Create app.py

app.py

@@ -0,0 +1,88 @@
+## Imports
+import tensorflow as tf
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+from tensorflow.keras.preprocessing.text import Tokenizer
+from tensorflow.keras.preprocessing.sequence import pad_sequences
+import gradio
+
+## Load Data
+dataset = pd.read_csv('/content/drive/MyDrive/SPAMtextmessage.csv')
+
+## Data Preprocessing 
+# Convert ham to 0 and spam to 1
+dataset['Category']= dataset['Category'].str.replace('ham','0')
+dataset['Category']= dataset['Category'].str.replace('spam','1')
+dataset['Category']= dataset['Category'].astype(int)
+sentences = dataset['Message'].tolist()
+labels = dataset['Category'].tolist()
+# Separate out the sentences and labels into training and test sets
+training_size = int(len(sentences) * 0.8)
+# Sentence variables
+training_sentences = sentences[0:training_size]
+testing_sentences = sentences[training_size:]
+# Labels variables
+training_labels = labels[0:training_size]
+testing_labels = labels[training_size:]
+# Make labels into numpy arrays for use with the network later
+training_labels_final = np.array(training_labels)
+testing_labels_final = np.array(testing_labels)
+
+## Text Preprocessing
+vocab_size = 1000
+embedding_dim = 16
+max_length = 100
+trunc_type='post'
+padding_type='post'
+oov_tok = ""
+tokenizer = Tokenizer(num_words = vocab_size, oov_token=oov_tok)
+tokenizer.fit_on_texts(training_sentences)
+word_index = tokenizer.word_index
+sequences = tokenizer.texts_to_sequences(training_sentences)
+padded = pad_sequences(sequences,maxlen=max_length, padding=padding_type,
+                       truncating=trunc_type)
+testing_sequences = tokenizer.texts_to_sequences(testing_sentences)
+testing_padded = pad_sequences(testing_sequences,maxlen=max_length,
+                               padding=padding_type, truncating=trunc_type)
+
+## Modeling 
+# Set lr = 0.01
+model = tf.keras.Sequential([
+    tf.keras.layers.Embedding(vocab_size,embedding_dim,input_length=max_length),
+    tf.keras.layers.Flatten(),
+    tf.keras.layers.Dense(20,activation='relu'),
+    tf.keras.layers.Dense(10,activation= 'relu'),
+    tf.keras.layers.Dense(1,activation= 'sigmoid')
+])
+
+model.compile(loss='binary_crossentropy',metrics=['accuracy'],
+              optimizer=tf.keras.optimizers.Adam(learning_rate=0.01))
+model.fit(padded,training_labels_final,batch_size=128,epochs=50,
+           validation_data=(testing_padded,testing_labels_final))
+
+## Gradio App
+def spam_detection(message):
+    # Preprocess the input message
+    sequence = tokenizer.texts_to_sequences([message])
+    padded_sequence = pad_sequences(sequence, maxlen=max_length, padding=padding_type, truncating=trunc_type)
+    
+    # Make prediction
+    prediction = model.predict(padded_sequence)[0, 0]
+    
+    # Return the result
+    return "Spam" if prediction >= 0.5 else "Not Spam"
+
+# Gradio Interface
+iface = gr.Interface(
+    fn=spam_detection,
+    inputs=gr.Textbox(prompt="Enter a message:"),
+    outputs="text",
+    live=True,
+    theme="huggingface",
+    title="Spam Message Detection",
+    description="A demo app for learning purposes. Detects spam messages with 98% accuracy based on the dataset."
+)
+
+# Launch the app
+iface.launch()