blade
ea81197bcd
update for apply ALI QWEN as Demo
2025-11-11 13:33:57 +08:00
blade
255752c848
add mini_main.py
2025-11-10 16:55:55 +08:00
37cf06a394
Confirm span pages calculation, the management fee and costs page only with management_fee_and_costs and management_fee datapoints
2025-04-03 18:08:27 -05:00
f333cc30f5
1. fit the scenario when document type is not 1 or 4, 5
...
2. support the scenario:
"investment fees and costs including performance" statement in performance fee data page, instead of in management fee and costs data page.
2025-04-03 17:06:43 -05:00
4b896f4460
update latest metrics based on optimized matching algorithm
...
Support: 2663 -> 2773
percentage of Share Matched: 80.08 -> 82.59
F1: 0.956 -> 0.943
2025-04-02 20:39:31 -05:00
427a379b3b
1. support re-call ChatGPT API to match non-matched prediction fund/ share names
...
2. If document fund amount less than 3, cancel the production name judgment logic
2025-04-02 16:34:41 -05:00
4cee95db9a
fix issue for post actions
2025-03-31 22:04:31 -05:00
50e51e0894
recover main.py
2025-03-31 17:16:05 -05:00
a42033f848
Merge branch 'aus_prospectus_ravi' of https://msstash.morningstar.com/scm/dc/dc-ml-emea-ar into aus_prospectus_ravi
2025-03-31 17:09:06 -05:00
984c686bf3
support separate tables and pages data which with specific biz rules
2025-03-31 17:08:49 -05:00
ac6332ad46
Merge branch 'aus_prospectus_ravi' of https://msstash.morningstar.com/scm/dc/dc-ml-emea-ar into aus_prospectus_ravi
2025-03-28 08:36:58 -05:00
f7b9652c75
gitignore a virtual environment directory
2025-03-28 08:36:44 -05:00
355b145cf7
If found total_annual_dollar_based_charges and could be divisible by 52 or 12,
...
then set the fund name and share name to be document production name
2025-03-28 01:33:33 -05:00
46f86b124b
update instructions fund name section structure
2025-03-28 00:51:51 -05:00
8a5723c150
optimize for Entry Fee/ Nil Entry case
2025-03-27 21:10:33 -05:00
d925992326
1. Support the keywords of complex special cases to be regex
...
2. Support set sub-datapoints list to complex special cases node.
3. Simplify the common management fee and costs instructions.
4. Add markdown title characters: ## or ### to instructions.
2025-03-27 16:00:19 -05:00
dc560e1e01
update metrics
2025-03-26 23:14:28 -05:00
ff2325c72d
1. fix issue for assign values based on production name
...
2. optimize instructions for extract non-necessary data by Cost of Product message
2025-03-26 18:58:45 -05:00
8ad472fb39
UPDATE metrics code file
2025-03-24 18:00:53 -05:00
dd1f8f76ae
update for metrics
2025-03-24 17:12:13 -05:00
4edc4b4768
clean code
2025-03-24 17:10:16 -05:00
9be6d1296d
update benchmark check logic
2025-03-19 00:52:25 -05:00
5ba39a394b
1. keep fund/ share db list before applying LLM
...
2. add key words for interposed_vehicle_performance_fee_cost
2025-03-18 22:15:31 -05:00
c71936c5ff
1. optimize benchmark_name instructions
...
2. consider possible with multiple same raw fund names in documents, not to remove unmatched_db_list when match relevant raw fund/ share name
Otherwise, it will occur some raw names couldn't match db name issue.
2025-03-18 17:22:21 -05:00
0cea2e501b
For AUS Prospectus, cancel visiting Vision ChatGPT when page contents without any numeric text or perhaps with messy code.
...
(But should keep this logic for EMEA LUX AR, because of some special providers cases for this market documents.)
2025-03-18 14:15:43 -05:00
6614972849
Raw Code added to identify benchmark names
2025-03-18 18:57:08 +05:30
0ad17e338e
Code added to save anomilities
2025-03-18 17:56:50 +05:30
2817490652
Code added to save anomilities
2025-03-18 17:54:33 +05:30
ad371f6584
Changed Performance matrix code to get all anomilities to analyze and Prompt to get better accuracy
2025-03-18 16:43:55 +05:30
b3941ee4b3
update instructions for total_annual_dollar_based_charges
2025-03-17 15:07:02 -05:00
0ce604021c
updating gitignore
2025-03-17 18:52:49 +05:30
dc9180ca1b
Rollback file name
2025-03-17 17:09:48 +05:30
af3d1222a6
Changes done for Bugfix: 1. SSL issue \n2. Ignore Example Tables \n3. Performacne fee
2025-03-17 17:07:08 +05:30
dd15c1c48e
Optimize for benchmark name
2025-03-14 11:51:10 -05:00
bceff71fa4
Set re-run parameters to be True
...
re_run_extract_data = True
re_run_mapping_data = True
force_save_total_data = True
2025-03-14 04:03:22 -05:00
0f65537478
optimize instructions for minimum_initial_investment
2025-03-14 04:02:15 -05:00
f539340d04
1. optimize instructions
...
Only load relevant fund name for investment objective, instead of full page text with the most recent investment objective
2. Exclude the table which with only one numeric column: Cost Product
2025-03-14 01:04:51 -05:00
551f754379
Fix issue when saving data extraction data
2025-03-13 18:36:04 -05:00
a48af9ddf0
A. Metrics score
...
Blade's updates
1. Set the secondary key to be the share class name, instead of the fund name
2. Remove the data point which support is 0 to calculate the metrics
3. Add the message list to store the error message
4. Support save metrics/ error message to excel file
5. Support statistics for different document list
6. Set F1-Score to the first column in the metrics table
B. Optimize instructions for benchmark_name
2025-03-13 17:52:06 -05:00
a090b5cc9e
1. metrics's key should be share class name: sec_name
...
2. support output metrics data as Excel file
3. Optimize instructions for performance_fee_costs
2025-03-13 11:53:27 -05:00
1f6b781b12
Merge branches 'aus_prospectus_ravi' and 'aus_prospectus_ravi' of https://msstash.morningstar.com/scm/dc/dc-ml-emea-ar into aus_prospectus_ravi
2025-03-13 17:34:35 +05:30
97da7e4961
Added code to identify anomaly cases and performance matrix and updated for pdf downloading code
2025-03-13 17:31:54 +05:30
fd2430082c
optimize instructions for management_fee_and_costs and buy_spread, sell_spread
2025-03-13 02:59:19 -05:00
336fd9a24f
Merge branch 'aus_prospectus_ravi' of https://msstash.morningstar.com/scm/dc/dc-ml-emea-ar into aus_prospectus_ravi
2025-03-13 11:39:30 +05:30
fb5dda2170
1. optimize performance_fee_costs prompts
...
2. support calculate metrics by zero equal with empty
2025-03-12 23:45:52 -05:00
c2c0b33015
align fund name based on production name
...
optimize performance relevant prompts
2025-03-12 21:52:00 -05:00
6f17c2253c
optimize instructions for document 412778803
2025-03-12 17:24:39 -05:00
765772e5a8
optimize performance_fee_costs by document 391080133
2025-03-12 14:45:48 -05:00
76fbb7c071
Merge branches 'aus_prospectus_ravi' and 'aus_prospectus_ravi' of https://msstash.morningstar.com/scm/dc/dc-ml-emea-ar into aus_prospectus_ravi
2025-03-12 14:16:27 +05:30
c7c36dbdd2
1. update performance_fee name to performance_fee_costs
...
2. support extract data for total_annual_dollar_based_charges
2025-03-11 17:15:39 -05:00
blade
Blade He
Russell Spence
Ravi Maheshwari