Spaces:
Sleeping
Sleeping
File size: 4,095 Bytes
09ee545 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 |
# install packages
!pip install langchain openai chromadb tiktoken pypdf panel
# import packages
import os
from langchain.chains import RetrievalQA
from langchain.llms import OpenAI
from langchain.document_loaders import TextLoader
from langchain.document_loaders import PyPDFLoader
from langchain.indexes import VectorstoreIndexCreator
from langchain.text_splitter import CharacterTextSplitter
from langchain.embeddings import OpenAIEmbeddings
from langchain.vectorstores import Chroma
import panel as pn
import tempfile
# sets Panel Framework
pn.extension('texteditor', template="bootstap", sizing_mode='stretch_width')
pn.state.template.param.update(
main_max_width="690px",
header_background="#F08080",
)
# set widgets
file_input = pn.widgets.FileInput(width=300)
openaikey = pn.widgets.PasswordInput(
value="", placeholder="Enter your OpenAI API key here...", width=300
)
prompt = pn.widgets.TextEditor(
value="", placeholder="Enter your questions here...", height=160, toolbar=False
)
run_button = pn.widgets.Button(name="Run!")
select_k = pn.widgets.IntSlider(
name="Number of relevant chunks", start=1, end=5, step=1, value=2
)
select_chain_type = pn.widgets.RadioButtonGroup(
name='Chain type',
options =['stuff', 'map_reduce', "refine", "map_rerank"]
)
widgets = pn.Row(
pn.Column(prompt, run_button, margin=5),
pn.Card(
"Chain type:",
pn.Column(select_chain_type, select_k),
title="Advanced settings", margin=10
), width=600
)
# define the question answering function
def qa(file, query, chain_type, k):
# load document
loader = PyPDFLoader(file)
documents = loader.load()
# split the documents into chunks
text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)
texts = text_splitter.split_documents(documents)
# select which embeddings we want to use
embeddings = OpenAIEmbeddings()
# create the vectorstore to use as the index
db = Chroma.from_documents(texts, embeddings)
# expose this index in a retriever interface
retriever = db.as_retriever(search_type="similarity", search_kwargs={"k": k})
# to create a chain to answer questions
qa = RetreivalQA.from_chain_type(
llm = OPenAI(), chain_type=chain_type, retriever=retriever, return_source_documents=True)
result = qa({"query": query})
print(result['result'])
return result
# store all Panel objects in a list
convos = []
def qa_result(_):
os.environ["OPENAI_API_KEY"] = openaikey.value
# save pdf file to a temp file
if file_input.value is not None:
file_input.save(temp.pdf)
prompt_text = prompt.value
if prompt_text:
result = qa(file="temp.pdf", query=prompt_text, chain_type=select_chain_type.value, k=select_k.value)
convos.extend([
pn.Row(
pn.panel("\U0001F60A", width=10),
prompt_text,
width=600
),
pn.Row(
pn.panel(\"U0001F916", width=10),
pn.Column(
result["result"],
"Relevant source text:",
pn.pane.Markdown('\n--------------------------------------------------------------------\n'.join(doc.page_content for docu in result["source_documents"]))
)
)
])
return pn.Column(*convos, margin=15, width=575, min_height=400)
# bind run button with the qa_result function
qa_interactive = pn.panel(
pn.bind(qa_result, run_button),
loading_indicator=True,
)
output pn.WidgetBox('*Output will show up here:*', qa_interactive, width=630, scroll=True)
# define the layout
pn.Column(
pn.pane.Markdown("""
## \U0001F60A! Question Answering with your PDF file
1) Upload a PDF. 2) Enter OpenAI API key. This costs $. Set up billing at [OpenAI](https://platform.openai.com/account). 3) Type a question and click "Run".
"""),
pn.Row(file_input, openaikey)
output,
widgets
).servable() |