muellerzr HF staff commited on
Commit
4f90d22
1 Parent(s): ebe49f6

Process documentation files

Browse files
process_documentation.ipynb ADDED
@@ -0,0 +1,198 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "code",
5
+ "execution_count": null,
6
+ "metadata": {
7
+ "id": "lIYdn1woOS1n"
8
+ },
9
+ "outputs": [],
10
+ "source": [
11
+ "!git clone https://huggingface.co/datasets/muellerzr/RAG-accelerate"
12
+ ]
13
+ },
14
+ {
15
+ "cell_type": "code",
16
+ "source": [
17
+ "%cd RAG-accelerate"
18
+ ],
19
+ "metadata": {
20
+ "id": "N_0YZlykKdVS"
21
+ },
22
+ "execution_count": null,
23
+ "outputs": []
24
+ },
25
+ {
26
+ "cell_type": "code",
27
+ "source": [
28
+ "!pip install langchain"
29
+ ],
30
+ "metadata": {
31
+ "id": "AK8XGa24Kekk"
32
+ },
33
+ "execution_count": null,
34
+ "outputs": []
35
+ },
36
+ {
37
+ "cell_type": "code",
38
+ "source": [
39
+ "from pathlib import Path"
40
+ ],
41
+ "metadata": {
42
+ "id": "Qt4Zu6mfLvW2"
43
+ },
44
+ "execution_count": 19,
45
+ "outputs": []
46
+ },
47
+ {
48
+ "cell_type": "code",
49
+ "source": [
50
+ "from langchain.text_splitter import MarkdownHeaderTextSplitter, RecursiveCharacterTextSplitter"
51
+ ],
52
+ "metadata": {
53
+ "id": "a9AHp1eZKgUw"
54
+ },
55
+ "execution_count": 12,
56
+ "outputs": []
57
+ },
58
+ {
59
+ "cell_type": "code",
60
+ "source": [
61
+ "docs_folder = Path(\"docs/source\")"
62
+ ],
63
+ "metadata": {
64
+ "id": "1gxt3ZcTLxAb"
65
+ },
66
+ "execution_count": 20,
67
+ "outputs": []
68
+ },
69
+ {
70
+ "cell_type": "code",
71
+ "source": [
72
+ "headers_to_split_on = [\n",
73
+ " (\"#\", \"Header 1\"),\n",
74
+ " (\"##\", \"Header 2\"),\n",
75
+ " (\"###\", \"Header 3\"),\n",
76
+ " (\"####\", \"Header 4\")\n",
77
+ "]"
78
+ ],
79
+ "metadata": {
80
+ "id": "EFwAEg2mKsNL"
81
+ },
82
+ "execution_count": 29,
83
+ "outputs": []
84
+ },
85
+ {
86
+ "cell_type": "code",
87
+ "source": [
88
+ "markdown_splitter = MarkdownHeaderTextSplitter(\n",
89
+ " headers_to_split_on=headers_to_split_on,\n",
90
+ " strip_headers=False\n",
91
+ ")"
92
+ ],
93
+ "metadata": {
94
+ "id": "HpNxkjoZKs2v"
95
+ },
96
+ "execution_count": 30,
97
+ "outputs": []
98
+ },
99
+ {
100
+ "cell_type": "code",
101
+ "source": [
102
+ "chunk_size = 1024\n",
103
+ "chunk_overlap = 256\n",
104
+ "text_splitter = RecursiveCharacterTextSplitter(\n",
105
+ " chunk_size=chunk_size, chunk_overlap=chunk_overlap\n",
106
+ ")"
107
+ ],
108
+ "metadata": {
109
+ "id": "ch_UDrYIM4GT"
110
+ },
111
+ "execution_count": 33,
112
+ "outputs": []
113
+ },
114
+ {
115
+ "cell_type": "code",
116
+ "source": [
117
+ "categories = [\"package_reference\", \"basic_tutorials\", \"concept_guides\", \"usage_guides\"]"
118
+ ],
119
+ "metadata": {
120
+ "id": "kKRNe4qPMM_6"
121
+ },
122
+ "execution_count": 34,
123
+ "outputs": []
124
+ },
125
+ {
126
+ "cell_type": "code",
127
+ "source": [
128
+ "data = []"
129
+ ],
130
+ "metadata": {
131
+ "id": "D-rajLfOMynB"
132
+ },
133
+ "execution_count": 44,
134
+ "outputs": []
135
+ },
136
+ {
137
+ "cell_type": "code",
138
+ "source": [
139
+ "for path in docs_folder.glob(\"**/*.md\"):\n",
140
+ " content = path.read_text()\n",
141
+ " md_header_splits = markdown_splitter.split_text(content)\n",
142
+ " for i,split in enumerate(md_header_splits):\n",
143
+ " if path.parent.name in categories:\n",
144
+ " md_header_splits[i].metadata.update({\"category\":path.parent.name})\n",
145
+ " splits = text_splitter.split_documents(md_header_splits)\n",
146
+ " data += splits"
147
+ ],
148
+ "metadata": {
149
+ "id": "zAqjYMdMMJRH"
150
+ },
151
+ "execution_count": 46,
152
+ "outputs": []
153
+ },
154
+ {
155
+ "cell_type": "code",
156
+ "source": [
157
+ "import numpy as np"
158
+ ],
159
+ "metadata": {
160
+ "id": "OAaI9OV_NsTC"
161
+ },
162
+ "execution_count": 50,
163
+ "outputs": []
164
+ },
165
+ {
166
+ "cell_type": "code",
167
+ "source": [
168
+ "np.save(\"processed_documentation_chunks.npy\", data)"
169
+ ],
170
+ "metadata": {
171
+ "id": "H9FzuYdvOGyM"
172
+ },
173
+ "execution_count": 51,
174
+ "outputs": []
175
+ },
176
+ {
177
+ "cell_type": "code",
178
+ "source": [],
179
+ "metadata": {
180
+ "id": "BgSCAljkOKLG"
181
+ },
182
+ "execution_count": null,
183
+ "outputs": []
184
+ }
185
+ ],
186
+ "metadata": {
187
+ "colab": {
188
+ "name": "scratchpad",
189
+ "provenance": []
190
+ },
191
+ "kernelspec": {
192
+ "display_name": "Python 3",
193
+ "name": "python3"
194
+ }
195
+ },
196
+ "nbformat": 4,
197
+ "nbformat_minor": 0
198
+ }
processed_documentation_chunks.npy ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7bde0bceeea343ca0ff9317da31769854f0382fc4fe2a5eb56c269cabadf3ee7
3
+ size 390777