Refactor app.py to enhance spaces import handling, provide a fallback for non-Space environments, and restore necessary function calls for resource initialization.

Refactor app.py to streamline spaces import logic and remove commented-out code for clarity.

Refactor spaces import handling in app.py to conditionally import and provide a fallback for non-Space environments.

Refactor GPU function definitions in app.py to improve structure and ensure warmup function is called before embedding creation.

Add warmup function in app.py to initialize GPU processing with a print statement.

Remove unnecessary import of spaces in app.py

Add GPU-aware function to create embeddings for input texts in app.py

add spaces import

Refactor data processing in app.py to improve handling of dictionary-like strings in DataFrame and update file upload options to support both CSV and Pickle formats.

Comment out torch import and remove it from requirements.txt

printout running statement

bounced torch to 2.0.0

Kicked out pkl upload, fixed csv-upload. Update requirements.txt to specify torch version and add gradio dependency.

Enhance data processing in app.py and openalex_utils.py by improving handling of referenced works and filling missing publication values with spaces.

Update project funding information in app.py for clarity and consistency.

rm working paper, added citation

commented out numba diagnostics, remove later.

Toggle use_medoids parameter in prediction function back to True and update datamapplot dependency to point to alternative version by MNoichl.

Toggle use_medoids parameter in prediction function to False for testing purposes.

added import for datamapplot.medoids and printed its help documentation

set Numba threads to 16 and print updated thread count

added numba threads-check

check numba config

set up caching dirs.

Refactor datamapplot version retrieval to use subprocess and update requirements to point to the main branch

force-bumped gradio added datamaplot version printout

Enhance data processing by adding CSV and Pickle file upload support, improving error handling, and refining the prediction pipeline. Update the `process_records_to_df` function to handle existing DataFrames and ensure required fields are processed correctly.

Add JavaScript function to enforce light theme by refreshing the page

Add retry mechanism and local publication year approximation to data processing

Update datamapplot dependency and minor UI text adjustments

linechange, test 2nd space push

Call is_running_in_hf_zero_gpu() function

added tier separation

adjusted_pointsize

seting ssr to false

Improve Uvicorn server configuration for local and HF Spaces

Enable SSR mode for Hugging Face Spaces

added note on Safari

fixing local run

run without ssr on local

ssr actually back on!

imports fixed

print token, ssr back on

disabled ssr

reintegrated old code

Increase GPU duration for prediction function

Update static file serving route in app.py

reintroduced uvicorn

adapted to aboklute path

cheger pder