Spaces:
Runtime error
Runtime error
Commit History
-
updated the latestcontent of:
Roland Dingrevised the UI of the app to add dropdown for IFU added features for incorporating IFU data in prompting updated upload conditions as per latest search_term sheet. On branch main Changes to be committed: modified: __pycache__/supplier.cpython-310.pyc modified: __pycache__/utility.cpython-310.pyc modified: backend_update.ipynb modified: chains.py modified: features.py modified: ui_study.py498a219
-
updated the chains.py by llm retry number and async_generate error message
Roland Ding3a74227 -
revised the async_generate from chains.py
Roland Dingremoved the retry_decorator from async_generate. async_generate will now add the timeout error message in the output instead of retrying. On branch main Changes to be committed: modified: chains.py5cef2da
-
use llm from chains.py as universal AI agent.
Roland Ding721214f
-
11.11.27.77
Roland Dingupdated as per request 2023-11-23 + added pre_view table heading and others extension + updated the assessment categories to lower cases. + added logic to the article revision. + extended the task generation into input_variables and chat_prompt generation functions. + added reformat_instruction function to the application.py due to generalized instruction format. On branch main Changes to be committed: modified: .data/instruction_agg_performance.json modified: application.py modified: backend_update.ipynb modified: chains.py modified: features.pyc769f48
-
11.11.27.77 major update
Roland Dingfeatures: + all features revised to align with the latest prompts and terms structure from aws cloud (search term v 14i) + added gen_task for post_process and execute_concurrent to declutter the code + optimized the async generation to generalize its functionality and remove the cluttering code. + removed gen_segment and uses async_generation for segment generation and post_process. + revised the term validation + removed unused functions like execute_prompts, run_gpt, f_replacement_term, f_summary_term, run_executor + revised the post_process to make it run async. external: + updated openai llm instance with retries and timeout to avoid the extensive wait and hitting rate limit issue. data: + added current and term in the session data to resolve the issue of consecutive prompts were called but not triggered by any terms. ui: + simplified the detailed siplay view to show only the aggregated result. On branch main Changes to be committed: modified: application.py modified: cha…8dfdf87
-
Major backend_update
Roland DingUI update: + Revised ui_studies to: - show all studies articles - added control for multiple files upload + Revised ui_study to: - update the article list in real time - realigned the layout to make it more compact Features: + added backend methods to upload multiple files + added post_process function to execute Futable commends + added retry_decorator for retrying failed openai inquiries + touchup for including futable results in extraction list for display + changed extraction collections to set + move the llm instance to global in chains.py + added root message for agent declaration before the first inquiry Cloud: + updated get table for handling pagination. Others: + touchup remove_symbols in utility.py for better filtering article content and characters. cleanup: + removed cloud_textract.py On branch main Changes to be committed: modified: chains.py modified: cloud_db.py deleted: cloud_textract.py modified: features.py modified: ui_studies…fdccd1a
-
9.10.25.78 house keeping
Roland Ding+ Cleaning up the chains.py file + added function place holder for the reference table. (May not be necessary since originally, it was created to fix the article reference capture issue. Since now we no longer using textract at all, this may not be necessary) + added set datatype transformation in all aws data conversion functions in utility.py + added a set of new code blocks to the backend_update.ipynb file to update the backend database as per search term sheet. On branch main Changes to be committed: modified: backend_update.ipynb modified: chains.py modified: cloud_textract.py modified: utility.py5b59400
-
9.9.24.74 updated the async function terminal
Roland DingAfter the updated of async functions, the original terminal_print function is not working for async functions. This commit is to add an aterminal_print as the async version of terminal_print to wrap around async functions.da2ce27
-
9.9.24.72 updated recursion for chains.py
Roland Dingupdate execute_concurrent with recursion for traverse along the prompt chain. On branch main modified: chains.pya6d0043
-
9.9.23.71 updated llm initialization with openai api key in gen_segments and execute_concurrent.
Roland Dingd5f1fce -
9.9.23.69 try to fix huggingface openai api key by changing application.py os.envron.get to os.getenv.
Roland Dingprovided the api key to the chatmodel in chains.py modified: application.py modified: chains.py6d0778c
-
9.9.23.68 revise and correction as per email
Roland DingAs per email from 2023-10-09, after review of the cross check results, the following changes were made to the algorithm: + chains.py updated term argument detection in the dictionary for the case of keyword abscence + features.py updated term validation from any to all + features.py excluded introduction from the key_content in the article for the term search. + features.py removed all return symbol before search for any key terms to ensure the terms were picked up properly. Next step: + features.py - add the ability to stream line the prompts for processing cascading and combination prompts. modified: chains.py modified: features.py0baa835
-
9.9.22.67 mass update of the application
Roland Ding+ Revised internally to apply usage of langchain and async call + Realigned the application to use the new terms and prompts from search term 13n + Added the new chains.py module to support the new langchain + revised the application.py to align with all the new backend data structure. modified: app.py modified: application.py new file: chains.py modified: features.py modified: requirements.txt modified: supplier.py modified: ui_studies.py modified: ui_study.py82b9d78