Update app.py

app.py

@@ -1,63 +1,106 @@
+from langchain import PromptTemplate, LLMChain
+from langchain.llms import CTransformers
+import os
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain.vectorstores import Chroma
+from langchain.chains import RetrievalQA
+from langchain.embeddings import HuggingFaceBgeEmbeddings
+from io import BytesIO
+from langchain.document_loaders import PyPDFLoader
 import gradio as gr
-from huggingface_hub import InferenceClient
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 
+local_llm = "final_model_maybe_gguf-unsloth.Q5_K_M.gguf"
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+config = {
+'max_new_tokens': 2048,
+'repetition_penalty': 1.1,
+'temperature': 0.6,
+'top_k': 50,
+'top_p': 0.9,
+'stream': True,
+'threads': int(os.cpu_count() / 2)
+}
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+llm = CTransformers(
+    model=local_llm,
+    model_type="mistral",
+    lib="avx2", #for CPU use
+    **config
+)
 
-    messages.append({"role": "user", "content": message})
+print("LLM Initialized...")
 
-    response = ""
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+prompt_template = """Use the following pieces of information to answer the user's question.
+If you don't know the answer, just say that you don't know, don't try to make up an answer.
 
-        response += token
-        yield response
+Context: {context}
+Question: {question}
 
+Only return the helpful answer below and nothing else.
+Helpful answer:
 """
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
+
+model_name = "ko-sroberta-multitask"
+model_kwargs = {'device': 'cpu'}
+encode_kwargs = {'normalize_embeddings': False}
+embeddings = HuggingFaceBgeEmbeddings(
+    model_name=model_name,
+    model_kwargs=model_kwargs,
+    encode_kwargs=encode_kwargs
 )
 
 
-if __name__ == "__main__":
-    demo.launch()
+prompt = PromptTemplate(template=prompt_template, input_variables=['context', 'question'])
+load_vector_store = Chroma(persist_directory="stores/pet_cosine", embedding_function=embeddings)
+retriever = load_vector_store.as_retriever(search_kwargs={"k":1})
+# query = "what is the fastest speed for a greyhound dog?"
+# semantic_search = retriever.get_relevant_documents(query)
+# print(semantic_search)
+
+print("######################################################################")
+
+chain_type_kwargs = {"prompt": prompt}
+
+# qa = RetrievalQA.from_chain_type(
+#     llm=llm,
+#     chain_type="stuff",
+#     retriever=retriever,
+#     return_source_documents = True,
+#     chain_type_kwargs= chain_type_kwargs,
+#     verbose=True
+# )
+
+# response = qa(query)
+
+# print(response)
+
+sample_prompts = ["what is the fastest speed for a greyhound dog?", "Why should we not feed chocolates to the dogs?", "Name two factors which might contribute to why some dogs might get scared?"]
+
+def get_response(input):
+  query = input
+  chain_type_kwargs = {"prompt": prompt}
+  qa = RetrievalQA.from_chain_type(llm=llm, chain_type="stuff", retriever=retriever, return_source_documents=True, chain_type_kwargs=chain_type_kwargs, verbose=True)
+  response = qa(query)
+  return response
+
+input = gr.Text(
+                label="Prompt",
+                show_label=False,
+                max_lines=1,
+                placeholder="Enter your prompt",
+                container=False,
+            )
+
+iface = gr.Interface(fn=get_response, 
+             inputs=input, 
+             outputs="text",
+             title="My Dog PetCare Bot",
+             description="This is a RAG implementation based on Zephyr 7B Beta LLM.",
+             examples=sample_prompts,
+             allow_screenshot=False,
+             allow_flagging=False
+             )
+
+iface.launch()