-
- Downloads
- Unignore results;
- Even if hyperparameters file is ignore with skip_best_hyperparams option, still use the same forest_size to be comparable; - Update experiment files for stage1 wo_param experiments (using the same forest size as the with_params experiments); - In compute_results: remove useless folder creation; temporary add extracted_forest_sizes_number option to specify the extracted forest sizes number; temporary not plotting train and dev losses in stage1 loss values figure; - In plotter, clean-up stage1 figure generation; - Add first unbiased losses plot (stage1: best params vs default params in california housing dataset).
parent
76a58da9
No related branches found
No related tags found
This commit is part of merge request !9. Comments created here will be created in the context of that merge request.
Showing
- .gitignore 0 additions, 4 deletions.gitignore
- code/train.py 7 additions, 3 deletionscode/train.py
- experiments/california_housing/stage1/none_wo_params_4.json 5 additions, 5 deletionsexperiments/california_housing/stage1/none_wo_params_4.json
- experiments/california_housing/stage1/omp_wo_params_6.json 5 additions, 5 deletionsexperiments/california_housing/stage1/omp_wo_params_6.json
- experiments/california_housing/stage1/random_wo_params_5.json 5 additions, 5 deletions...riments/california_housing/stage1/random_wo_params_5.json
Please register or sign in to comment