', 'sd', 'nlargest', 'nsmallest', 'count_windows', 'hrv_num_windows_non_nan']
         STANDARDIZE_FEATURES: False
       SRC_SCRIPT: src/features/empatica_inter_beat_interval/cr/main.py
 

environment.yml

@@ -111,7 +111,7 @@ dependencies:
     - biosppy==0.8.0
     - cached-property==1.5.2
     - configargparse==0.15.1
-    - cr-features==0.1.13
+    - cr-features==0.1.15
     - cycler==0.11.0
     - decorator==4.4.2
     - fonttools==4.33.2
@@ -131,6 +131,7 @@ dependencies:
     - pyrsistent==0.15.5
     - pywavelets==1.3.0
     - ratelimiter==1.2.0.post0
+    - seaborn==0.11.2
     - shortuuid==1.0.8
     - snakemake==5.30.2
     - toposort==1.5

src/features/cr_features_helper_methods.py

@@ -22,13 +22,13 @@ def extract_second_order_features(intraday_features, so_features_names, prefix="
         if "sd" in so_features_names:
             so_features = pd.concat([so_features, intraday_features.drop(prefix+"level_1", axis=1).groupby(groupby_cols).std().add_suffix("_SO_sd")], axis=1)
         
-        if "nlargest_mean" in so_features_names: # largest 5 -- maybe there is a faster groupby solution?
+        if "nlargest" in so_features_names: # largest 5 -- maybe there is a faster groupby solution?
             for column in intraday_features.loc[:, ~intraday_features.columns.isin(groupby_cols+[prefix+"level_1"])]:
-                so_features[column+"_SO_nlargest_mean"] = intraday_features.drop(prefix+"level_1", axis=1).groupby(groupby_cols)[column].apply(lambda x: x.nlargest(5).mean())
+                so_features[column+"_SO_nlargest"] = intraday_features.drop(prefix+"level_1", axis=1).groupby(groupby_cols)[column].apply(lambda x: x.nlargest(5).mean())
         
-        if "nsmallest_mean" in so_features_names: # smallest 5 -- maybe there is a faster groupby solution?
+        if "nsmallest" in so_features_names: # smallest 5 -- maybe there is a faster groupby solution?
             for column in intraday_features.loc[:, ~intraday_features.columns.isin(groupby_cols+[prefix+"level_1"])]:
-                so_features[column+"_SO_nsmallest_mean"] = intraday_features.drop(prefix+"level_1", axis=1).groupby(groupby_cols)[column].apply(lambda x: x.nsmallest(5).mean())
+                so_features[column+"_SO_nsmallest"] = intraday_features.drop(prefix+"level_1", axis=1).groupby(groupby_cols)[column].apply(lambda x: x.nsmallest(5).mean())
         
         if "count_windows" in so_features_names:
             so_features["SO_windowsCount"] = intraday_features.groupby(groupby_cols).count()[prefix+"level_1"]

tests/scripts/zero_vals.py

@@ -14,22 +14,20 @@ df = pd.read_csv(path)
 df_num_peaks_zero = df[df["empatica_electrodermal_activity_cr_numPeaks"] == 0]
 columns_num_peaks_zero = df_num_peaks_zero.columns[df_num_peaks_zero.isna().any()].tolist()
 
-df_num_peaks_non_zero_t = df[df["empatica_electrodermal_activity_cr_numPeaks"] != 0]
+df_num_peaks_non_zero = df[df["empatica_electrodermal_activity_cr_numPeaks"] != 0]
-df_num_peaks_non_zero = df_num_peaks_non_zero_t[columns_num_peaks_zero]
+df_num_peaks_non_zero = df_num_peaks_non_zero[columns_num_peaks_zero]
+
+print(df_num_peaks_non_zero[df_num_peaks_non_zero["empatica_electrodermal_activity_cr_maxPeakAmplitudeChangeBefore"] != 0])
 
 
-# row_has_NaN = is_NaN. any(axis=1)
-# rows_with_NaN = df[row_has_NaN]
-# print(rows_with_NaN.size)
 pd.set_option('display.max_columns', None)
-# # pd.set_option('display.max_rows', None)
+
-# print(df_num_peaks_non_zero)
 df_q = pd.DataFrame()
 for col in df_num_peaks_non_zero:
     df_q[col] = pd.to_numeric(pd.cut(df_num_peaks_non_zero[col], bins=[-1,0,0.000000000001,1000], labels=[-1,0,1], right=False))
 
 sns.heatmap(df_q)
-plt.savefig('eda_windows_p01_window_non_zero.png', bbox_inches='tight')
+plt.savefig('eda_windows_p01_window_values_non_zero_peak_distribution_0thresh.png', bbox_inches='tight')
 plt.close()
 
 # Filter columns that do not contain 0
@@ -39,12 +37,4 @@ zero_cols = list(set(columns_num_peaks_zero) - set(non_zero_cols))
 print(non_zero_cols, "\n")
 print(zero_cols)
 
-# maxPeakAmplitudeChangeBefore 
-
-mpacb = df_num_peaks_non_zero_t\
-    [(df_num_peaks_non_zero_t['empatica_electrodermal_activity_cr_avgPeakAmplitudeChangeBefore'] != 0) \
-    & (df_num_peaks_non_zero_t['empatica_electrodermal_activity_cr_numPeaks'] != 0)]
-print(mpacb['empatica_electrodermal_activity_cr_numPeaks'])
-sns.heatmap(mpacb['empatica_electrodermal_activity_cr_numPeaks'])
-plt.savefig('maxPeakAmplitudeChangeBefore.png', bbox_inches='tight')