Upload 8 Files

.gitattributes

@@ -1,35 +1,38 @@
-*.7z filter=lfs diff=lfs merge=lfs -text
-*.arrow filter=lfs diff=lfs merge=lfs -text
-*.bin filter=lfs diff=lfs merge=lfs -text
-*.bz2 filter=lfs diff=lfs merge=lfs -text
-*.ckpt filter=lfs diff=lfs merge=lfs -text
-*.ftz filter=lfs diff=lfs merge=lfs -text
-*.gz filter=lfs diff=lfs merge=lfs -text
-*.h5 filter=lfs diff=lfs merge=lfs -text
-*.joblib filter=lfs diff=lfs merge=lfs -text
-*.lfs.* filter=lfs diff=lfs merge=lfs -text
-*.mlmodel filter=lfs diff=lfs merge=lfs -text
-*.model filter=lfs diff=lfs merge=lfs -text
-*.msgpack filter=lfs diff=lfs merge=lfs -text
-*.npy filter=lfs diff=lfs merge=lfs -text
-*.npz filter=lfs diff=lfs merge=lfs -text
-*.onnx filter=lfs diff=lfs merge=lfs -text
-*.ot filter=lfs diff=lfs merge=lfs -text
-*.parquet filter=lfs diff=lfs merge=lfs -text
-*.pb filter=lfs diff=lfs merge=lfs -text
-*.pickle filter=lfs diff=lfs merge=lfs -text
-*.pkl filter=lfs diff=lfs merge=lfs -text
-*.pt filter=lfs diff=lfs merge=lfs -text
-*.pth filter=lfs diff=lfs merge=lfs -text
-*.rar filter=lfs diff=lfs merge=lfs -text
-*.safetensors filter=lfs diff=lfs merge=lfs -text
-saved_model/**/* filter=lfs diff=lfs merge=lfs -text
-*.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tar filter=lfs diff=lfs merge=lfs -text
-*.tflite filter=lfs diff=lfs merge=lfs -text
-*.tgz filter=lfs diff=lfs merge=lfs -text
-*.wasm filter=lfs diff=lfs merge=lfs -text
-*.xz filter=lfs diff=lfs merge=lfs -text
-*.zip filter=lfs diff=lfs merge=lfs -text
-*.zst filter=lfs diff=lfs merge=lfs -text
-*tfevents* filter=lfs diff=lfs merge=lfs -text
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+Models/Roberta-base-5_3k-epoch2/*.safetensors filter=lfs diff=lfs merge=lfs -text
+Models/Roberta-base-5_3k-epoch2/*.json filter=lfs diff=lfs merge=lfs -text
+Models/Roberta-base-5_3k-epoch2/*.txt filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,2 @@
+.env
+__pycache__

Dockerfile

@@ -0,0 +1,6 @@
+FROM python:3.10-slim
+WORKDIR /app
+COPY . .
+RUN pip install --no-cache-dir -r requirements.txt
+EXPOSE 7860
+CMD ["uvicorn","app:app","--port","7860"]

README.md

@@ -1,12 +1,12 @@
----
-title: HallucinationDetectionServer
-emoji: 🏃
-colorFrom: yellow
-colorTo: blue
-sdk: gradio
-sdk_version: 5.45.0
-app_file: app.py
-pinned: false
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+---
+title: HallucinationDetectionServer
+emoji: 🏃
+colorFrom: yellow
+colorTo: blue
+sdk: gradio
+sdk_version: 5.45.0
+app_file: app.py
+pinned: false
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

Routes.py

@@ -0,0 +1,113 @@
+from fastapi.routing import APIRouter
+from pydantic import BaseModel
+from Pipeline.HallucinationPipeline import HallucinationPipeline
+from Pipeline.CorrectionLLMs import DeepseekAPI
+from nltk.tokenize import sent_tokenize
+from utils import detectionProcess
+import os
+
+
+deepseek_apikey="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"
+
+
+router=APIRouter(prefix="/api")
+pipeline=HallucinationPipeline("Razor2507/Roberta-Base-Finetuned","cpu")
+deepseek=DeepseekAPI(api_key=deepseek_apikey)
+
+# {'predictions': [0],
+#  'corrected_summary': [],
+#  'sent_predicted': [array([2, 0, 0, 0])],
+#  'factual_score': [0.21776947937905788],
+#  'contradiction_score': [0.7815729230642319]}
+
+
+
+# Detection Endpoint
+class DetectionRequest(BaseModel):
+    article:str
+    summary:str
+    arbiter:str
+
+@router.post("/detect")
+def detect(data:DetectionRequest):
+    try:
+        article=data.article.strip().replace("\n"," ").replace("\t"," ")
+        summary=data.summary.strip().replace("\n"," ").replace("\t"," ")
+        arbiter=True if data.arbiter=="on" else False
+        print("Arbiter : ",arbiter)
+        result=detectionProcess(article=article,summary=summary,pipeline=pipeline,arbiter=arbiter)
+        result["status"]=200
+        return result
+    except Exception as e:
+        print(e)
+        return {"status":404}
+
+
+
+#  Correction Endpoint
+class correctionRequest(BaseModel):
+    article:str
+    tag_summary:str
+    model:str
+
+@router.post("/correct")
+def correct(data:correctionRequest):
+    try:
+        
+        if data.model=="mistral":
+            pass
+        elif data.model=="gemini":
+            pass
+        elif data.model=="deepseek":
+            correction=deepseek.correct(premise=data.article, summary=data.tag_summary)
+        
+        print(correction)
+
+        result=detectionProcess(article=data.article,summary=correction,pipeline=pipeline)
+        result["corrected_summary"]=correction
+        result["status"]=200
+        return result
+    except Exception as e:
+        print(e)
+        return {"status":404}
+    
+
+@router.get("/test")
+def keyTest():
+    print("Testing ",os.getenv("deepseek_apikey"))
+    return {"msg":"testing_works"}
+
+
+
+
+
+
+
+# @router.post("/detect")
+# def detect(data:DetectionRequest):
+    # article=data.article.strip().replace("\n","").replace("\t"," ")
+    # summary=data.summary.strip().replace("\n","").replace("\t"," ")
+
+#     result=pipeline.process([[article,summary]],correct_the_summary=False)
+#     all_sentences=sent_tokenize(summary)
+#     print(result)
+#     summary=pipeline.addTags(all_sentences,result["sent_predicted"][0],len(all_sentences))
+#     score=str(result["factual_score"][0])
+#     sentenceLabels=list(result["sent_predicted"][0])
+#     labelCounts=[sentenceLabels.count(0),sentenceLabels.count(2)]
+  
+#     prompt=f"""
+# Here is a summary with hallucinated parts marked using <xx> tags.
+
+# Please correct only the text inside the <xx> tags to make it factually accurate based on the original article. Leave the rest of the summary unchanged and remove the <xx> tags after correction.
+
+# Return the summary with hallucinated parts fixed and you can remove those <xx></xx> tags. Don't remove that entire sentence.
+            
+# Original Article:
+#     {data.article}
+
+# Summary:
+#     {summary}
+
+# """
+    # return {"summary":summary,"score":score,"counts":labelCounts,"copy_prompt":prompt}

app.py

@@ -0,0 +1,20 @@
+from fastapi import FastAPI
+from Routes import router
+from fastapi.middleware.cors import CORSMiddleware
+from fastapi.staticfiles import StaticFiles
+from fastapi.responses import FileResponse
+from dotenv import load_dotenv
+
+
+load_dotenv()
+
+
+app=FastAPI()
+app.include_router(router)
+app.add_middleware(CORSMiddleware, allow_origins=["*"], allow_methods=["*"], allow_credentials=True)
+app.mount("/assets",StaticFiles(directory="dist/assets"),name="assets")
+
+
+@app.get("/{full_path:path}")
+async def main():
+    return FileResponse("dist/index.html")

requirements.txt

@@ -0,0 +1,15 @@
+fastapi==0.111.0
+uvicorn==0.29.0
+pydantic==2.1.1
+python-dotenv==1.0.1
+requests==2.32.3
+nltk==3.9.1
+transformers==4.53.2
+torch==2.5.1+cu121
+xgboost==3.0.0
+scikit-learn==1.5.1
+numpy==1.26.4
+pandas==2.2.3
+openai==1.3.7
+google-generativeai==0.8.5
+llama-cpp-python==0.3.12

utils.py

@@ -0,0 +1,30 @@
+from nltk.tokenize import sent_tokenize
+
+
+# This method is created because it will be required for 
+# Detection as well as Correction (Like results after correction)
+
+def detectionProcess(article,summary,pipeline,arbiter):
+    result=pipeline.process([[article,summary]],correct_the_summary=False,arbiter=arbiter)
+    all_sentences=sent_tokenize(summary)
+    print(result)
+    summary=pipeline.addTags(all_sentences,result["sent_predicted"][0],len(all_sentences))
+    score=str(result["factual_score"][0])
+    sentenceLabels=list(result["sent_predicted"][0])
+    labelCounts=[sentenceLabels.count(0),sentenceLabels.count(2)]
+  
+    prompt=f"""
+Here is a summary with hallucinated parts marked using <xx> tags.
+
+Please correct only the text inside the <xx> tags to make it factually accurate based on the original article. Leave the rest of the summary unchanged and remove the <xx> tags after correction.
+
+Return the summary with hallucinated parts fixed and you can remove those <xx></xx> tags. Don't remove that entire sentence.
+            
+Original Article:
+    {article}
+
+Summary:
+    {summary}
+
+"""
+    return {"summary":summary,"score":score,"counts":labelCounts,"copy_prompt":prompt}