Spaces:
Sleeping
Sleeping
File size: 7,695 Bytes
b8bc9e3 69283f8 b8bc9e3 c413f56 b8bc9e3 69283f8 b8bc9e3 69283f8 b8bc9e3 7da8f6b b8bc9e3 11d734e a5af360 11d734e a402973 b8bc9e3 69283f8 b8bc9e3 69283f8 b8bc9e3 a5af360 b8bc9e3 69283f8 b8bc9e3 c413f56 b8bc9e3 3ef7248 b8bc9e3 a081905 065db74 b8bc9e3 69283f8 b8bc9e3 69283f8 b8bc9e3 69283f8 b8bc9e3 69283f8 b8bc9e3 c256eaf b8bc9e3 69283f8 b8bc9e3 69283f8 b8bc9e3 69283f8 b8bc9e3 d551216 b8bc9e3 69283f8 c413f56 d551216 b8bc9e3 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 |
{
"cells": [
{
"cell_type": "code",
"execution_count": null,
"id": "543ba231",
"metadata": {},
"outputs": [],
"source": [
"#%pip install langchain openai chromadb tiktoken pypdf panel bokeh"
]
},
{
"cell_type": "code",
"execution_count": 21,
"id": "4545d020",
"metadata": {},
"outputs": [],
"source": [
"from langchain.chains import RetrievalQA\n",
"from langchain.llms import OpenAI\n",
"from langchain.document_loaders import PyPDFLoader\n",
"from langchain.indexes import VectorstoreIndexCreator\n",
"from langchain.text_splitter import CharacterTextSplitter\n",
"from langchain.embeddings import OpenAIEmbeddings\n",
"from langchain.vectorstores import Chroma\n",
"import panel as pn\n",
"import os\n",
"import dotenv\n",
"dotenv.load_dotenv()\n",
"api_key = os.environ.get(\"API_KEY\")"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "2a643296",
"metadata": {},
"outputs": [],
"source": [
"\n",
"pn.extension('texteditor', template=\"bootstrap\", sizing_mode='stretch_width')\n",
"pn.state.template.param.update(\n",
" main_max_width=\"690px\",\n",
" header_background=\"#F08080\",\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 23,
"id": "f843fe71",
"metadata": {},
"outputs": [],
"source": [
"file_input = pn.widgets.FileInput(width=300)\n",
"\n",
"openaikey = pn.widgets.PasswordInput(\n",
" value=api_key, placeholder=\"Insert your Openai API key here\", width=250\n",
")\n",
"prompt = pn.widgets.TextEditor(\n",
" value=\"\", placeholder=\"How do you become a good programmer?\", height=160, toolbar=False\n",
")\n",
"search_button = pn.widgets.Button(name=\"Search!\")\n",
"\n",
"select_k = pn.widgets.IntSlider(\n",
" name=\"Number of relevant chunks\", start=1, end=10, step=1, value=2\n",
")\n",
"select_chain_type = pn.widgets.RadioButtonGroup(\n",
" name='Chain type', \n",
" options=['stuff', 'map_reduce', \"refine\"]\n",
")\n",
"\n",
"legend = pn.pane.Markdown(\"\"\"\n",
"**Chain types:**\n",
"\n",
"- `stuff`: Parse the entire document at one go. (for smaller documents)\n",
"- `map_reduce`: Split the document into smaller chunks in parallel. (faster processing but less detailed answer)\n",
"- `refine`: Split the document into smaller chunks subsequently. (slower processing but more detailed answer)\n",
"\"\"\")\n",
"\n",
"widgets = pn.Row(\n",
" pn.Column(prompt, search_button, margin=5),\n",
" pn.Card(\n",
" \"Chain type:\",\n",
" pn.Column(select_chain_type, select_k),\n",
" title=\"Advanced settings\", margin=20,\n",
" style={\"width\": \"3000px\"}\n",
" ), \n",
" pn.Column(legend)\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 24,
"id": "7e7daca0",
"metadata": {},
"outputs": [],
"source": [
"def qa(file, query, chain_type, k):\n",
" # load document pdf\n",
" loader = PyPDFLoader(file)\n",
" documents = loader.load()\n",
" # split the documents into chunks\n",
" text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)\n",
" texts = text_splitter.split_documents(documents)\n",
" # select which embeddings we want to use\n",
" embeddings = OpenAIEmbeddings()\n",
" # create the vectorestore to use as the index\n",
" db = Chroma.from_documents(texts, embeddings)\n",
" # expose this index in a retriever interface\n",
" retriever = db.as_retriever(search_type=\"similarity\", search_kwargs={\"k\": k})\n",
" # create a chain to answer questions \n",
" qa = RetrievalQA.from_chain_type(\n",
" llm=OpenAI(), chain_type=chain_type, retriever=retriever, return_source_documents=True)\n",
" result = qa({\"query\": query})\n",
" print(result['result'])\n",
" return result"
]
},
{
"cell_type": "code",
"execution_count": 25,
"id": "1ddd25b6",
"metadata": {},
"outputs": [],
"source": [
"convos = [] # store all panel objects in a list\n",
"\n",
"def qa_result(_):\n",
" os.environ[\"OPENAI_API_KEY\"] = openaikey.value\n",
" \n",
" # save pdf file to a temp file \n",
" if file_input.value is not None:\n",
" file_input.save(\"/.cache/temp.pdf\")\n",
" \n",
" prompt_text = prompt.value\n",
" if prompt_text:\n",
" result = qa(file=\"/.cache/temp.pdf\", query=prompt_text, chain_type=select_chain_type.value, k=select_k.value)\n",
" convos.extend([\n",
" pn.Row(\n",
" pn.panel(\"\\U0001F60A\", width=10),\n",
" prompt_text,\n",
" width=600\n",
" ),\n",
" pn.Row(\n",
" pn.panel(\"\\U0001F916\", width=10),\n",
" pn.Column(\n",
" result[\"result\"],\n",
" \"Relevant source text:\",\n",
" pn.pane.Markdown('\\n--------------------------------------------------------------------\\n'.join(doc.page_content for doc in result[\"source_documents\"]))\n",
" )\n",
" )\n",
" ])\n",
" #return convos\n",
" return pn.Column(*convos, margin=15, width=575, min_height=400)"
]
},
{
"cell_type": "code",
"execution_count": 26,
"id": "7341d7df",
"metadata": {},
"outputs": [],
"source": [
"qa_interactive = pn.panel(\n",
" pn.bind(qa_result, search_button),\n",
" loading_indicator=True,\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 27,
"id": "88d43948",
"metadata": {},
"outputs": [],
"source": [
"output = pn.WidgetBox('*Output will show up here:*', qa_interactive, width=630, scroll=True)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "19041fc0",
"metadata": {},
"outputs": [],
"source": [
"# layout\n",
"pn.Column(\n",
" pn.pane.Markdown(\"\"\"\n",
" ## Query anything in your files \\U0001F60A!\n",
" \n",
" 1) Upload a pdf file. \\n\n",
" 2) Enter OpenAI API key(Using Aaron's by default). \\n\n",
" 3) Ask any question and click \"Search\". (Click twice on first startup)\n",
" \n",
" \"\"\"),\n",
" pn.Row(file_input,pn.pane.HTML(\"OpenAI API Key: \", width=110, align=\"end\"), openaikey),\n",
" output,\n",
" widgets\n",
"\n",
").servable()"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "3a9b908d",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "100ec49c",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "e533a218",
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.8"
}
},
"nbformat": 4,
"nbformat_minor": 5
}
|