Update app.py

app.py

@@ -48,14 +48,17 @@ def load_model(_docs):
     text_splitter = RecursiveCharacterTextSplitter(chunk_size=1024, chunk_overlap=256)
     texts = text_splitter.split_documents(docs)
     db = FAISS.from_documents(texts, embeddings)
-    model_name_or_path = "/home/user/app/Llama-2-13B-chat-GPTQ/"
+    #model_name_or_path = "/home/user/app/Llama-2-13B-chat-GPTQ/"
+    model_name_or_path = "/home/user/app/codeLlama"
+
     model_basename = "model"
 
     tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)
 
     model = AutoGPTQForCausalLM.from_quantized(
         model_name_or_path,
-        revision="gptq-8bit-128g-actorder_False",
+        #revision="gptq-8bit-128g-actorder_False",
+        revision="gptq-8bit-128g-actorder_True",
         model_basename=model_basename,
         use_safetensors=True,
         trust_remote_code=True,
@@ -82,7 +85,6 @@ def load_model(_docs):
 
     DEFAULT_SYSTEM_PROMPT = """
     You are a helpful, respectful and honest assistant with knowledge of machine learning, data science, computer science, Python programming language, mathematics, probability and statistics.
-    Take a deep breath and work on the given problem step-by-step. 
     """.strip()
 
     def generate_prompt(prompt: str, system_prompt: str = DEFAULT_SYSTEM_PROMPT) -> str:
@@ -99,9 +101,12 @@ def load_model(_docs):
                              streamer=streamer,)
     llm = HuggingFacePipeline(pipeline=text_pipeline, model_kwargs={"temperature": 0.5})
 
-    SYSTEM_PROMPT = ("Use the following pieces of context to answer the question at the end. "
-                     "If you don't know the answer, just say that you don't know, "
-                     "don't try to make up an answer.")
+    # SYSTEM_PROMPT = ("Use the following pieces of context to answer the question at the end. "
+    #                  "If you don't know the answer, just say that you don't know, "
+    #                  "don't try to make up an answer.")
+    SYSTEM_PROMPT = ("Use the following pieces of context along with general information you possess to answer the question at the end. "
+                 "If you don't know the answer, just say that you don't know, "
+                 "don't try to make up an answer.")
 
     template = generate_prompt("""{context}  Question: {question} """,system_prompt=SYSTEM_PROMPT,) #Enter memory here!
     prompt = PromptTemplate(template=template, input_variables=["context",  "question"]) #Add history here