mrmr
mrmr copied to clipboard
mRMR (minimum-Redundancy-Maximum-Relevance) for automatic feature selection at scale.
Import fail for silicon-based Mac OS -- help?
There is a new syntax for computing Pearson correlations in Polars, the previous one errors for me. It appears that `pearson_corr` is no longer a function in Polars. https://docs.pola.rs/py-polars/html/reference/expressions/api/polars.corr.html
Running the example code Im receiving the following error: --------------------------------------------------------------------------- _RemoteTraceback Traceback (most recent call last) _RemoteTraceback: """ Traceback (most recent call last): File "/opt/conda/envs/feature_creation_env/lib/python3.12/site-packages/joblib/externals/loky/process_executor.py", line 426, in _process_worker call_item...
Hello, I would like to use mrmr_regression to perform some feature selection but when I run the following line : feature_select_MRMR = mrmr_regression(X=initial_data_set, y=target, K=10, n_jobs=1) Where my data and...
Is there any way that I can get the scores for the selected k features?
~\anaconda3\lib\site-packages\mrmr\main.py in mrmr_base(K, relevance_func, redundancy_func, relevance_args, redundancy_args, denominator_func, only_same_domain, return_scores, show_progress) 96 """ 97 ---> 98 relevance = relevance_func(**relevance_args) 99 features = relevance[relevance.fillna(0) > 0].index.to_list() 100 relevance = relevance.loc[features] ~\anaconda3\lib\site-packages\mrmr\pandas.py...
Looks like `polars` made a breaking change at some point and renamed it to `.corr` (https://docs.pola.rs/api/python/stable/reference/expressions/api/polars.corr.html#polars.corr)
Based on my user experience, `polars` seem to be as it is an required and not just an optional package. --- I base this on the following error message, ```...
I switched to direct imports, but this could be reverted... ### Do not import polars: An issue for M1 Macs is polars, which needs to use the arm instructions. Apparently...
First, great library and related blog posts. I was beginning to code this procedure and then stumbled upon your work. Here is my question / concern. I am using data...