Update app.py

app.py

@@ -1,45 +1,71 @@
 import gradio as gr
 from transformers import pipeline
 
+# 감성 분석 파이프라인 초기화
 sentiment = pipeline("sentiment-analysis")
 
+# 사용자 입력에 대한 감성 분석 결과를 반환하는 함수
 def get_sentiment(입력):
-    # from transformers import AutoTokenizer, AutoModelForCausalLM
-    # model_name = "heegyu/koalpaca-355m"
-    # tokenizer = AutoTokenizer.from_pretrained(model_name)
-    # tokenizer.truncation_side = "right"
-    # model = AutoModelForCausalLM.from_pretrained(model_name)
-    return sentiment(입력)
-
-def get_response(output):
-    context = f"<usr>{context}\n<sys>"
-    inputs = tokenizer(
-        context, 
-        truncation=True,
-        max_length=512,
-        return_tensors="pt")
+    # 감성 분석 실행
+    result = sentiment(입력)
+    # 결과 포맷팅 및 반환
+    return result[0]
+
+# Gradio 앱 인터페이스 구성
+iface = gr.Interface(
+    fn=get_sentiment,  # 실행할 함수
+    inputs=gr.inputs.Textbox(lines=2, placeholder="여기에 텍스트를 입력하세요..."),  # 입력란 설정
+    outputs="json",  # 출력 형태
+    clear_on_submit=True,  # 제출 후 입력란 클리어
+    title="텍스트 감성 분석",  # UI 제목
+    description="텍스트를 입력하고 제출 버튼을 클릭하여 감성 분석 결과를 확인하세요."  # UI 설명
+)
+
+# Gradio 앱 실행
+iface.launch()
+
+# import gradio as gr
+# from transformers import pipeline
+
+# sentiment = pipeline("sentiment-analysis")
+
+# def get_sentiment(입력):
+#     # from transformers import AutoTokenizer, AutoModelForCausalLM
+#     # model_name = "heegyu/koalpaca-355m"
+#     # tokenizer = AutoTokenizer.from_pretrained(model_name)
+#     # tokenizer.truncation_side = "right"
+#     # model = AutoModelForCausalLM.from_pretrained(model_name)
+#     return sentiment(입력)
+
+# def get_response(output):
+#     context = f"<usr>{context}\n<sys>"
+#     inputs = tokenizer(
+#         context, 
+#         truncation=True,
+#         max_length=512,
+#         return_tensors="pt")
     
-    generation_args = dict(
-        max_length=256,
-        min_length=64,
-        eos_token_id=2,
-        do_sample=True,
-        top_p=1.0,
-        early_stopping=True
-    )
-
-    outputs = model.generate(**inputs, **generation_args)
-    response = tokenizer.decode(outputs[0])
-    print(context)
-    print(response)
-    response = response[len(context):].replace("</s>", "")
-
-    return response
-
-model, tokenizer = get_pipe()
-
-def ask_question(input_):
-    response = get_response(tokenizer, model, input_)
-    return response
-
-gr.Interface(fn=ask_question, inputs="text", outputs="text", title="KoAlpaca-355M", description="한국어로 질문하세요.").launch()
+#     generation_args = dict(
+#         max_length=256,
+#         min_length=64,
+#         eos_token_id=2,
+#         do_sample=True,
+#         top_p=1.0,
+#         early_stopping=True
+#     )
+
+#     outputs = model.generate(**inputs, **generation_args)
+#     response = tokenizer.decode(outputs[0])
+#     print(context)
+#     print(response)
+#     response = response[len(context):].replace("</s>", "")
+
+#     return response
+
+# model, tokenizer = get_pipe()
+
+# def ask_question(input_):
+#     response = get_response(tokenizer, model, input_)
+#     return response
+
+# gr.Interface(fn=ask_question, inputs="text", outputs="text", title="KoAlpaca-355M", description="한국어로 질문하세요.").launch()