Update app.py

app.py

@@ -1,35 +1,27 @@
-from fastapi import FastAPI
-from fastapi.middleware.cors import CORSMiddleware
-from pydantic import BaseModel
-from transformers import pipeline
 import gradio as gr
+from transformers import pipeline
 
 print("Loading models...")
 quality_pipe = pipeline("text-classification", model="FareehaAly/fator-argument-quality", top_k=None)
 fallacy_pipe = pipeline("text-classification", model="FareehaAly/fator-fallacy-detector", top_k=None)
 print("Models loaded!")
 
-app = FastAPI()
-app.add_middleware(CORSMiddleware, allow_origins=["*"], allow_methods=["*"], allow_headers=["*"])
-
-class Input(BaseModel):
-    text: str
+def predict_quality(text):
+    return quality_pipe(text[:512])
 
-@app.post("/quality")
-def quality(inp: Input):
-    return quality_pipe(inp.text[:512])
-
-@app.post("/fallacy")
-def fallacy(inp: Input):
-    return fallacy_pipe(inp.text[:512])
-
-def q(text): return quality_pipe(text[:512])
-def f(text): return fallacy_pipe(text[:512])
+def predict_fallacy(text):
+    return fallacy_pipe(text[:512])
 
 with gr.Blocks() as demo:
-    gr.Markdown("## FATOR NLP API")
-    t = gr.Textbox(label="Input text")
-    gr.Button("Run Quality").click(q, t, gr.JSON(label="Quality result"))
-    gr.Button("Run Fallacy").click(f, t, gr.JSON(label="Fallacy result"))
-
-app = gr.mount_gradio_app(app, demo, path="/")
+    gr.Markdown("## FATOR NLP API - Running")
+    with gr.Row():
+        txt = gr.Textbox(label="Input text", scale=4)
+    with gr.Row():
+        btn_q = gr.Button("Quality")
+        btn_f = gr.Button("Fallacy")
+    out_q = gr.JSON(label="Quality Result")
+    out_f = gr.JSON(label="Fallacy Result")
+    btn_q.click(predict_quality, inputs=txt, outputs=out_q, api_name="quality")
+    btn_f.click(predict_fallacy, inputs=txt, outputs=out_f, api_name="fallacy")
+
+demo.launch(server_name="0.0.0.0", server_port=7860)