Update app.py

app.py

@@ -0,0 +1,45 @@
+import gradio as gr
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+import torch
+import numpy as np
+
+# Load the model and tokenizer
+model = AutoModelForSequenceClassification.from_pretrained('kietnt0603/bertweet-base-hate-speech-offensive')
+tokenizer = AutoTokenizer.from_pretrained('kietnt0603/bertweet-base-hate-speech-offensive')
+
+# Define the labels
+labels = ["Hate", "Offensive", "Neither"]
+
+# Function for prediction
+def predict(inputs: str) -> Dict[str, Any]:
+    # Tokenize input text
+    inputs_dict = tokenizer(inputs, return_tensors="pt")
+
+    # Forward pass
+    with torch.no_grad():
+        outputs = model(**inputs_dict)
+
+    # Softmax to get probabilities
+    probabilities = torch.nn.functional.softmax(outputs.logits, dim=1)
+
+    # Get probabilities for each label
+    label_probabilities = {label: round(prob.item(), 4) for label, prob in zip(labels, probabilities[0].tolist())}
+
+    # Return the result
+    return label_probabilities
+
+# Create title and description for the task
+title = "Text Classification Demo"
+description = "Classify text into categories: Hate, Offensive, Neither"
+article = "Model loaded from https://huggingface.co/kietnt0603/bertweet-base-hate-speech-offensive"
+
+# Create the Gradio interface
+iface = gr.Interface(fn=predict,
+                     inputs="textbox",
+                     outputs="dictionary",
+                     title=title,
+                     description=description,
+                     article=article)
+
+# Launch the interface
+iface.launch()