Upload 4 files

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+gru/save/gru_model.keras filter=lfs diff=lfs merge=lfs -text
+input/emotions.csv filter=lfs diff=lfs merge=lfs -text

gru/index_gru.py

@@ -0,0 +1,71 @@
+import numpy as np
+import pandas as pd
+import seaborn as sns
+import tensorflow as tf
+import matplotlib.pyplot as plt
+data = pd.read_csv('../input/emotions.csv')
+data['label'].value_counts()
+data.isnull().sum()
+label_mapping = {'NEGATIVE': 0, 'NEUTRAL': 1, 'POSITIVE': 2}
+data['label'] = data['label'].replace(label_mapping)
+X = data.drop('label', axis=1).copy()
+y = data['label'].copy()
+from sklearn.model_selection import train_test_split
+X_train, X_test, y_train, y_test = train_test_split(X, y, train_size=0.8, random_state=42)
+inputs = tf.keras.Input(shape=(X_train.shape[1],))
+expand_dims = tf.expand_dims(inputs, axis=2)
+gru = tf.keras.layers.GRU(256, return_sequences=True)(expand_dims)
+flatten = tf.keras.layers.Flatten()(gru)
+outputs = tf.keras.layers.Dense(3, activation='softmax')(flatten)
+model = tf.keras.Model(inputs=inputs, outputs=outputs)
+
+new_learning_rate = 1  # 设置学习率
+model.compile(
+    optimizer=tf.keras.optimizers.Adam(learning_rate=new_learning_rate),
+    loss='sparse_categorical_crossentropy',
+    metrics=['accuracy']
+)
+
+model.compile(
+    optimizer='adam',
+    loss='sparse_categorical_crossentropy',
+    metrics=['accuracy']
+)
+print(model.summary())
+history = model.fit(  # 开始训练
+    X_train,
+    y_train,
+    validation_split=0.2,
+    batch_size=32,
+    epochs=50,   # 训练50轮
+    callbacks=[
+        tf.keras.callbacks.EarlyStopping(
+            monitor='val_loss',
+            patience=5,
+            restore_best_weights=True
+        )
+    ]
+)
+model_acc = model.evaluate(X_test, y_test, verbose=0)[1]
+
+print("Test Accuracy: {:.2f}%".format(model_acc * 100))
+from sklearn.metrics import confusion_matrix
+y_pred = np.array(list(map(lambda x: np.argmax(x), model.predict(X_test))))
+cm = confusion_matrix(y_test, y_pred)
+
+print("Confusion Matrix:\n",(cm))
+
+plt.figure(figsize=(8, 8))
+sns.heatmap(cm, annot=True, vmin=0, fmt='g', cbar=False, cmap='Blues')
+plt.xticks(np.arange(3) + 0.5, label_mapping.keys())
+plt.yticks(np.arange(3) + 0.5, label_mapping.keys())
+plt.xlabel("Predicted")
+plt.ylabel("Actual")
+plt.title("Confusion Matrix")
+plt.show()
+from sklearn.metrics import classification_report
+# Generate classification report
+clr = classification_report(y_test, y_pred, target_names=label_mapping.keys())
+print("Classification Report:\n----------------------\n", clr)
+# Save the trained model to a file
+model.save('save/gru_model.keras')

gru/run.py

@@ -0,0 +1,26 @@
+import numpy as np
+import pandas as pd
+import tensorflow as tf
+
+# Load the trained GRU model
+loaded_model = tf.keras.models.load_model('save/gru_model.keras')
+
+# Load new data for prediction
+new_data = pd.read_csv('input/data.csv')  # Replace with the path to your new data CSV file
+
+# Preprocess the new data (similar to how you preprocessed the training data)
+# Assuming the new_data has the same features as the training data
+X_new = new_data.drop('label', axis=1)
+
+# Make predictions on new data
+predicted_labels = loaded_model.predict_classes(X_new)
+
+# Map predicted labels back to emotions using the label_mapping dictionary
+reverse_label_mapping = {v: k for k, v in label_mapping.items()}
+predicted_emotions = [reverse_label_mapping[label] for label in predicted_labels]
+
+# Add predicted emotions to the new_data DataFrame
+new_data['predicted_emotion'] = predicted_emotions
+
+# Print the new_data DataFrame with predicted emotions
+print(new_data)

gru/save/gru_model.keras

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:374c6df9a7c7c64105c3c1b3e8a3d87cd688a84101599116ae44d1a272710b61
+size 25902232

input/emotions.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a51d6be460818bd6dd39ead707d7f2228c0ae8c4e88f8b79957e8853a0d5fdfa
+size 51200113