-
Notifications
You must be signed in to change notification settings - Fork 0
/
7_train_cnn.py
39 lines (31 loc) · 1.38 KB
/
7_train_cnn.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
import pandas as pd
from sklearn.preprocessing import LabelEncoder
from keras.models import Sequential
from keras.layers import Conv1D, MaxPooling1D, Flatten, Dense
from keras.utils import to_categorical
# Load the training data
train_data = pd.read_excel("output/Actor_01/train_test_data.xlsx", sheet_name='Training')
# Separate features (MFCCs) and target variable (Emotion)
X_train = train_data.drop(columns=['Audio File', 'Emotion'])
y_train = train_data['Emotion']
# Encode the target variable
label_encoder = LabelEncoder()
y_train_encoded = label_encoder.fit_transform(y_train)
y_train_categorical = to_categorical(y_train_encoded)
# Reshape X_train to 3D tensor for CNN input
X_train_cnn = X_train.values.reshape(X_train.shape[0], X_train.shape[1], 1)
# Define the CNN model
model = Sequential()
model.add(Conv1D(64, 3, activation='relu', input_shape=(X_train_cnn.shape[1], X_train_cnn.shape[2])))
model.add(MaxPooling1D(2))
model.add(Conv1D(128, 3, activation='relu'))
model.add(MaxPooling1D(2))
model.add(Flatten())
model.add(Dense(256, activation='relu'))
model.add(Dense(len(label_encoder.classes_), activation='softmax'))
# Compile the model
model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
# Fit the model to the training data
model.fit(X_train_cnn, y_train_categorical, epochs=500, batch_size=64)
# Save the trained model
model.save("models/cnn_model.h5")