klbr/khoj
1
0
Fork 0
mirror of https://github.com/khoaliber/khoj.git synced 2026-03-06 21:29:12 +00:00
Code Issues Packages Projects Releases Wiki Activity

Split entries by max tokens while converting Markdown entries To JSONL

Browse Source
This commit is contained in:
Debanjum Singh Solanky
2022-12-26 13:14:15 -03:00
parent 24676f95d8
commit f209e30a3b
1 changed files with 6 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
src/processor/markdown/markdown_to_jsonl.py
View File

@@ -35,6 +35,12 @@ class MarkdownToJsonl(TextToJsonl):
end = time.time() end = time.time()
logger.debug(f"Parse entries from Markdown files into dictionaries: {end - start} seconds") logger.debug(f"Parse entries from Markdown files into dictionaries: {end - start} seconds")
# Split entries by max tokens supported by model
start = time.time()
current_entries = self.split_entries_by_max_tokens(current_entries, max_tokens=256)
end = time.time()
logger.debug(f"Split entries by max token size supported by model: {end - start} seconds")
# Identify, mark and merge any new entries with previous entries # Identify, mark and merge any new entries with previous entries
start = time.time() start = time.time()
if not previous_entries: if not previous_entries: