chore: improve code quality

.vscode/launch.json

@@ -0,0 +1,15 @@
+{
+    // Use IntelliSense to learn about possible attributes.
+    // Hover to view descriptions of existing attributes.
+    // For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
+    "version": "0.2.0",
+    "configurations": [
+        {
+            "name": "Python: Current File",
+            "type": "python",
+            "request": "launch",
+            "program": "${file}",
+            "console": "integratedTerminal"
+        }
+    ]
+}

.vscode/settings.json

@@ -1,3 +1,4 @@
 {
-    "python.pythonPath": "/Users/Markus/opt/Anaconda3/python"
+    "python.pythonPath": "/Users/Markus/opt/anaconda3/envs/Slovakia 2021/bin/python",
+    //"terminal.integrated.inheritEnv": false
 }

Handle_emg_data.py

@@ -501,7 +501,7 @@ class DL_data_handler:
 
     def __init__(self, csv_handler:CSV_handler) -> None:
         self.csv_handler = csv_handler
-        # Should med 4 sessions * split nr of samples per person. Each sample is structured like [sample_df, samplerate]
+        # Should med 4 sessions * split nr of samples per person. Each sample is structured like this: [sample_df, samplerate]
         self.samples_per_subject = {1: [],  
                                     2: [], 
                                     3: [],
@@ -573,6 +573,8 @@ class DL_data_handler:
             
             self.samples_per_subject[subject_nr+1] = subj_samples
 
+        
+    
     def reshape_session_df_to_signal(self, df:DataFrame):
         main_df = df[['timestamp', 1]].rename(columns={1: 'emg'})
         for i in range(2, 17):
@@ -593,30 +595,42 @@ class DL_data_handler:
         raw_data_dict = self.get_samples_dict()
     
         # loop through all subjects to get samples
+        mfcc_list = []
+        mfcc_frame_list = []
+
         for key, value in raw_data_dict.items():
 
 
-            # save genre label (i.e., sub-folder name) in the mapping
+            # save subject label in the mapping
             subject_label = 'Subject ' + str(key)
             data["mapping"].append(subject_label)
             print("\nProcessing: {}".format(subject_label))
 
-            # process all audio files in genre sub-dir
+            # process all samples per subject
             for i, (sample) in enumerate(value):
 
-                # load audio file
+                # load signal from sample
                 signal, sample_rate = sample[0], sample[1]
                 signal = signal['emg'].to_numpy()
                 test_df_for_bugs(signal, key, i)
+                #print(sample_rate)
 
                 # extract mfcc
-
                 mfcc = mfcc_custom(signal, sample_rate, MFCC_WINDOWSIZE, MFCC_STEPSIZE, NR_COEFFICIENTS, NR_MEL_BINS)
-                mfcc = mfcc.T
                 
-                data["mfcc"].append(mfcc.tolist())
+                mfcc_list.append(mfcc.tolist())
+                mfcc_frame_list.append(mfcc.shape[0])
+
+                #data["mfcc"].append(mfcc.tolist())
                 data["labels"].append(key)
-                print("sample:{}".format(i+1))
+                print("sample:{} is done".format(i+1))
+
+        minimum = min(mfcc_frame_list)
+
+        for mfcc_data in mfcc_list:
+
+            data["mfcc"].append(mfcc_data[:minimum])
+            print(np.array(mfcc_data[:minimum]).shape)
 
         # save MFCCs to json file
         with open(json_path, "w") as fp:

Neural_Network_Analysis.py

@@ -1,8 +1,9 @@
 import json
+from python_speech_features.python_speech_features.base import mfcc
 import numpy as np
 from sklearn.model_selection import train_test_split
-import tensorflow.keras as keras
 import tensorflow as tf
+import tensorflow.keras as keras
 from pathlib import Path
 import pandas as pd
 import matplotlib.pyplot as plt
@@ -15,12 +16,16 @@ def load_data(data_path):
     with open(data_path, "r") as fp:
         data = json.load(fp)
 
-    # convert lists to numpy arrays
+    # convert lists to numpy arraysls
-    X = np.array(data["mfcc"])
+    #print('\n', data['mfcc'], '\n')
+    X = np.array(data['mfcc'])
+    print(X.shape)
+    #print((len(X), len(X[0]), len(X[0][0])))
+    #print((len(X), len(X[5]), len(X[5][0])))
+
     y = np.array(data["labels"])
-    #X = np.asarray(X).astype('float32')
+    print(y.shape)
-    #y = np.asarray(y).astype('float32')
+    
-    #y = tf.expand_dims(y, axis=1)
 
     print("Data succesfully loaded!")
 
@@ -102,17 +107,9 @@ if __name__ == "__main__":
     # get train, validation, test splits
     X_train, X_validation, X_test, y_train, y_validation, y_test = prepare_datasets(0.25, 0.2)
 
+    print(X_train.shape[1], X_train.shape[2])
     # create network
-
+    input_shape = (X_train.shape[1], X_train.shape[2])  # 18, 13
-    print(X_train.shape)
-    X_train = np.reshape(X_train, (X_train.shape[0], 1, X_train.shape[1]))
-    X_test = np.reshape(X_test, (X_test.shape[0], 1, X_test.shape[1]))
-    #X_validation = np.reshape(X_validation, (X_test.shape[0], 1, X_test.shape[1]))
-    print(X_train.shape)
-    print(X_train.shape[0])
-    print(X_train.shape[1])
-
-    input_shape = (X_train.shape[1], X_train.shape[2])  # 300, 13
     model = build_model(input_shape)
 
     # compile model
@@ -124,7 +121,7 @@ if __name__ == "__main__":
     model.summary()
 
     # train model
-    history = model.fit(X_train, y_train, validation_data=(X_validation, y_validation), batch_size=16, epochs=30)
+    history = model.fit(X_train, y_train, validation_data=(X_validation, y_validation), batch_size=100, epochs=30)
 
     # plot accuracy/error for training and validation
     plot_history(history)
@@ -133,3 +130,4 @@ if __name__ == "__main__":
     test_loss, test_acc = model.evaluate(X_test, y_test, verbose=2)
     print('\nTest accuracy:', test_acc)
 
+

Present_data.py

@@ -221,11 +221,18 @@ def main():
     csv_handler = CSV_handler()
     csv_handler.load_data('soft')
     dl_data_handler = DL_data_handler(csv_handler)
-    dl_data_handler.store_samples(15)
+    dl_data_handler.store_samples(5)
-    #dict = dl_data_handler.samples_per_subject
+    dict = dl_data_handler.samples_per_subject
-    #print(dict.get(1)[10][0], dict.get(1)[10][1])
+    #print(dict.get(1)[2][0], dict.get(1)[10][1])
-    #print(dict.get(1)[12][0], dict.get(1)[12][1])
+    #print(dict.get(1)[15][0], dict.get(1)[10][1])
-    
+    #print(dict.get(1))
+    '''
+    print(len(dict.get(1)))
+    print(len(dict.get(2)))
+    print(len(dict.get(3)))
+    print(len(dict.get(4)))
+    print(len(dict.get(5)))
+    '''
     dl_data_handler.save_mfcc()