Update app.py

app.py

@@ -38,47 +38,39 @@ model_race.to(device)
 model_race.load_state_dict(torch.load('best_model_race.pt', map_location=torch.device('cpu')))
 
 
-def predict(text):
-    sentences = [
-        text
-    ]
+max_textboxes = 10
 
-    tokenizer = AutoTokenizer.from_pretrained("vinai/bertweet-base", normalization=True)
 
-    encoded_sentences = tokenizer(
-        sentences,
-        padding=True,
-        truncation=True,
-        return_tensors='pt',
-        max_length=128,
-    )
+def update_textboxes(k):
+    components = []
+    if k is None:
+        k = 0
+    for i in range(max_textboxes):
+        if i < k:
+            components.append(gr.update(visible=True))
+        else:
+            components.append(gr.update(visible=False))
+    return components
 
-    input_ids = encoded_sentences["input_ids"].to(device)
-    attention_mask = encoded_sentences["attention_mask"].to(device)
 
-    model_race.eval()
-    with torch.no_grad():
-        outputs = model_race(input_ids, attention_mask)
-        probs = torch.nn.functional.softmax(outputs, dim=1)
-        predictions = torch.argmax(outputs, dim=1)
-        predictions = predictions.cpu().numpy()
+def clear_textboxes():
+    return [gr.update(value='') for _ in range(max_textboxes)]
 
-    output_string = ""
-    for i, prob in enumerate(probs[0]):
-        print(f"{labels[i]}: %{round(prob.item() * 100, 2)}")
-        output_string += f"{labels[i]}: %{round(prob.item() * 100, 2)}\n"
 
-    print(labels[predictions[0]])
-    output_string += f"Predicted as: {labels[predictions[0]]}"
+with gr.Blocks() as demo:
+    with gr.Row():
+        with gr.Column(scale=1):
+            s = gr.Slider(1, max_textboxes, value=1, step=1, label="How many tweets do you want to enter:")
+            textboxes = [gr.Textbox(label=f"Tweet {i + 1}", visible=(i == 0)) for i in range(max_textboxes)]
+            s.change(fn=update_textboxes, inputs=s, outputs=textboxes)
+            btn = gr.Button("Predict")
+        with gr.Column(scale=1):
+            output = gr.Textbox(label="Profile of User")
 
-    return output_string
-
-
-demo = gr.Interface(
-    fn=predict,
-    inputs=["text"],
-    outputs=["text"],
-)
+    btn.click(fn=predict, inputs=textboxes, outputs=output)
+    btn_clear = gr.Button("Clear")
+    btn_clear.click(fn=clear_textboxes, outputs=textboxes)
 
 demo.launch()
 
+