Instructions to use Zipeng365/WISP with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Scikit-learn
How to use Zipeng365/WISP with Scikit-learn:
from huggingface_hub import hf_hub_download import joblib model = joblib.load( hf_hub_download("Zipeng365/WISP", "sklearn_model.joblib") ) # only load pickle files from sources you trust # read more about it here https://skops.readthedocs.io/en/stable/persistence.html - Notebooks
- Google Colab
- Kaggle
Download docs/figures/cost_accuracy_plotted_values.csv from Zipeng365/WISP: direct link, hf CLI and curl.
- Browser
- Download file 2.73 kB
-
https://huggingface.co/Zipeng365/WISP/resolve/main/docs/figures/cost_accuracy_plotted_values.csv
- Command line
-
hf download hf://Zipeng365/WISP/docs/figures/cost_accuracy_plotted_values.csv
-
curl -L -o cost_accuracy_plotted_values.csv https://huggingface.co/Zipeng365/WISP/resolve/main/docs/figures/cost_accuracy_plotted_values.csv
2.73 kB
| key,name,group,final_model_core_h,final_model_vs_wisp_cis,macro_f1,accuracy,balanced_accuracy,worst_class_f1,accuracy_scope,hardware_note | |
| wisp_random,WISP-Random,wisp_search,0.311,0.8963,70.1996,77.2824,70.875,28.2521,3 seeds,"Xeon 8570, 1 thread (final model of the 16-thread search run)" | |
| wisp_evolution,WISP-Evolution,wisp_search,0.712,2.0519,69.5918,76.8443,70.1693,27.9627,3 seeds,"Xeon 8570, 1 thread (final model of the 16-thread search run)" | |
| wisp_cis,WISP-CIS,wisp_fixed,0.347,1.0,66.1391,74.0655,66.0421,24.7785,3 seeds,"Xeon 8570, 1 thread" | |
| rdst_tuned,RDST (tuned),cpu,34.15,98.415,67.07,74.93,68.21,25.64,3 seeds,"mixed BlueBEAR CPUs (Xeon 8360Y / 8480CL / 8570), 8-thread timing reruns of the final refit" | |
| mr_hydra_tuned,MR-Hydra (tuned),cpu,45.52,131.1816,66.57,74.75,67.11,27.82,3 seeds,"mixed BlueBEAR CPUs, final refit from 8-/1-thread timing reruns" | |
| multirocket_tuned,MultiRocket (tuned),cpu,84.63,243.8905,66.52,74.45,66.06,26.95,3 seeds,"mixed BlueBEAR CPUs, final refit from 8-/1-thread timing reruns" | |
| minirocket_tuned,MiniRocket (tuned),cpu,15.53,44.755,64.03,73.59,64.08,22.73,3 seeds,"mixed BlueBEAR CPUs, final refit from 8-/1-thread timing reruns" | |
| aeon15_rdst_default_v1,RDST,cpu,33.15,95.5331,60.0974,70.1516,60.5947,18.0501,3 seeds,"mixed BlueBEAR CPUs, 8 threads" | |
| aeon15_mr_hydra_default_v1,MR-Hydra,cpu,91.01,262.2767,59.8006,68.4927,60.4243,19.4399,3 seeds,"mixed BlueBEAR CPUs, 1/32/8 threads" | |
| aeon15_multirocket_default_v1,MultiRocket†,cpu,74.55,214.8415,59.0983,67.8303,59.7245,19.12,3 seeds,"mixed BlueBEAR CPUs, 22/32/8 threads; WISDM fold-0 refit on the workstation (AMD Threadripper PRO 9965WX), wall time x 22 threads, which overstates its CPU time" | |
| aeon15_quant_default_v1,QUANT,cpu,1.74,5.0144,55.2813,66.1314,55.8404,15.8511,3 seeds,"mixed BlueBEAR CPUs, 4 threads" | |
| minirocket_official_v1,MiniRocket,cpu,2.54,7.3199,59.0362,69.3873,59.7813,15.4518,3 seeds,"mixed BlueBEAR CPUs, 4 threads" | |
| simple_stats_rf_balanced_v1,Balanced-RF,cpu,0.22,0.634,47.8504,60.3891,48.6058,9.0427,3 seeds,"mixed BlueBEAR CPUs, 4 threads" | |
| hc2,HC2,cpu,448.0,1291.0663,61.6773,72.1287,61.9067,18.0373,3 seeds; 120-min contract,"Xeon 8360Y / 8480CL, 32 threads" | |
| harnet_ft_refit,HARNet (fine-tuned),deep,16.89,48.6744,54.78,61.69,59.36,21.3,3 seeds,1 x A100 (GPU wall x 64) + 18 host cores | |
| tf_c_3ch,TF-C (3 channels),deep,92.78,267.3775,47.21,55.21,51.14,15.31,3 seeds,1 x A100 (GPU wall x 64) + 18 host cores | |
| attend_discriminate_official_v1,Attend-and-Discriminate,deep,489.39,1410.3458,56.8609,64.574,62.9973,23.4789,3 seeds,"1 x A100 (GPU wall x 64), one job at a time" | |
| h_inceptiontime_aeon_v1,H-InceptionTime,deep,102.36,294.9856,57.6317,66.7665,59.8324,19.128,3 seeds,"1 x A100 (GPU wall x 64), one job at a time" | |