Spaces:
Sleeping
Sleeping
Commit History
-
Update README.md
Verified signatureEdit README to reflect current status as of 20/09/25 for purpose of passdown.e3e814a
-
feat: Ignore brand associations from training data
Analysis of responses shows that the LLM tends to tag quotes with characteristics marketed by the brand even when only the brand name is given. Example: Quote: "Brand X", tagged: "Natural" Modify prompt to ignore all learned brand associations from training data and just tag based on what the respondent has said. Also create new folder for prompt archival. Not relying on git versioning to track prompts as of now to make it easy to reconfigure old prompts for comparing and contrasting outputs.6ac00ea
-
feat: add translation to English for quotes
Add quote translation. Translation is not threaded so it will take a short while after tagging all the quotes. Translation is also done by OpenAI so results may vary if compared with Google Translate.01355c1 -
feat: add random quote retrieval for each tag
-Also add one hot encoding for tags -Fix bug where app throws error when retained_columns is empty -Remove limit from ThreadPoolExecutor after determining impact on rate limit is insignificantea8f271
-
feat: Introduce threading for OpenAI API calls
Implement concurrent processing for quotes tagging to speed up processing time.424ebdd
-
fix: update Python version for compatibility
e0618e8 -
feat: \n-separated taglist, progress bar, etc
Added new features: 1)Split tag list input by user by newlines instead of commas to facilitate copy-paste input from Excel 2)Add progress bar to display what the app is doing and how many quotes have been tagged. Accordingly the main function now yields instead of returns. 3)Add option to allow user to specify columns to be retained in output 4)Tags in output are converted from list to string to remove brackets from Excel output file76b64ac
-
feat: add setup for temp users
Add setup for alternative signin credentials for temp users and a function check_auth to handle both these and the permanent credentials. To setup a temp user for the app, their username and password should be set in environment variables TEMP_USERNAME and TEMP_PASSWORD and the expiry of their access should be set in TEMP_EXPIRY_TIME_SG_ISO_8601, in Singapore time and ISO 8601 format.f792567
-
feat: add invalid tag filtering
Add filtering for all tags to ensure they are taken from tags list. Warning will be raised if tags not from tag list are encountered.5024bb9 -
fix: change pydantic version
Encountering error: TypeError: argument of type 'bool' is not iterable ERROR: Exception in ASGI application Updating pydantic version to 2.10.6 as dependency issue is suspected3bf592f -
feat: add per-quote logging
Add logging to track progress. Change from applying the tag_quote func to using an explicit for loop to deliver logs after each quote.95e3994
-
feat: include hash table of tagged tags as output
-Added function count_tags which counts occurrences of each tag tagged and returns the data in a DataFrame -Updated Interface to also display above DataFrame -Output file link now includes data from both DataFrames in separate sheets -Updated user instructions to reflect above changes -Amend: update documentation comments for Counter import243f5cb -
fix: added ssr_mode=False to fix login
Auth is not working on HF spaces, trying a few fixes found online.0c5d2b6 -
fix: update requirements.txt
dotenv->python-dotenv2df2718 -
feat: add authentication
Now enabled authentication by anyone with a password to be defined in the environment variables on HF. This enables going public with the app rather than restricting it to HF organisation members only.7e3947d
-
feat: Allow result to be downloaded as Excel file
-Updated gradio version in requirements.txt from 4.31.5 to 5.12.0 to fix compatibilty issue with python 3.13. audioop-lts module was added to gradio in 5.4.0 to compensate for removal of audioop module in Python 3.13+. See https://github.com/gradio-app/gradio/pull/9757. -Added newlines to ends of .txt files and app.py -Modified Gradio interface in app.py to give an additional file as output. File is created by converting the original output DataFrame using df.to_excel. -Added dotenv package for easier local testing with .env file. -Added .gitignore to ignore venv folder and .env file when pushing to origin. -Amend: added dotenv 1.0.1 to requirements.txt. -Amend 2: Non-urgent issue requiring fix: output.xlsx is created in local directory when testing locally. Will be good to cleanup when local testing server is shut down. -Amend 3: Added output.xlsx to .gitignore.553817d
-
Merge remote-tracking branch 'origin/main'
32f45ca
-
docs: Docstrings, comments and type hints added
Also moved the prompt and user instructions to new files prompt.txt and user_instructions.txt to de-clutter the app.py.f6a99c0
-
Update app.py
Verified signatureeaa8a1d -
Update app.py
Verified signature5681111 -
fix: now retrieves org key and proj key from env
4716193
-
feat: version 1.0
Allows user to input a file containing quotes in a column, the column name, and a list of tags. OpenAI gpt-4o-mini is used to tag each quote with appropriate tags from the list. Future potential improvements: 1: Allow upload of tags as a file if deemed to be useful by users. 2: Introduce tag recommendations by getting GPT to crunch the input quotes. Amend with bugfix: added a missing = sign in requirements.txt23192ed
-
Adding boilerplate code
9a3ebb6 -
initial commit
Verified signature8a9f41f