PanoEvJ commited on
Commit
927867b
β€’
1 Parent(s): 0605939
Files changed (6) hide show
  1. .gitattributes +1 -0
  2. README.md +19 -13
  3. app.py +68 -0
  4. chainlit.md +14 -0
  5. hitchhikers.pdf +3 -0
  6. requirements.txt +3 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ *.pdf filter=lfs diff=lfs merge=lfs -text
README.md CHANGED
@@ -1,13 +1,19 @@
1
- ---
2
- title: Ask GalacticHitchHikers With FineTunedGPT Using LlamaIndex
3
- emoji: πŸ‘€
4
- colorFrom: red
5
- colorTo: red
6
- sdk: streamlit
7
- sdk_version: 1.26.0
8
- app_file: app.py
9
- pinned: false
10
- license: bigscience-openrail-m
11
- ---
12
-
13
- Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
 
 
 
 
 
 
 
1
+ # Building A Finetuning Machine Using Chainlit
2
+
3
+ Today we'll take a look at how we can wrap our finetuning process into Chainlit and have our own dynamic fine-tuning machine.
4
+
5
+ We'll be leveraging [this](https://github.com/AI-Maker-Space/LLM-Ops-Cohort-1/blob/main/Week%202/Thursday/Automated%20Fine-tuning%20with%20LLamaIndex.ipynb) notebook for finetuning.
6
+
7
+ Find worked-out notebook at this [repo](https://github.com/PanoEvJ/LLMOps/tree/main/Week%202/Thursday).
8
+
9
+ # Build πŸ—οΈ
10
+ There are 2 main tasks for this assignment:
11
+
12
+ - Verify you can run the notebook
13
+ - Wrap the notebook using Chainlit
14
+
15
+ # Ship 🚒
16
+ Construct a Chainlit application using the notebook that allows users to interface with the application.
17
+
18
+ # Share πŸš€
19
+ Make a social media post about your final application and tag @AIMakerspace
app.py ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import os
2
+ import openai
3
+
4
+ from llama_index.query_engine.retriever_query_engine import RetrieverQueryEngine
5
+ from llama_index.callbacks.base import CallbackManager
6
+ from llama_index import (
7
+ LLMPredictor,
8
+ ServiceContext,
9
+ StorageContext,
10
+ load_index_from_storage,
11
+ )
12
+ from llama_index.llms import OpenAI
13
+ import chainlit as cl
14
+
15
+
16
+ openai.api_key = os.environ.get("OPENAI_API_KEY")
17
+
18
+ try:
19
+ # rebuild storage context
20
+ storage_context = StorageContext.from_defaults(persist_dir="./storage")
21
+ # load index
22
+ index = load_index_from_storage(storage_context)
23
+ except:
24
+ from llama_index import GPTVectorStoreIndex, SimpleDirectoryReader
25
+
26
+ documents = SimpleDirectoryReader(input_files=["hitchhikers.pdf"]).load_data()
27
+ index = GPTVectorStoreIndex.from_documents(documents)
28
+ index.storage_context.persist()
29
+
30
+
31
+ @cl.on_chat_start
32
+ async def factory():
33
+ llm_predictor = LLMPredictor(
34
+ llm=OpenAI(
35
+ temperature=0,
36
+ model="ft:gpt-3.5-turbo-0613:personal::7rG4voK4",
37
+ streaming=True,
38
+ context_window=2048,
39
+ ),
40
+ )
41
+ service_context = ServiceContext.from_defaults(
42
+ llm_predictor=llm_predictor,
43
+ chunk_size=512,
44
+ callback_manager=CallbackManager([cl.LlamaIndexCallbackHandler()]),
45
+ )
46
+
47
+ query_engine = index.as_query_engine(
48
+ service_context=service_context,
49
+ streaming=True,
50
+ )
51
+
52
+ cl.user_session.set("query_engine", query_engine)
53
+
54
+
55
+ @cl.on_message
56
+ async def main(message):
57
+ query_engine = cl.user_session.get("query_engine") # type: RetrieverQueryEngine
58
+ response = await cl.make_async(query_engine.query)(message)
59
+
60
+ response_message = cl.Message(content="")
61
+
62
+ for token in response.response_gen:
63
+ await response_message.stream_token(token=token)
64
+
65
+ if response.response_txt:
66
+ response_message.content = response.response_txt
67
+
68
+ await response_message.send()
chainlit.md ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Welcome to Chainlit! πŸš€πŸ€–
2
+
3
+ Hi there, Developer! πŸ‘‹ We're excited to have you on board. Chainlit is a powerful tool designed to help you prototype, debug and share applications built on top of LLMs.
4
+
5
+ ## Useful Links πŸ”—
6
+
7
+ - **Documentation:** Get started with our comprehensive [Chainlit Documentation](https://docs.chainlit.io) πŸ“š
8
+ - **Discord Community:** Join our friendly [Chainlit Discord](https://discord.gg/ZThrUxbAYw) to ask questions, share your projects, and connect with other developers! πŸ’¬
9
+
10
+ We can't wait to see what you create with Chainlit! Happy coding! πŸ’»πŸ˜Š
11
+
12
+ ## Welcome screen
13
+
14
+ To modify the welcome screen, edit the `chainlit.md` file at the root of your project. If you do not want a welcome screen, just leave this file empty.
hitchhikers.pdf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fbf013abdefb5a9f2bccd3f7345467712213d4f247ce29846045056e8cf603c1
3
+ size 3402125
requirements.txt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ chainlit==0.6.3
2
+ llama_index==0.8.9
3
+ openai==0.27.9