Spaces:
Build error
Build error
XThomasBU
commited on
Commit
•
57b7b8d
1
Parent(s):
fe158b7
modularied dataloader + Added Chroma
Browse files- code/config.yml +2 -2
- code/modules/data_loader.py +165 -223
- code/modules/embedding_model_loader.py +2 -0
- code/modules/helpers.py +11 -7
- code/modules/llm_tutor.py +15 -13
- code/modules/vector_db.py +75 -22
code/config.yml
CHANGED
@@ -1,6 +1,5 @@
|
|
1 |
embedding_options:
|
2 |
embedd_files: False # bool
|
3 |
-
persist_directory: null # str or None
|
4 |
data_path: 'storage/data' # str
|
5 |
url_file_path: 'storage/data/urls.txt' # str
|
6 |
expand_urls: True # bool
|
@@ -8,8 +7,9 @@ embedding_options:
|
|
8 |
db_path : 'vectorstores' # str
|
9 |
model : 'sentence-transformers/all-MiniLM-L6-v2' # str [sentence-transformers/all-MiniLM-L6-v2, text-embedding-ada-002']
|
10 |
search_top_k : 3 # int
|
|
|
11 |
llm_params:
|
12 |
-
use_history:
|
13 |
memory_window: 3 # int
|
14 |
llm_loader: 'local_llm' # str [local_llm, openai]
|
15 |
openai_params:
|
|
|
1 |
embedding_options:
|
2 |
embedd_files: False # bool
|
|
|
3 |
data_path: 'storage/data' # str
|
4 |
url_file_path: 'storage/data/urls.txt' # str
|
5 |
expand_urls: True # bool
|
|
|
7 |
db_path : 'vectorstores' # str
|
8 |
model : 'sentence-transformers/all-MiniLM-L6-v2' # str [sentence-transformers/all-MiniLM-L6-v2, text-embedding-ada-002']
|
9 |
search_top_k : 3 # int
|
10 |
+
score_threshold : 0.5 # float
|
11 |
llm_params:
|
12 |
+
use_history: True # bool
|
13 |
memory_window: 3 # int
|
14 |
llm_loader: 'local_llm' # str [local_llm, openai]
|
15 |
openai_params:
|
code/modules/data_loader.py
CHANGED
@@ -1,6 +1,7 @@
|
|
|
|
1 |
import re
|
|
|
2 |
import pysrt
|
3 |
-
from langchain.text_splitter import RecursiveCharacterTextSplitter
|
4 |
from langchain.document_loaders import (
|
5 |
PyMuPDFLoader,
|
6 |
Docx2txtLoader,
|
@@ -8,49 +9,32 @@ from langchain.document_loaders import (
|
|
8 |
WebBaseLoader,
|
9 |
TextLoader,
|
10 |
)
|
|
|
|
|
11 |
from langchain.schema import Document
|
12 |
-
import tempfile
|
13 |
-
from tempfile import NamedTemporaryFile
|
14 |
import logging
|
15 |
-
import
|
|
|
|
|
16 |
|
17 |
logger = logging.getLogger(__name__)
|
18 |
|
19 |
|
20 |
-
class
|
21 |
-
def __init__(self
|
22 |
-
|
23 |
-
Class for handling all data extraction and chunking
|
24 |
-
Inputs:
|
25 |
-
config - dictionary from yaml file, containing all important parameters
|
26 |
-
"""
|
27 |
-
self.config = config
|
28 |
-
self.remove_leftover_delimiters = config["splitter_options"][
|
29 |
-
"remove_leftover_delimiters"
|
30 |
-
]
|
31 |
|
32 |
-
|
33 |
-
|
34 |
-
|
35 |
|
36 |
-
|
37 |
-
|
38 |
-
|
39 |
-
|
40 |
-
|
41 |
-
|
42 |
-
|
43 |
-
)
|
44 |
-
else:
|
45 |
-
self.splitter = RecursiveCharacterTextSplitter(
|
46 |
-
chunk_size=config["splitter_options"]["chunk_size"],
|
47 |
-
chunk_overlap=config["splitter_options"]["chunk_overlap"],
|
48 |
-
separators=config["splitter_options"]["chunk_separators"],
|
49 |
-
disallowed_special=()
|
50 |
-
)
|
51 |
-
else:
|
52 |
-
self.splitter = None
|
53 |
-
logger.info("InfoLoader instance created")
|
54 |
|
55 |
def extract_text_from_pdf(self, pdf_path):
|
56 |
text = ""
|
@@ -73,215 +57,173 @@ class DataLoader:
|
|
73 |
print("Failed to download PDF from URL:", pdf_url)
|
74 |
return None
|
75 |
|
76 |
-
def
|
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 |
-
def get_pdf(temp_file_path: str, title: str):
|
114 |
-
"""
|
115 |
-
Function to process PDF files
|
116 |
-
"""
|
117 |
-
loader = PyMuPDFLoader(
|
118 |
-
temp_file_path
|
119 |
-
) # This loader preserves more metadata
|
120 |
-
|
121 |
-
if self.splitter:
|
122 |
-
document_chunks = self.splitter.split_documents(loader.load())
|
123 |
-
else:
|
124 |
-
document_chunks = loader.load()
|
125 |
-
|
126 |
-
if "title" in document_chunks[0].metadata.keys():
|
127 |
-
title = document_chunks[0].metadata["title"]
|
128 |
-
|
129 |
-
logger.info(
|
130 |
-
f"\t\tOriginal no. of pages: {document_chunks[0].metadata['total_pages']}"
|
131 |
-
)
|
132 |
-
|
133 |
-
return title, document_chunks
|
134 |
-
|
135 |
-
def get_txt(temp_file_path: str, title: str):
|
136 |
-
"""
|
137 |
-
Function to process TXT files
|
138 |
-
"""
|
139 |
-
loader = TextLoader(temp_file_path, autodetect_encoding=True)
|
140 |
-
|
141 |
-
if self.splitter:
|
142 |
-
document_chunks = self.splitter.split_documents(loader.load())
|
143 |
-
else:
|
144 |
-
document_chunks = loader.load()
|
145 |
-
|
146 |
-
# Update the metadata
|
147 |
-
for chunk in document_chunks:
|
148 |
-
chunk.metadata["source"] = title
|
149 |
-
chunk.metadata["page"] = "N/A"
|
150 |
-
|
151 |
-
return title, document_chunks
|
152 |
-
|
153 |
-
def get_srt(temp_file_path: str, title: str):
|
154 |
-
"""
|
155 |
-
Function to process SRT files
|
156 |
-
"""
|
157 |
-
subs = pysrt.open(temp_file_path)
|
158 |
-
|
159 |
-
text = ""
|
160 |
-
for sub in subs:
|
161 |
-
text += sub.text
|
162 |
-
document_chunks = [Document(page_content=text)]
|
163 |
-
|
164 |
-
if self.splitter:
|
165 |
-
document_chunks = self.splitter.split_documents(document_chunks)
|
166 |
-
|
167 |
-
# Update the metadata
|
168 |
-
for chunk in document_chunks:
|
169 |
-
chunk.metadata["source"] = title
|
170 |
-
chunk.metadata["page"] = "N/A"
|
171 |
-
|
172 |
-
return title, document_chunks
|
173 |
-
|
174 |
-
def get_docx(temp_file_path: str, title: str):
|
175 |
-
"""
|
176 |
-
Function to process DOCX files
|
177 |
-
"""
|
178 |
-
loader = Docx2txtLoader(temp_file_path)
|
179 |
|
180 |
-
|
181 |
-
|
182 |
-
|
183 |
-
|
|
|
184 |
|
185 |
-
|
186 |
-
|
187 |
-
|
188 |
-
chunk.metadata["page"] = "N/A"
|
189 |
|
190 |
-
return title, document_chunks
|
191 |
|
192 |
-
|
193 |
-
|
194 |
-
|
195 |
-
|
196 |
-
|
197 |
-
|
198 |
-
|
|
|
199 |
|
200 |
-
|
201 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
202 |
else:
|
203 |
-
|
204 |
-
|
205 |
-
|
206 |
-
|
207 |
-
|
208 |
-
|
209 |
-
|
210 |
-
|
211 |
-
|
212 |
-
|
213 |
-
|
214 |
-
|
215 |
-
"""
|
216 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
217 |
|
218 |
-
|
219 |
-
|
220 |
-
|
221 |
-
|
222 |
|
223 |
-
|
224 |
-
logger.info(document_chunks[0].metadata)
|
225 |
|
226 |
-
|
|
|
|
|
227 |
|
228 |
-
# Handle file by file
|
229 |
for file_index, file_path in enumerate(uploaded_files):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
230 |
|
231 |
-
|
232 |
-
|
|
|
|
|
233 |
|
234 |
-
|
235 |
-
if file_type == "pdf":
|
236 |
-
try:
|
237 |
-
title, document_chunks = get_pdf(file_path, file_name)
|
238 |
-
except:
|
239 |
-
title, document_chunks = get_pdf_from_url(file_path)
|
240 |
-
elif file_type == "txt":
|
241 |
-
title, document_chunks = get_txt(file_path, file_name)
|
242 |
-
elif file_type == "docx":
|
243 |
-
title, document_chunks = get_docx(file_path, file_name)
|
244 |
-
elif file_type == "srt":
|
245 |
-
title, document_chunks = get_srt(file_path, file_name)
|
246 |
-
|
247 |
-
# Additional wrangling - Remove leftover delimiters and any specified chunks
|
248 |
-
if self.remove_leftover_delimiters:
|
249 |
-
document_chunks = remove_delimiters(document_chunks)
|
250 |
-
if self.config["splitter_options"]["remove_chunks"]:
|
251 |
-
document_chunks = remove_chunks(document_chunks)
|
252 |
-
|
253 |
-
logger.info(f"\t\tExtracted no. of chunks: {len(document_chunks)} from {file_name}")
|
254 |
-
self.document_names.append(title)
|
255 |
-
self.document_chunks_full.extend(document_chunks)
|
256 |
-
|
257 |
-
# Handle youtube links:
|
258 |
if weblinks[0] != "":
|
259 |
logger.info(f"Splitting weblinks: total of {len(weblinks)}")
|
260 |
|
261 |
-
# Handle link by link
|
262 |
for link_index, link in enumerate(weblinks):
|
263 |
try:
|
264 |
logger.info(f"\tSplitting link {link_index+1} : {link}")
|
265 |
if "youtube" in link:
|
266 |
-
|
267 |
else:
|
268 |
-
|
269 |
-
|
270 |
-
# Additional wrangling - Remove leftover delimiters and any specified chunks
|
271 |
-
if self.remove_leftover_delimiters:
|
272 |
-
document_chunks = remove_delimiters(document_chunks)
|
273 |
-
if self.config["splitter_options"]["remove_chunks"]:
|
274 |
-
document_chunks = remove_chunks(document_chunks)
|
275 |
|
276 |
-
|
277 |
-
self.document_names.append(
|
278 |
self.document_chunks_full.extend(document_chunks)
|
279 |
-
except:
|
280 |
-
logger.
|
281 |
-
|
|
|
282 |
|
283 |
-
logger.info(
|
284 |
-
f"\tNumber of document chunks extracted in total: {len(self.document_chunks_full)}\n\n"
|
285 |
-
)
|
286 |
|
287 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import os
|
2 |
import re
|
3 |
+
import requests
|
4 |
import pysrt
|
|
|
5 |
from langchain.document_loaders import (
|
6 |
PyMuPDFLoader,
|
7 |
Docx2txtLoader,
|
|
|
9 |
WebBaseLoader,
|
10 |
TextLoader,
|
11 |
)
|
12 |
+
from langchain_community.document_loaders import UnstructuredMarkdownLoader
|
13 |
+
from llama_parse import LlamaParse
|
14 |
from langchain.schema import Document
|
|
|
|
|
15 |
import logging
|
16 |
+
from langchain.text_splitter import RecursiveCharacterTextSplitter
|
17 |
+
from langchain_experimental.text_splitter import SemanticChunker
|
18 |
+
from langchain_openai.embeddings import OpenAIEmbeddings
|
19 |
|
20 |
logger = logging.getLogger(__name__)
|
21 |
|
22 |
|
23 |
+
class PDFReader:
|
24 |
+
def __init__(self):
|
25 |
+
pass
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
26 |
|
27 |
+
def get_loader(self, pdf_path):
|
28 |
+
loader = PyMuPDFLoader(pdf_path)
|
29 |
+
return loader
|
30 |
|
31 |
+
def get_documents(self, loader):
|
32 |
+
return loader.load()
|
33 |
+
|
34 |
+
|
35 |
+
class FileReader:
|
36 |
+
def __init__(self):
|
37 |
+
self.pdf_reader = PDFReader()
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
38 |
|
39 |
def extract_text_from_pdf(self, pdf_path):
|
40 |
text = ""
|
|
|
57 |
print("Failed to download PDF from URL:", pdf_url)
|
58 |
return None
|
59 |
|
60 |
+
def read_pdf(self, temp_file_path: str):
|
61 |
+
# parser = LlamaParse(
|
62 |
+
# api_key="",
|
63 |
+
# result_type="markdown",
|
64 |
+
# num_workers=4,
|
65 |
+
# verbose=True,
|
66 |
+
# language="en",
|
67 |
+
# )
|
68 |
+
# documents = parser.load_data(temp_file_path)
|
69 |
+
|
70 |
+
# with open("temp/output.md", "a") as f:
|
71 |
+
# for doc in documents:
|
72 |
+
# f.write(doc.text + "\n")
|
73 |
+
|
74 |
+
# markdown_path = "temp/output.md"
|
75 |
+
# loader = UnstructuredMarkdownLoader(markdown_path)
|
76 |
+
# loader = PyMuPDFLoader(temp_file_path) # This loader preserves more metadata
|
77 |
+
# return loader.load()
|
78 |
+
loader = self.pdf_reader.get_loader(temp_file_path)
|
79 |
+
documents = self.pdf_reader.get_documents(loader)
|
80 |
+
return documents
|
81 |
+
|
82 |
+
def read_txt(self, temp_file_path: str):
|
83 |
+
loader = TextLoader(temp_file_path, autodetect_encoding=True)
|
84 |
+
return loader.load()
|
85 |
+
|
86 |
+
def read_docx(self, temp_file_path: str):
|
87 |
+
loader = Docx2txtLoader(temp_file_path)
|
88 |
+
return loader.load()
|
89 |
+
|
90 |
+
def read_srt(self, temp_file_path: str):
|
91 |
+
subs = pysrt.open(temp_file_path)
|
92 |
+
text = ""
|
93 |
+
for sub in subs:
|
94 |
+
text += sub.text
|
95 |
+
return [Document(page_content=text)]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
96 |
|
97 |
+
def read_youtube_transcript(self, url: str):
|
98 |
+
loader = YoutubeLoader.from_youtube_url(
|
99 |
+
url, add_video_info=True, language=["en"], translation="en"
|
100 |
+
)
|
101 |
+
return loader.load()
|
102 |
|
103 |
+
def read_html(self, url: str):
|
104 |
+
loader = WebBaseLoader(url)
|
105 |
+
return loader.load()
|
|
|
106 |
|
|
|
107 |
|
108 |
+
class ChunkProcessor:
|
109 |
+
def __init__(self, config):
|
110 |
+
self.config = config
|
111 |
+
self.remove_leftover_delimiters = config["splitter_options"][
|
112 |
+
"remove_leftover_delimiters"
|
113 |
+
]
|
114 |
+
self.document_chunks_full = []
|
115 |
+
self.document_names = []
|
116 |
|
117 |
+
if config["splitter_options"]["use_splitter"]:
|
118 |
+
if config["splitter_options"]["split_by_token"]:
|
119 |
+
self.splitter = RecursiveCharacterTextSplitter.from_tiktoken_encoder(
|
120 |
+
chunk_size=config["splitter_options"]["chunk_size"],
|
121 |
+
chunk_overlap=config["splitter_options"]["chunk_overlap"],
|
122 |
+
separators=config["splitter_options"]["chunk_separators"],
|
123 |
+
disallowed_special=(),
|
124 |
+
)
|
125 |
else:
|
126 |
+
self.splitter = RecursiveCharacterTextSplitter(
|
127 |
+
chunk_size=config["splitter_options"]["chunk_size"],
|
128 |
+
chunk_overlap=config["splitter_options"]["chunk_overlap"],
|
129 |
+
separators=config["splitter_options"]["chunk_separators"],
|
130 |
+
disallowed_special=(),
|
131 |
+
)
|
132 |
+
else:
|
133 |
+
self.splitter = None
|
134 |
+
logger.info("ChunkProcessor instance created")
|
135 |
+
|
136 |
+
def remove_delimiters(self, document_chunks: list):
|
137 |
+
for chunk in document_chunks:
|
138 |
+
for delimiter in self.config["splitter_options"]["delimiters_to_remove"]:
|
139 |
+
chunk.page_content = re.sub(delimiter, " ", chunk.page_content)
|
140 |
+
return document_chunks
|
141 |
+
|
142 |
+
def remove_chunks(self, document_chunks: list):
|
143 |
+
front = self.config["splitter_options"]["front_chunk_to_remove"]
|
144 |
+
end = self.config["splitter_options"]["last_chunks_to_remove"]
|
145 |
+
for _ in range(front):
|
146 |
+
del document_chunks[0]
|
147 |
+
for _ in range(end):
|
148 |
+
document_chunks.pop()
|
149 |
+
logger.info(f"\tNumber of pages after skipping: {len(document_chunks)}")
|
150 |
+
return document_chunks
|
151 |
+
|
152 |
+
def process_chunks(self, documents):
|
153 |
+
if self.splitter:
|
154 |
+
document_chunks = self.splitter.split_documents(documents)
|
155 |
+
else:
|
156 |
+
document_chunks = documents
|
157 |
|
158 |
+
if self.remove_leftover_delimiters:
|
159 |
+
document_chunks = self.remove_delimiters(document_chunks)
|
160 |
+
if self.config["splitter_options"]["remove_chunks"]:
|
161 |
+
document_chunks = self.remove_chunks(document_chunks)
|
162 |
|
163 |
+
return document_chunks
|
|
|
164 |
|
165 |
+
def get_chunks(self, file_reader, uploaded_files, weblinks):
|
166 |
+
self.document_chunks_full = []
|
167 |
+
self.document_names = []
|
168 |
|
|
|
169 |
for file_index, file_path in enumerate(uploaded_files):
|
170 |
+
file_name = os.path.basename(file_path)
|
171 |
+
file_type = file_name.split(".")[-1].lower()
|
172 |
+
|
173 |
+
try:
|
174 |
+
if file_type == "pdf":
|
175 |
+
documents = file_reader.read_pdf(file_path)
|
176 |
+
elif file_type == "txt":
|
177 |
+
documents = file_reader.read_txt(file_path)
|
178 |
+
elif file_type == "docx":
|
179 |
+
documents = file_reader.read_docx(file_path)
|
180 |
+
elif file_type == "srt":
|
181 |
+
documents = file_reader.read_srt(file_path)
|
182 |
+
else:
|
183 |
+
logger.warning(f"Unsupported file type: {file_type}")
|
184 |
+
continue
|
185 |
+
|
186 |
+
document_chunks = self.process_chunks(documents)
|
187 |
+
self.document_names.append(file_name)
|
188 |
+
self.document_chunks_full.extend(document_chunks)
|
189 |
+
|
190 |
+
except Exception as e:
|
191 |
+
logger.error(f"Error processing file {file_name}: {str(e)}")
|
192 |
+
|
193 |
+
self.process_weblinks(file_reader, weblinks)
|
194 |
|
195 |
+
logger.info(
|
196 |
+
f"Total document chunks extracted: {len(self.document_chunks_full)}"
|
197 |
+
)
|
198 |
+
return self.document_chunks_full, self.document_names
|
199 |
|
200 |
+
def process_weblinks(self, file_reader, weblinks):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
201 |
if weblinks[0] != "":
|
202 |
logger.info(f"Splitting weblinks: total of {len(weblinks)}")
|
203 |
|
|
|
204 |
for link_index, link in enumerate(weblinks):
|
205 |
try:
|
206 |
logger.info(f"\tSplitting link {link_index+1} : {link}")
|
207 |
if "youtube" in link:
|
208 |
+
documents = file_reader.read_youtube_transcript(link)
|
209 |
else:
|
210 |
+
documents = file_reader.read_html(link)
|
|
|
|
|
|
|
|
|
|
|
|
|
211 |
|
212 |
+
document_chunks = self.process_chunks(documents)
|
213 |
+
self.document_names.append(link)
|
214 |
self.document_chunks_full.extend(document_chunks)
|
215 |
+
except Exception as e:
|
216 |
+
logger.error(
|
217 |
+
f"Error splitting link {link_index+1} : {link}: {str(e)}"
|
218 |
+
)
|
219 |
|
|
|
|
|
|
|
220 |
|
221 |
+
class DataLoader:
|
222 |
+
def __init__(self, config):
|
223 |
+
self.file_reader = FileReader()
|
224 |
+
self.chunk_processor = ChunkProcessor(config)
|
225 |
+
|
226 |
+
def get_chunks(self, uploaded_files, weblinks):
|
227 |
+
return self.chunk_processor.get_chunks(
|
228 |
+
self.file_reader, uploaded_files, weblinks
|
229 |
+
)
|
code/modules/embedding_model_loader.py
CHANGED
@@ -1,6 +1,7 @@
|
|
1 |
from langchain_community.embeddings import OpenAIEmbeddings
|
2 |
from langchain.embeddings import HuggingFaceEmbeddings
|
3 |
from langchain.embeddings import LlamaCppEmbeddings
|
|
|
4 |
try:
|
5 |
from modules.constants import *
|
6 |
except:
|
@@ -19,6 +20,7 @@ class EmbeddingModelLoader:
|
|
19 |
model=self.config["embedding_options"]["model"],
|
20 |
show_progress_bar=True,
|
21 |
openai_api_key=OPENAI_API_KEY,
|
|
|
22 |
)
|
23 |
else:
|
24 |
embedding_model = HuggingFaceEmbeddings(
|
|
|
1 |
from langchain_community.embeddings import OpenAIEmbeddings
|
2 |
from langchain.embeddings import HuggingFaceEmbeddings
|
3 |
from langchain.embeddings import LlamaCppEmbeddings
|
4 |
+
|
5 |
try:
|
6 |
from modules.constants import *
|
7 |
except:
|
|
|
20 |
model=self.config["embedding_options"]["model"],
|
21 |
show_progress_bar=True,
|
22 |
openai_api_key=OPENAI_API_KEY,
|
23 |
+
disallowed_special=(),
|
24 |
)
|
25 |
else:
|
26 |
embedding_model = HuggingFaceEmbeddings(
|
code/modules/helpers.py
CHANGED
@@ -4,6 +4,7 @@ from tqdm import tqdm
|
|
4 |
from urllib.parse import urlparse
|
5 |
import chainlit as cl
|
6 |
from langchain import PromptTemplate
|
|
|
7 |
try:
|
8 |
from modules.constants import *
|
9 |
except:
|
@@ -60,7 +61,7 @@ class WebpageCrawler:
|
|
60 |
|
61 |
def get_subpage_links(self, l, base_url):
|
62 |
for link in tqdm(l):
|
63 |
-
print(
|
64 |
if not link.endswith("/"):
|
65 |
l[link] = "Checked"
|
66 |
dict_links_subpages = {}
|
@@ -109,6 +110,7 @@ def get_base_url(url):
|
|
109 |
base_url = f"{parsed_url.scheme}://{parsed_url.netloc}/"
|
110 |
return base_url
|
111 |
|
|
|
112 |
def get_prompt(config):
|
113 |
if config["llm_params"]["use_history"]:
|
114 |
if config["llm_params"]["llm_loader"] == "local_llm":
|
@@ -134,6 +136,7 @@ def get_prompt(config):
|
|
134 |
)
|
135 |
return prompt
|
136 |
|
|
|
137 |
def get_sources(res, answer):
|
138 |
source_elements_dict = {}
|
139 |
source_elements = []
|
@@ -144,21 +147,22 @@ def get_sources(res, answer):
|
|
144 |
for idx, source in enumerate(res["source_documents"]):
|
145 |
source_metadata = source.metadata
|
146 |
url = source_metadata["source"]
|
|
|
147 |
|
148 |
if url not in source_dict:
|
149 |
-
source_dict[url] = [source.page_content]
|
150 |
else:
|
151 |
-
source_dict[url].append(source.page_content)
|
152 |
|
153 |
for source_idx, (url, text_list) in enumerate(source_dict.items()):
|
154 |
full_text = ""
|
155 |
-
for url_idx, text in enumerate(text_list):
|
156 |
-
full_text += f"Source {url_idx+1}:\n
|
157 |
source_elements.append(cl.Text(name=url, content=full_text))
|
158 |
-
found_sources.append(url)
|
159 |
|
160 |
if found_sources:
|
161 |
-
answer += f"\n\nSources: {', '.join(found_sources)}
|
162 |
else:
|
163 |
answer += f"\n\nNo source found."
|
164 |
|
|
|
4 |
from urllib.parse import urlparse
|
5 |
import chainlit as cl
|
6 |
from langchain import PromptTemplate
|
7 |
+
|
8 |
try:
|
9 |
from modules.constants import *
|
10 |
except:
|
|
|
61 |
|
62 |
def get_subpage_links(self, l, base_url):
|
63 |
for link in tqdm(l):
|
64 |
+
print("checking link:", link)
|
65 |
if not link.endswith("/"):
|
66 |
l[link] = "Checked"
|
67 |
dict_links_subpages = {}
|
|
|
110 |
base_url = f"{parsed_url.scheme}://{parsed_url.netloc}/"
|
111 |
return base_url
|
112 |
|
113 |
+
|
114 |
def get_prompt(config):
|
115 |
if config["llm_params"]["use_history"]:
|
116 |
if config["llm_params"]["llm_loader"] == "local_llm":
|
|
|
136 |
)
|
137 |
return prompt
|
138 |
|
139 |
+
|
140 |
def get_sources(res, answer):
|
141 |
source_elements_dict = {}
|
142 |
source_elements = []
|
|
|
147 |
for idx, source in enumerate(res["source_documents"]):
|
148 |
source_metadata = source.metadata
|
149 |
url = source_metadata["source"]
|
150 |
+
score = source_metadata.get("score", "N/A")
|
151 |
|
152 |
if url not in source_dict:
|
153 |
+
source_dict[url] = [(source.page_content, score)]
|
154 |
else:
|
155 |
+
source_dict[url].append((source.page_content, score))
|
156 |
|
157 |
for source_idx, (url, text_list) in enumerate(source_dict.items()):
|
158 |
full_text = ""
|
159 |
+
for url_idx, (text, score) in enumerate(text_list):
|
160 |
+
full_text += f"Source {url_idx + 1} (Score: {score}):\n{text}\n\n\n"
|
161 |
source_elements.append(cl.Text(name=url, content=full_text))
|
162 |
+
found_sources.append(f"{url} (Score: {score})")
|
163 |
|
164 |
if found_sources:
|
165 |
+
answer += f"\n\nSources: {', '.join(found_sources)}"
|
166 |
else:
|
167 |
answer += f"\n\nNo source found."
|
168 |
|
code/modules/llm_tutor.py
CHANGED
@@ -12,7 +12,7 @@ import os
|
|
12 |
from modules.constants import *
|
13 |
from modules.helpers import get_prompt
|
14 |
from modules.chat_model_loader import ChatModelLoader
|
15 |
-
from modules.vector_db import VectorDB
|
16 |
|
17 |
|
18 |
class LLMTutor:
|
@@ -34,19 +34,25 @@ class LLMTutor:
|
|
34 |
|
35 |
# Retrieval QA Chain
|
36 |
def retrieval_qa_chain(self, llm, prompt, db):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
37 |
if self.config["llm_params"]["use_history"]:
|
38 |
memory = ConversationBufferWindowMemory(
|
39 |
-
|
40 |
-
|
|
|
|
|
41 |
)
|
42 |
qa_chain = ConversationalRetrievalChain.from_llm(
|
43 |
llm=llm,
|
44 |
chain_type="stuff",
|
45 |
-
retriever=
|
46 |
-
search_kwargs={
|
47 |
-
"k": self.config["embedding_options"]["search_top_k"]
|
48 |
-
}
|
49 |
-
),
|
50 |
return_source_documents=True,
|
51 |
memory=memory,
|
52 |
combine_docs_chain_kwargs={"prompt": prompt},
|
@@ -55,11 +61,7 @@ class LLMTutor:
|
|
55 |
qa_chain = RetrievalQA.from_chain_type(
|
56 |
llm=llm,
|
57 |
chain_type="stuff",
|
58 |
-
retriever=
|
59 |
-
search_kwargs={
|
60 |
-
"k": self.config["embedding_options"]["search_top_k"]
|
61 |
-
}
|
62 |
-
),
|
63 |
return_source_documents=True,
|
64 |
chain_type_kwargs={"prompt": prompt},
|
65 |
)
|
|
|
12 |
from modules.constants import *
|
13 |
from modules.helpers import get_prompt
|
14 |
from modules.chat_model_loader import ChatModelLoader
|
15 |
+
from modules.vector_db import VectorDB, VectorDBScore
|
16 |
|
17 |
|
18 |
class LLMTutor:
|
|
|
34 |
|
35 |
# Retrieval QA Chain
|
36 |
def retrieval_qa_chain(self, llm, prompt, db):
|
37 |
+
retriever = VectorDBScore(
|
38 |
+
vectorstore=db,
|
39 |
+
search_type="similarity_score_threshold",
|
40 |
+
search_kwargs={
|
41 |
+
"score_threshold": self.config["embedding_options"]["score_threshold"],
|
42 |
+
"k": self.config["embedding_options"]["search_top_k"],
|
43 |
+
},
|
44 |
+
)
|
45 |
if self.config["llm_params"]["use_history"]:
|
46 |
memory = ConversationBufferWindowMemory(
|
47 |
+
k=self.config["llm_params"]["memory_window"],
|
48 |
+
memory_key="chat_history",
|
49 |
+
return_messages=True,
|
50 |
+
output_key="answer",
|
51 |
)
|
52 |
qa_chain = ConversationalRetrievalChain.from_llm(
|
53 |
llm=llm,
|
54 |
chain_type="stuff",
|
55 |
+
retriever=retriever,
|
|
|
|
|
|
|
|
|
56 |
return_source_documents=True,
|
57 |
memory=memory,
|
58 |
combine_docs_chain_kwargs={"prompt": prompt},
|
|
|
61 |
qa_chain = RetrievalQA.from_chain_type(
|
62 |
llm=llm,
|
63 |
chain_type="stuff",
|
64 |
+
retriever=retriever,
|
|
|
|
|
|
|
|
|
65 |
return_source_documents=True,
|
66 |
chain_type_kwargs={"prompt": prompt},
|
67 |
)
|
code/modules/vector_db.py
CHANGED
@@ -1,7 +1,10 @@
|
|
1 |
import logging
|
2 |
import os
|
3 |
import yaml
|
4 |
-
from langchain.vectorstores import FAISS
|
|
|
|
|
|
|
5 |
|
6 |
try:
|
7 |
from modules.embedding_model_loader import EmbeddingModelLoader
|
@@ -15,6 +18,24 @@ except:
|
|
15 |
from helpers import *
|
16 |
|
17 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
18 |
class VectorDB:
|
19 |
def __init__(self, config, logger=None):
|
20 |
self.config = config
|
@@ -61,10 +82,12 @@ class VectorDB:
|
|
61 |
return files, urls
|
62 |
|
63 |
def clean_url_list(self, urls):
|
64 |
-
# get lecture pdf links
|
65 |
lecture_pdfs = [link for link in urls if link.endswith(".pdf")]
|
66 |
lecture_pdfs = [link for link in lecture_pdfs if "lecture" in link.lower()]
|
67 |
-
urls = [
|
|
|
|
|
68 |
|
69 |
return urls, lecture_pdfs
|
70 |
|
@@ -81,6 +104,18 @@ class VectorDB:
|
|
81 |
self.vector_db = FAISS.from_documents(
|
82 |
documents=document_chunks, embedding=self.embedding_model
|
83 |
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
84 |
self.logger.info("Completed initializing vector_db")
|
85 |
|
86 |
def create_database(self):
|
@@ -89,7 +124,8 @@ class VectorDB:
|
|
89 |
files, urls = self.load_files()
|
90 |
urls, lecture_pdfs = self.clean_url_list(urls)
|
91 |
files += lecture_pdfs
|
92 |
-
|
|
|
93 |
document_chunks, document_names = data_loader.get_chunks(files, urls)
|
94 |
self.logger.info("Completed loading data")
|
95 |
|
@@ -97,29 +133,46 @@ class VectorDB:
|
|
97 |
self.initialize_database(document_chunks, document_names)
|
98 |
|
99 |
def save_database(self):
|
100 |
-
self.
|
101 |
-
|
102 |
-
|
103 |
-
|
104 |
-
|
105 |
-
|
106 |
-
|
|
|
|
|
107 |
)
|
108 |
-
|
|
|
|
|
109 |
self.logger.info("Saved database")
|
110 |
|
111 |
def load_database(self):
|
112 |
self.create_embedding_model()
|
113 |
-
self.
|
114 |
-
|
115 |
-
|
116 |
-
|
117 |
-
|
118 |
-
|
119 |
-
|
120 |
-
|
121 |
-
|
122 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
123 |
self.logger.info("Loaded database")
|
124 |
return self.vector_db
|
125 |
|
|
|
1 |
import logging
|
2 |
import os
|
3 |
import yaml
|
4 |
+
from langchain.vectorstores import FAISS, Chroma
|
5 |
+
from langchain.schema.vectorstore import VectorStoreRetriever
|
6 |
+
from langchain.callbacks.manager import CallbackManagerForRetrieverRun
|
7 |
+
from langchain.schema.document import Document
|
8 |
|
9 |
try:
|
10 |
from modules.embedding_model_loader import EmbeddingModelLoader
|
|
|
18 |
from helpers import *
|
19 |
|
20 |
|
21 |
+
class VectorDBScore(VectorStoreRetriever):
|
22 |
+
# See https://github.com/langchain-ai/langchain/blob/61dd92f8215daef3d9cf1734b0d1f8c70c1571c3/libs/langchain/langchain/vectorstores/base.py#L500
|
23 |
+
def _get_relevant_documents(
|
24 |
+
self, query: str, *, run_manager: CallbackManagerForRetrieverRun
|
25 |
+
):
|
26 |
+
docs_and_similarities = (
|
27 |
+
self.vectorstore.similarity_search_with_relevance_scores(
|
28 |
+
query, **self.search_kwargs
|
29 |
+
)
|
30 |
+
)
|
31 |
+
# Make the score part of the document metadata
|
32 |
+
for doc, similarity in docs_and_similarities:
|
33 |
+
doc.metadata["score"] = similarity
|
34 |
+
|
35 |
+
docs = [doc for doc, _ in docs_and_similarities]
|
36 |
+
return docs
|
37 |
+
|
38 |
+
|
39 |
class VectorDB:
|
40 |
def __init__(self, config, logger=None):
|
41 |
self.config = config
|
|
|
82 |
return files, urls
|
83 |
|
84 |
def clean_url_list(self, urls):
|
85 |
+
# get lecture pdf links
|
86 |
lecture_pdfs = [link for link in urls if link.endswith(".pdf")]
|
87 |
lecture_pdfs = [link for link in lecture_pdfs if "lecture" in link.lower()]
|
88 |
+
urls = [
|
89 |
+
link for link in urls if link.endswith("/")
|
90 |
+
] # only keep links that end with a '/'. Extract Files Seperately
|
91 |
|
92 |
return urls, lecture_pdfs
|
93 |
|
|
|
104 |
self.vector_db = FAISS.from_documents(
|
105 |
documents=document_chunks, embedding=self.embedding_model
|
106 |
)
|
107 |
+
elif self.db_option == "Chroma":
|
108 |
+
self.vector_db = Chroma.from_documents(
|
109 |
+
documents=document_chunks,
|
110 |
+
embedding=self.embedding_model,
|
111 |
+
persist_directory=os.path.join(
|
112 |
+
self.config["embedding_options"]["db_path"],
|
113 |
+
"db_"
|
114 |
+
+ self.config["embedding_options"]["db_option"]
|
115 |
+
+ "_"
|
116 |
+
+ self.config["embedding_options"]["model"],
|
117 |
+
),
|
118 |
+
)
|
119 |
self.logger.info("Completed initializing vector_db")
|
120 |
|
121 |
def create_database(self):
|
|
|
124 |
files, urls = self.load_files()
|
125 |
urls, lecture_pdfs = self.clean_url_list(urls)
|
126 |
files += lecture_pdfs
|
127 |
+
if "storage/data/urls.txt" in files:
|
128 |
+
files.remove("storage/data/urls.txt")
|
129 |
document_chunks, document_names = data_loader.get_chunks(files, urls)
|
130 |
self.logger.info("Completed loading data")
|
131 |
|
|
|
133 |
self.initialize_database(document_chunks, document_names)
|
134 |
|
135 |
def save_database(self):
|
136 |
+
if self.db_option == "FAISS":
|
137 |
+
self.vector_db.save_local(
|
138 |
+
os.path.join(
|
139 |
+
self.config["embedding_options"]["db_path"],
|
140 |
+
"db_"
|
141 |
+
+ self.config["embedding_options"]["db_option"]
|
142 |
+
+ "_"
|
143 |
+
+ self.config["embedding_options"]["model"],
|
144 |
+
)
|
145 |
)
|
146 |
+
elif self.db_option == "Chroma":
|
147 |
+
# db is saved in the persist directory during initialization
|
148 |
+
pass
|
149 |
self.logger.info("Saved database")
|
150 |
|
151 |
def load_database(self):
|
152 |
self.create_embedding_model()
|
153 |
+
if self.db_option == "FAISS":
|
154 |
+
self.vector_db = FAISS.load_local(
|
155 |
+
os.path.join(
|
156 |
+
self.config["embedding_options"]["db_path"],
|
157 |
+
"db_"
|
158 |
+
+ self.config["embedding_options"]["db_option"]
|
159 |
+
+ "_"
|
160 |
+
+ self.config["embedding_options"]["model"],
|
161 |
+
),
|
162 |
+
self.embedding_model,
|
163 |
+
allow_dangerous_deserialization=True,
|
164 |
+
)
|
165 |
+
elif self.db_option == "Chroma":
|
166 |
+
self.vector_db = Chroma(
|
167 |
+
persist_directory=os.path.join(
|
168 |
+
self.config["embedding_options"]["db_path"],
|
169 |
+
"db_"
|
170 |
+
+ self.config["embedding_options"]["db_option"]
|
171 |
+
+ "_"
|
172 |
+
+ self.config["embedding_options"]["model"],
|
173 |
+
),
|
174 |
+
embedding_function=self.embedding_model,
|
175 |
+
)
|
176 |
self.logger.info("Loaded database")
|
177 |
return self.vector_db
|
178 |
|