Spaces:
Sleeping
Sleeping
Asaad Almutareb
commited on
Commit
•
1314610
1
Parent(s):
aae1d57
added persistent_directory to chroma client
Browse files
app.py
CHANGED
@@ -20,6 +20,7 @@ import os
|
|
20 |
dotenv.load_dotenv()
|
21 |
config = ConfigParser()
|
22 |
config.read('innovation_pathfinder_ai/config.ini')
|
|
|
23 |
|
24 |
logger = logger.get_console_logger("app")
|
25 |
|
@@ -28,7 +29,9 @@ app = FastAPI()
|
|
28 |
def initialize_chroma_db() -> Chroma:
|
29 |
collection_name = config.get('main', 'CONVERSATION_COLLECTION_NAME')
|
30 |
|
31 |
-
client = chromadb.PersistentClient(
|
|
|
|
|
32 |
|
33 |
collection = client.get_or_create_collection(
|
34 |
name=collection_name,
|
|
|
20 |
dotenv.load_dotenv()
|
21 |
config = ConfigParser()
|
22 |
config.read('innovation_pathfinder_ai/config.ini')
|
23 |
+
persist_directory = config.get('main', 'VECTOR_DATABASE_LOCATION')
|
24 |
|
25 |
logger = logger.get_console_logger("app")
|
26 |
|
|
|
29 |
def initialize_chroma_db() -> Chroma:
|
30 |
collection_name = config.get('main', 'CONVERSATION_COLLECTION_NAME')
|
31 |
|
32 |
+
client = chromadb.PersistentClient(
|
33 |
+
path=persist_directory
|
34 |
+
)
|
35 |
|
36 |
collection = client.get_or_create_collection(
|
37 |
name=collection_name,
|
innovation_pathfinder_ai/knowledge_base/placeholder.txt
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
this file is needed
|
innovation_pathfinder_ai/structured_tools/structured_tools.py
CHANGED
@@ -37,6 +37,7 @@ from configparser import ConfigParser
|
|
37 |
|
38 |
config = ConfigParser()
|
39 |
config.read('innovation_pathfinder_ai/config.ini')
|
|
|
40 |
|
41 |
@tool
|
42 |
def memory_search(query:str) -> str:
|
@@ -44,7 +45,7 @@ def memory_search(query:str) -> str:
|
|
44 |
This is your primary source to start your search with checking what you already have learned from the past, before going online."""
|
45 |
# Since we have more than one collections we should change the name of this tool
|
46 |
client = chromadb.PersistentClient(
|
47 |
-
|
48 |
)
|
49 |
|
50 |
collection_name = config.get('main', 'CONVERSATION_COLLECTION_NAME')
|
@@ -70,7 +71,7 @@ def knowledgeBase_search(query:str) -> str:
|
|
70 |
"""Search the internal knowledge base for research papers and relevent chunks"""
|
71 |
# Since we have more than one collections we should change the name of this tool
|
72 |
client = chromadb.PersistentClient(
|
73 |
-
|
74 |
)
|
75 |
|
76 |
collection_name="ArxivPapers"
|
@@ -144,7 +145,7 @@ def embed_arvix_paper(paper_id:str) -> None:
|
|
144 |
paper.download_pdf(dirpath=pdf_directory, filename=f"{number_without_period}.pdf")
|
145 |
|
146 |
client = chromadb.PersistentClient(
|
147 |
-
|
148 |
)
|
149 |
|
150 |
collection_name="ArxivPapers"
|
|
|
37 |
|
38 |
config = ConfigParser()
|
39 |
config.read('innovation_pathfinder_ai/config.ini')
|
40 |
+
persist_directory = config.get('main', 'VECTOR_DATABASE_LOCATION')
|
41 |
|
42 |
@tool
|
43 |
def memory_search(query:str) -> str:
|
|
|
45 |
This is your primary source to start your search with checking what you already have learned from the past, before going online."""
|
46 |
# Since we have more than one collections we should change the name of this tool
|
47 |
client = chromadb.PersistentClient(
|
48 |
+
path=persist_directory,
|
49 |
)
|
50 |
|
51 |
collection_name = config.get('main', 'CONVERSATION_COLLECTION_NAME')
|
|
|
71 |
"""Search the internal knowledge base for research papers and relevent chunks"""
|
72 |
# Since we have more than one collections we should change the name of this tool
|
73 |
client = chromadb.PersistentClient(
|
74 |
+
path=persist_directory,
|
75 |
)
|
76 |
|
77 |
collection_name="ArxivPapers"
|
|
|
145 |
paper.download_pdf(dirpath=pdf_directory, filename=f"{number_without_period}.pdf")
|
146 |
|
147 |
client = chromadb.PersistentClient(
|
148 |
+
path=persist_directory,
|
149 |
)
|
150 |
|
151 |
collection_name="ArxivPapers"
|
innovation_pathfinder_ai/vector_store/chroma_vector_store.py
CHANGED
@@ -31,9 +31,7 @@ import os
|
|
31 |
dotenv.load_dotenv()
|
32 |
config = ConfigParser()
|
33 |
config.read('innovation_pathfinder_ai/config.ini')
|
34 |
-
|
35 |
-
VECTOR_DATABASE_LOCATION = config.get('main', 'VECTOR_DATABASE_LOCATION')
|
36 |
-
|
37 |
|
38 |
|
39 |
def read_markdown_file(file_path: str) -> str:
|
@@ -91,7 +89,7 @@ def add_markdown_to_collection(
|
|
91 |
splits = text_splitter.split_documents(md_header_splits)
|
92 |
|
93 |
client = chromadb.PersistentClient(
|
94 |
-
|
95 |
)
|
96 |
|
97 |
|
@@ -176,7 +174,7 @@ def add_pdf_to_vector_store(
|
|
176 |
|
177 |
|
178 |
client = chromadb.PersistentClient(
|
179 |
-
|
180 |
)
|
181 |
|
182 |
collection = client.get_or_create_collection(
|
@@ -210,7 +208,7 @@ if __name__ == "__main__":
|
|
210 |
collection_name="ArxivPapers"
|
211 |
|
212 |
client = chromadb.PersistentClient(
|
213 |
-
|
214 |
)
|
215 |
|
216 |
# delete existing collection
|
@@ -238,7 +236,7 @@ if __name__ == "__main__":
|
|
238 |
|
239 |
#create the cliient using Chroma's library
|
240 |
client = chromadb.PersistentClient(
|
241 |
-
|
242 |
)
|
243 |
|
244 |
# This is an example collection name
|
|
|
31 |
dotenv.load_dotenv()
|
32 |
config = ConfigParser()
|
33 |
config.read('innovation_pathfinder_ai/config.ini')
|
34 |
+
persist_directory = config.get('main', 'VECTOR_DATABASE_LOCATION')
|
|
|
|
|
35 |
|
36 |
|
37 |
def read_markdown_file(file_path: str) -> str:
|
|
|
89 |
splits = text_splitter.split_documents(md_header_splits)
|
90 |
|
91 |
client = chromadb.PersistentClient(
|
92 |
+
path=persist_directory,
|
93 |
)
|
94 |
|
95 |
|
|
|
174 |
|
175 |
|
176 |
client = chromadb.PersistentClient(
|
177 |
+
path=persist_directory,
|
178 |
)
|
179 |
|
180 |
collection = client.get_or_create_collection(
|
|
|
208 |
collection_name="ArxivPapers"
|
209 |
|
210 |
client = chromadb.PersistentClient(
|
211 |
+
path=persist_directory,
|
212 |
)
|
213 |
|
214 |
# delete existing collection
|
|
|
236 |
|
237 |
#create the cliient using Chroma's library
|
238 |
client = chromadb.PersistentClient(
|
239 |
+
path=persist_directory,
|
240 |
)
|
241 |
|
242 |
# This is an example collection name
|