Create main.py

main.py

@@ -0,0 +1,70 @@
+import tensorflow as tf
+from tensorflow import keras
+import pandas as pd
+import numpy as np
+print("keep in mind: the ai will periodicaly relearn from your chat and its own data, please dont close the ai's console window while it relearns, as the file could get corrupted, and uterly unusable, forcing you to retrain it from scratch, sincirely: the data-finder team(cross)")
+data_location = "data/"
+
+# Define the structure of the neural network
+def build_model(input_shape):
+    model = keras.Sequential([
+        keras.layers.Dense(64, activation='relu', input_shape=input_shape),
+        keras.layers.Dense(64, activation='relu'),
+        keras.layers.Dense(1, activation='sigmoid')
+    ])
+    model.compile(optimizer='adam',
+                  loss='binary_crossentropy',
+                  metrics=['accuracy'])
+    return model
+
+# Load the training data
+def load_data(train_file):
+    df = pd.read_csv(train_file)
+    print(df)
+    X = df['content'].values
+    y = df['expected_output'].values
+    return X, y
+
+# Convert text data to vectors using an embedding layer
+def preprocess_data(X_train, y_train, X_test, y_test, max_features=20000, maxlen=100):
+    tokenizer = keras.preprocessing.text.Tokenizer(num_words=max_features)
+    tokenizer.fit_on_texts(list(X_train))
+    X_train = tokenizer.texts_to_sequences(X_train)
+    X_test = tokenizer.texts_to_sequences(X_test)
+    X_train = keras.preprocessing.sequence.pad_sequences(X_train, maxlen=maxlen)
+    X_test = keras.preprocessing.sequence.pad_sequences(X_test, maxlen=maxlen)
+    y_train = np.array(y_train)
+    y_test = np.array(y_test)
+    return X_train, y_train, X_test, y_test
+
+# Train the neural network on the training data
+def train(model, X_train, y_train, X_test, y_test, epochs=10, batch_size=32):
+    model.fit(X_train, y_train, validation_data=(X_test, y_test), epochs=epochs, batch_size=batch_size)
+
+# Generate a response to user input using the trained model
+def generate_response(model, tokenizer, input_text):
+    input_vector = tokenizer.texts_to_sequences([input_text])
+    input_vector = keras.preprocessing.sequence.pad_sequences(input_vector, maxlen=100)
+    output_vector = model.predict(input_vector)
+    output_text = "I'm not sure what you're asking. Can you please clarify?"
+    if output_vector[0][0] >= 0.5:
+        output_text = "Yes, I am Chihiro Fujisaki."
+    return output_text
+
+# Main function to run the AI
+def run():
+    train_file = f"{data_location}initial_data.csv"
+    X, y = load_data(train_file)
+    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)
+    X_train, y_train, X_test, y_test = preprocess_data(X_train, y_train, X_test, y_test)
+    model = build_model(input_shape=(X_train.shape[1],))
+    train(model, X_train, y_train, X_test, y_test)
+    tokenizer = keras.preprocessing.text.Tokenizer(num_words=20000)
+    tokenizer.fit_on_texts(X)
+    while True:
+        input_text = input("You: ")
+        output_text = generate_response(model, tokenizer, input_text)
+        print(f"AI: {output_text}")
+
+if __name__ == '__main__':
+    run()