Update app.py

app.py

@@ -13,9 +13,6 @@ from llama_cpp import Llama
 from huggingface_hub import login
 login(token = os.getenv('HF_TOKEN'))
 
-# subprocess.run('pip install llama-cpp-python==0.2.75 --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cu124', shell=True)
-# subprocess.run('pip install llama-cpp-agent==0.2.10', shell=True)
-
 repo_id = "srijaydeshpande/Deid-Fine-Tuned"
 model_id = "deid_finetuned.Q4_K_M.gguf"
 
@@ -26,18 +23,6 @@ hf_hub_download(
     local_dir = "./models"
 )
 
-# hf_hub_download(
-#     repo_id="QuantFactory/Meta-Llama-3-8B-Instruct-GGUF",
-#     filename="Meta-Llama-3-8B-Instruct.Q8_0.gguf",
-#     local_dir = "./models"
-# )
-
-# hf_hub_download(
-#     repo_id="bartowski/Meta-Llama-3-70B-Instruct-GGUF",
-#     filename="Meta-Llama-3-70B-Instruct-Q3_K_M.gguf",
-#     local_dir = "./models"
-# )
-
 def process_document(pdf_path):
     extracted_pages = extract_pages(pdf_path)
     page2content = {}
@@ -139,15 +124,12 @@ def deidentify_doc(llm_type, pdftext, maxtokens, temperature, top_probability):
         
     else:
         client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
-        messages = [{"role": "assistant", "content": 'Help answeing questions'}]
-        # messages.append({"role": "user", "content": str(pdftext)})
-        messages.append({"role": "user", "content": "How are you doing?"})
+        messages = [{"role": "assistant", "content": prompt}]
+        messages.append({"role": "user", "content": pdftext})
         response = ""
-        print('PROMPT IS ',messages)
         for message in client.chat_completion(
             messages,
             max_tokens=max_tokens,
-            # stream=True,
             temperature=temperature,
             top_p=top_probability,
         ):