Spaces:
Sleeping
Sleeping
File size: 1,886 Bytes
6df9459 |
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 |
from langchain_text_splitters import CharacterTextSplitter
from langchain_huggingface import HuggingFaceEmbeddings
from langchain_chroma import Chroma
from langchain.docstore.document import Document
import pandas as pd
import os
import glob
# Define a function to perform vectorization for multiple CSV files
def vectorize_documents():
embeddings = HuggingFaceEmbeddings()
# Directory containing multiple CSV files
csv_directory = "Data" # Replace with your folder name
csv_files = glob.glob(os.path.join(csv_directory, "*.csv")) # Find all CSV files in the folder
documents = []
# Load and concatenate all CSV files
for file_path in csv_files:
df = pd.read_csv(file_path)
for _, row in df.iterrows():
# Combine all columns in the row into a single string
row_content = " ".join(row.astype(str))
documents.append(Document(page_content=row_content))
# Splitting the text and creating chunks of these documents
text_splitter = CharacterTextSplitter(
chunk_size=2000,
chunk_overlap=500
)
text_chunks = text_splitter.split_documents(documents)
# Process text chunks in batches
batch_size = 5000 # Chroma's batch size limit is 5461, set a slightly smaller size for safety
for i in range(0, len(text_chunks), batch_size):
batch = text_chunks[i:i + batch_size]
# Store the batch in Chroma vector DB
vectordb = Chroma.from_documents(
documents=batch,
embedding=embeddings,
persist_directory="vector_db_dir"
)
print("Documents Vectorized and saved in VectorDB")
# Expose embeddings if needed
embeddings = HuggingFaceEmbeddings()
# Main guard to prevent execution on import
if __name__ == "__main__":
vectorize_documents()
|