modle_change

app.py

@@ -1,68 +1,34 @@
-
-from transformers import pipeline
+from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
+import torch
 import gradio as gr
 import easyocr
 import pdfplumber
-import random 
-
-
-import os
-os.system("rm -rf /home/user/.cache/huggingface")
-
-
-#適用於Interface、Block
-title = "<h1>產生英文題目</h1>"
-description = """這是一個利用hugging face 產生英文題目的小專案"""
-textbox = gr.Textbox(label="請輸入英文文章:", placeholder="While lily is setting...", lines=5)
-
-#加入磚
-demo = gr.Blocks()
-
-# 加載 Hugging Face 上的問答模型
-question_generator = pipeline("text2text-generation", model="valhalla/t5-base-qg-hl")
+import random
 
-# def question_generate(context):
-#     # 讓模型根據文章生成問題
-#     question_result = []
-#     for i in range(5):
-#         question = question_generator(f"question: {context}",
-#             max_length=100,
-#             do_sample = True, # 啟用採樣以增加多樣性
-#             temperature=0.8 + (i * 0.1),  # 逐漸增加溫度參數來獲得更多樣的結果
-#             top_p=0.9
-#                         )
-#         question_result.append(f"Q{i+1}. {question[0]['generated_text']}")
+# 載入本地模型與 tokenizer
+MODEL_PATH = "models/t5-base-qg-hl"
+tokenizer = AutoTokenizer.from_pretrained(MODEL_PATH)
+model = AutoModelForSeq2SeqLM.from_pretrained(MODEL_PATH)
 
-#     return "\n".join(question_result)  #以換行符號返回多個問題
+def generate_text(prompt, max_length=100, temperature=0.8, top_p=0.9):
+    inputs = tokenizer(prompt, return_tensors="pt")
+    with torch.no_grad():
+        outputs = model.generate(**inputs, max_length=max_length, do_sample=True, temperature=temperature, top_p=top_p)
+    return tokenizer.decode(outputs[0], skip_special_tokens=True)
 
 def question_generator_with_answer(context):
-
-    #產生題目
-    question_data = question_generator(f"question:{context}", 
-    max_length=100, do_sample=True, temperature=0.8, top_p=0.9)
-    question = question_data[0]['generated_text']
- 
-    #產生正確答案
-    answer_data = question_generator(f"answer:{context}", 
-    max_length=100, do_sample=True, temperature=1, top_p=0.9)
-    correct_answer = answer_data[0]['generated_text']
-
-    #產生錯誤答案
+    question = generate_text(f"question: {context}")
+    correct_answer = generate_text(f"answer: {context}", temperature=1.0)
+    
     wrong_answers = set()
     while len(wrong_answers) < 3:
-        wrong_data = question_generator(f"answer: {context}", max_length=50, 
-                    do_sample=True, temperature=1.0, top_p=0.8)
-        wrong_answer = wrong_data[0]['generated_text']
-        if wrong_answer != correct_answer and "?" not in wrong_answer:  # 避免重複正確答案
+        wrong_answer = generate_text(f"answer: {context}", max_length=50, temperature=1.0, top_p=0.8)
+        if wrong_answer != correct_answer and "?" not in wrong_answer:
             wrong_answers.add(wrong_answer)
-
     
-    # 將正確答案加入選項，並打亂順序
     choices = list(wrong_answers) + [correct_answer]
     random.shuffle(choices)
-
-
-    # 回傳題目與選項
+    
     return {
         "question": question,
         "choices": choices,
@@ -70,114 +36,57 @@ def question_generator_with_answer(context):
     }
 
 def format_question_output(context):
-    question_result=[]
+    question_result = []
     for j in range(4):
         result = question_generator_with_answer(context)
         question_text = f"{result['question']}\n"
         choices_text = "\n".join([f"{chr(65+i)}. {choice}" for i, choice in enumerate(result['choices'])])
         question_result.append(f"\nQ{j+1}.{question_text}\n{choices_text}\n")
-    return "\n".join(question_result) #用換行來連接
-
-# def format_question_output(context):
-#     result = question_generator_with_answer(context)
-#     question_text = f"**{result['question']}**\n\n"
-#     choices_text = "\n".join([f"{chr(65+i)}. {choice}" for i, choice in enumerate(result['choices'])])
-#     return f"{question_text}\n{choices_text}\n\n✅ 正確答案: {result['correct_answer']}"
- 
+    return "\n".join(question_result)
 
-#pdf辨識
 def extract_text_from_pdf(pdf_path):
     text = ""
-    with pdfplumber.open(pdf_path.name) as pdf:  
+    with pdfplumber.open(pdf_path.name) as pdf:
         for page in pdf.pages:
             text += page.extract_text() + "\n"
-    ls = format_question_output(text)
-    return ls
+    return format_question_output(text)
 
-
-
-#圖片辨識(辨識度太低)
 def OCR(photo):
-    text_inner = ""
-    questions = []
     reader = easyocr.Reader(['en', 'ch_tra'])
     results = reader.readtext(photo)
-    for (bbox, text, prob) in results:
-        text_inner += text
-    return text_inner
+    return "".join([text for (_, text, _) in results])
 
-
-#確認辨識結果沒有問題後，產生題目
 def OCR_gen(text):
-    if not text.strip():  # 確保輸入的 text 不是空的
+    if not text.strip():
         return "錯誤：OCR 沒有輸出任何��用的文字，請重新檢查圖片內容。"
-    ls = format_question_output(text)
-    return ls
-
+    return format_question_output(text)
 
+demo = gr.Blocks()
 with demo:
-    gr.Markdown(title)
-    gr.Markdown(description)
+    gr.Markdown("<h1>產生英文題目</h1>")
+    gr.Markdown("這是一個利用 hugging face 產生英文題目的小專案")
+    
     with gr.Tabs():
         with gr.TabItem("輸入文字"):
-            with gr.Row():
-                text_input = gr.Textbox(label="請輸入英文文章:", placeholder="While lily is setting...", lines=5)
-                with gr.Column():
-                    text_output = gr.Textbox(label="題目")
-                    text_button = gr.Button("產生題目")
+            text_input = gr.Textbox(label="請輸入英文文章:", placeholder="While lily is setting...", lines=5)
+            text_output = gr.Textbox(label="題目")
+            text_button = gr.Button("產生題目")
+        
         with gr.TabItem("PDF文件辨識"):
-            with gr.Row():
-                PDF_input = gr.File(label="請上傳PDF文件")
-                with gr.Column():
-                    PDF_output = gr.Textbox()
-                    PDF_button = gr.Button("產生題目")
+            PDF_input = gr.File(label="請上傳PDF文件")
+            PDF_output = gr.Textbox()
+            PDF_button = gr.Button("產生題目")
+        
         with gr.TabItem("圖片辨識"):
-            with gr.Row():
-                image_input = gr.Image()
-                #解析圖片文字
-                with gr.Column():
-                    img_tem = gr.Textbox(placeholder="請確認辨識結果",label="辨識結果")
-                    img_button = gr.Button("開始解析")
-                    image_button = gr.Button("產生題目")
-                # #產生題目
-                with gr.Column():
-                    image_output = gr.Textbox(label="題目")
-
-    #判別有沒有輸入文章
-    def validate_and_generate(text):
-        if not text.strip():
-            return "請輸入文章以產生題目"
-        return format_question_output(text)
-
-    #文字輸入                 物件
-    text_button.click(validate_and_generate, inputs=text_input, outputs=text_output)
-    
-    #判別有沒有上傳檔案
-    def test_PDF(file):
-        if not file:
-            return "請上傳PDF文件以產生題目"
-        return extract_text_from_pdf(file)
-
-    #PDF輸入
-    PDF_button.click(test_PDF, inputs=PDF_input, outputs=PDF_output)
-
-    #判別有沒有上傳照片
-    def test_image(image):
-        if image is None:
-            return "請上傳圖片以產生題目"
-        return OCR(image)
-
-    #辨識文章
-    img_button.click(test_image, inputs=image_input, outputs=img_tem)
-
-
-    #檢查辨識結果有沒有存在
-    def test_finished(text):
-        if (not text.strip() or text == "請上傳圖片以產生題目"):
-            return "請確認文章已經輸入"
-        return OCR_gen(text)
-    image_button.click(test_finished, inputs=img_tem, outputs=image_output)
-
-
+            image_input = gr.Image()
+            img_tem = gr.Textbox(placeholder="請確認辨識結果", label="辨識結果")
+            img_button = gr.Button("開始解析")
+            image_button = gr.Button("產生題目")
+            image_output = gr.Textbox(label="題目")
+
+    text_button.click(format_question_output, inputs=text_input, outputs=text_output)
+    PDF_button.click(extract_text_from_pdf, inputs=PDF_input, outputs=PDF_output)
+    img_button.click(OCR, inputs=image_input, outputs=img_tem)
+    image_button.click(OCR_gen, inputs=img_tem, outputs=image_output)
 
 demo.launch()