f9ef4cec96
update sql_query cache file store location
...
At most cache 5 days, then clean from local disk.
2025-01-31 10:59:54 -06:00
7f37f3532f
switch example document
2025-01-27 14:59:26 -06:00
6f831e241c
Merge branch 'aus_prospectus_ravi'
2025-01-27 12:32:42 -06:00
41f8c307ff
a little change
2025-01-27 12:32:36 -06:00
47c41e492f
1. only get name mapping data from document mapping
...
2. Compare name mapping metrics between Ravi's and mine.
2025-01-27 12:29:49 -06:00
d9b0bed39a
a little change
2025-01-22 09:57:42 -06:00
350550d1b0
fix issue for removing item from list
2025-01-21 17:24:05 -06:00
e2b9bcbdbc
initial abbreviation configurations
2025-01-21 17:09:45 -06:00
b15d260a58
migrate name mapping algorithm from Ravi
2025-01-21 16:55:08 -06:00
d41fae3dba
prepare for 100 multi-funds document samples
2025-01-17 16:26:31 -06:00
b93a8d55e8
update for output data as template
2025-01-17 11:41:58 -06:00
f10ff8ee33
update for deployment
2025-01-16 20:34:43 -06:00
fb4a6402f0
support output merged data format
2025-01-16 16:31:04 -06:00
2eace81f51
support more configurable parts
2025-01-16 13:54:45 -06:00
db0827435b
supplement EMEA AR configuration files
2025-01-16 11:30:44 -06:00
9f0e77a11e
support load configurations by doc_source parameter
2025-01-16 11:17:48 -06:00
acc30d4b72
if fail to get text by pdf to html API, then try to get text by pymupdf.
2025-01-15 18:36:02 -06:00
ace0ac2674
a little change
2025-01-15 18:22:08 -06:00
a89aa9c4de
support fetch data from Prospectus
2025-01-14 16:21:48 -06:00
e230a5bf15
a little change
2025-01-09 12:19:24 -06:00
91c86bb983
update AUS Prospectus relevant configuration
2025-01-08 17:40:57 -06:00
0a867dcf07
complete configuration for AUS Prospectus
2025-01-07 16:25:13 -06:00
201a809ffa
comment remove_abundant_data function
2025-01-06 15:27:43 -06:00
c335992ced
update requirements.txt
2025-01-06 13:56:09 -06:00
9348e32caa
support more performance fee keywords
2025-01-06 13:14:20 -06:00
65e752e25a
realize merge_output_data function, whether to output as this format, depends on confirmation with data/ developer teams
2024-12-18 09:19:55 -06:00
309bb714f6
fix issue for parsing data via Vision Function.
2024-12-11 16:49:04 -06:00
d673a99e21
switch back to extract data from image stream directly, instead of getting text from image stream as the first step, then extract data from extracted text.
...
The reason is: the quality of getting text from image steam is not good enough.
2024-12-10 16:17:47 -06:00
f71e2968cc
simplify code
2024-12-09 22:24:40 -06:00
75ea5e70de
1. support fetch data from messy-code page by ChatGPT4o Vision function.
...
2. multilingual share features configuration
2024-12-09 17:47:42 -06:00
d96f77fe00
Split share class names which with multiple share classes in same line
2024-12-06 16:31:42 -06:00
d79b05885d
optimize prompts for TOR
2024-12-06 14:50:34 -06:00
a25991e2bb
1. Set TOR reported name priority
...
2. Optimize investment mapping logic
2024-12-06 09:54:43 -06:00
95c386911c
Clean fund name after getting response from ChatGPT
2024-12-04 22:08:09 -06:00
70362b554f
Fix issue for "The last fund name of previous PDF page" logic:
...
If current page fund name starts with "The last fund name of previous PDF page" and with more contents below, then remove "The last fund name of previous PDF page".
2024-12-04 16:57:52 -06:00
36fbaa946e
Add the statement when transferring the last fund name of previous PDF page:
...
The last fund name of previous PDF page:
page_text = f"\nThe last fund name of previous PDF page: {previous_page_fund_name}\n{page_text}"
2024-12-03 11:50:31 -06:00
a11a99fdc3
1. Optimize instructions: not to fetch the data with "up to" statement.
...
2. Add exception handler in function.
2024-12-03 11:27:28 -06:00
bc32860f87
remove_abundant_data
2024-12-02 17:16:56 -06:00
c146497052
optimize share feature judgment logic:
...
accumulation with capitalisation and institutional
income with distribution
Document: 337293427
2024-12-02 13:11:49 -06:00
352886ade2
update instructions for TER, OGC, Performance Fees
2024-12-02 11:45:19 -06:00
276ff93a1d
Optimize drilldown algorithm
...
Share class names with currency
Reason
The currency in document not next to share name
Solution
If can't get relevant text from PDF page contents, and the last word of share class name belongs to currency, remove currency from share class name, then try again.
After implementing this solution, recall is from 95% to 96%
Can't find relevant text from current PDF page text
Reason
Hence apply try to merge previous page text into current page, perhaps the text is from previous page text.
Solution
Try to get previous page and search relevant value.
After implementing this solution, recall is from 96% to 98%.
2024-11-26 16:35:07 -06:00
a09778d9d1
Create EMEA AR API code file.
...
Optimize annotation list for drilldown.
2024-11-26 11:24:29 -06:00
fb356fce76
1. optimize drilldown algorithm
...
2. support calculate drilldown recall metrics
2024-11-25 15:11:03 -06:00
78fb283130
update python libraries
2024-11-25 11:11:02 -06:00
fc80093557
optimize investment mapping
2024-11-22 14:54:52 -06:00
f1c0290588
Optimize investment mapping algorithm.
...
1. Get proper currency if exist multiple currencies in share name, e.g. CHF EUR
2. Default currency should be based on scenario: USD or EUR.
3. Remove special chars should be based on \W, instead of [^a-zA-Z0-9\s]
2024-11-21 16:36:58 -06:00
5b9f9416de
1. Update for mapping multilingual share class names.
...
2. Optimize getting currency logic
2024-11-21 11:37:58 -06:00
843bbbd13f
dynamic loading instructions for multilingual.
2024-11-20 17:00:22 -06:00
067d89e0f9
Add datapoint_reportedname.json for dynamic loading reported names based on document language.
2024-11-19 16:49:15 -06:00
8223ca9a5c
a little change
2024-11-18 16:13:24 -06:00
Blade He