Create app.py

app.py

@@ -0,0 +1,43 @@
+import gradio as gr
+from transformers import AutoModelForSequenceClassification, AutoTokenizer
+import torch
+
+# Load the pre-trained model and tokenizer
+model_name = "distilbert-base-uncased"  # Replace this with the desired model
+model = AutoModelForSequenceClassification.from_pretrained(model_name)
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+
+# Define a function for sentiment analysis
+def predict_sentiment(text):
+    # Tokenize the input text and prepare it to be used by the model
+    inputs = tokenizer(text, return_tensors="pt", padding=True, truncation=True)
+
+    # Forward pass through the model
+    with torch.no_grad():
+        outputs = model(**inputs)
+
+    # Get the predicted probabilities and convert them to percentages
+    probabilities = torch.softmax(outputs.logits, dim=1).squeeze().tolist()
+    positive_percent = probabilities[2] * 100
+    negative_percent = probabilities[0] * 100
+    neutral_percent = probabilities[1] * 100
+
+    # Construct the result dictionary
+    result = {
+        "Positive": round(positive_percent, 2),
+        "Negative": round(negative_percent, 2),
+        "Neutral": round(neutral_percent, 2)
+    }
+
+    return result
+
+iface = gr.Interface(
+    fn=predict_sentiment,
+    inputs=gr.inputs.Textbox(lines=10, label="Enter financial statement"),
+    outputs=gr.outputs.Label(num_top_classes=3, label="Sentiment Percentages"),
+    title="Financial Statement Sentiment Analysis",
+    description="Predict the sentiment percentages of a financial statement."
+)
+
+if __name__ == "__main__":
+    iface.launch()