# CLEANED DATA VISUAL ANALYSIS (signal, fft, fbank, mfcc) SECTION print("\n\n--------------------Started cleaned data visual analysis for adjusted and {gender} isolated dataset: {name}--------------------".format(gender=confv.gender_male, name=confv.dataset_shemo_male)) da.visual_analysis(df=data_info_shemo_df_m, database=confv.database_shemo, status=confv.clean, gender=confv.gender_male, envelope=False, resample=False) # This is same as, # da.visual_analysis(df=data_info_shemo_df_m, database=confv.database_shemo, status=confv.original, gender=confv.gender_male, envelope=True, resample=True) # Since these cleaned data are already equipped with envelope and resampling, setting them to False or True does not matter. # (envelope and resample does not matter when its clean) print("--------------------Finished cleaned data visual analysis for adjusted and {gender} isolated dataset: {name}--------------------".format(gender=confv.gender_male, name=confv.dataset_shemo_male)) ''' # Building Features print("\n\n--------------------Started building features for adjusted and {gender} isolated dataset: {name}--------------------".format(gender=confv.gender_male, name=confv.dataset_shemo_male)) classes = list(np.unique(data_info_shemo_df_m.stress_emotion)) mconf_shemo_m = confc.ModelConfig(database=confv.database_shemo, gender=confv.gender_male, mode=confv.ml_mode_convolutional, classes=classes) print(mconf_shemo_m.database) print(mconf_shemo_m.gender) print(mconf_shemo_m.mode) print(mconf_shemo_m.nfilt) print(mconf_shemo_m.nfeat) print(mconf_shemo_m.nfft) print(mconf_shemo_m.step) print(mconf_shemo_m.classes) print(mconf_shemo_m.features_save_name) print(mconf_shemo_m.model_config_save_name) print(mconf_shemo_m.training_log_name) print(mconf_shemo_m.model_save_name) print(mconf_shemo_m.model_h5_save_name) print(mconf_shemo_m.model_tflite_save_name) print(mconf_shemo_m.feature_path)
print("\n\n--------------------Started cleaned data visual analysis for adjusted and {gender} isolated dataset: {name}--------------------".format(gender=confv.gender_female, name=confv.dataset_cremad_female)) da.visual_analysis(df=data_info_cremad_df_f, database=confv.database_cremad, status=confv.clean, gender=confv.gender_female, envelope=False, resample=False) # This is same as, # da.visual_analysis(df=data_info_cremad_df_f, database=confv.database_cremad, status=confv.original, gender=confv.gender_female, envelope=True, resample=True) # Since these cleaned data are already equipped with envelope and resampling, setting them to False or True does not matter. # (envelope and resample does not matter when its clean) print("--------------------Finished cleaned data visual analysis for adjusted and {gender} isolated dataset: {name}--------------------".format(gender=confv.gender_female, name=confv.dataset_cremad_female)) ''' # Building Features print( "\n\n--------------------Started building features for adjusted and {gender} isolated dataset: {name}--------------------" .format(gender=confv.gender_female, name=confv.dataset_cremad_female)) classes = list(np.unique(data_info_cremad_df_f.stress_emotion)) mconf_cremad_f = confc.ModelConfig(database=confv.database_cremad, gender=confv.gender_female, mode=confv.ml_mode_convolutional, classes=classes) print(mconf_cremad_f.database) print(mconf_cremad_f.gender) print(mconf_cremad_f.mode) print(mconf_cremad_f.nfilt) print(mconf_cremad_f.nfeat) print(mconf_cremad_f.nfft) print(mconf_cremad_f.step) print(mconf_cremad_f.classes) print(mconf_cremad_f.features_save_name) print(mconf_cremad_f.model_config_save_name) print(mconf_cremad_f.training_log_name) print(mconf_cremad_f.model_save_name) print(mconf_cremad_f.model_h5_save_name) print(mconf_cremad_f.model_tflite_save_name)
print("\n\n--------------------Started cleaned data visual analysis for adjusted and {gender} isolated dataset: {name}--------------------".format(gender=confv.gender_male, name=confv.dataset_ravdess_male)) da.visual_analysis(df=data_info_ravdess_df_m, database=confv.database_ravdess, status=confv.clean, gender=confv.gender_male, envelope=False, resample=False) # This is same as, # da.visual_analysis(df=data_info_ravdess_df_m, database=confv.database_ravdess, status=confv.original, gender=confv.gender_male, envelope=True, resample=True) # Since these cleaned data are already equipped with envelope and resampling, setting them to False or True does not matter. # (envelope and resample does not matter when its clean) print("--------------------Finished cleaned data visual analysis for adjusted and {gender} isolated dataset: {name}--------------------".format(gender=confv.gender_male, name=confv.dataset_ravdess_male)) ''' # Building Features print( "\n\n--------------------Started building features for adjusted and {gender} isolated dataset: {name}--------------------" .format(gender=confv.gender_male, name=confv.dataset_ravdess_male)) classes = list(np.unique(data_info_ravdess_df_m.stress_emotion)) mconf_ravdess_m = confc.ModelConfig(database=confv.database_ravdess, gender=confv.gender_male, mode=confv.ml_mode_convolutional, classes=classes) print(mconf_ravdess_m.database) print(mconf_ravdess_m.gender) print(mconf_ravdess_m.mode) print(mconf_ravdess_m.nfilt) print(mconf_ravdess_m.nfeat) print(mconf_ravdess_m.nfft) print(mconf_ravdess_m.step) print(mconf_ravdess_m.classes) print(mconf_ravdess_m.features_save_name) print(mconf_ravdess_m.model_config_save_name) print(mconf_ravdess_m.training_log_name) print(mconf_ravdess_m.model_save_name) print(mconf_ravdess_m.model_h5_save_name) print(mconf_ravdess_m.model_tflite_save_name)