Run on NeSI, 100 iterations

Model pickle files and predictions.

Results from snakemake -j 1 -p --config USE_SLURM=True, running on https://jupyter.nesi.org.nz.

50 iter

Output of snakemake -j 16

Building DAG of jobs...
Using shell: /usr/bin/bash
Provided cores: 16
Rules claiming more threads will be scaled down.
Job counts:
        count   jobs
        1       all
        1       evaluate
        3       fit
        1       fit_gbdt
        1       fit_knn
        1       fit_radius
        6       predict
        1       prepare_data
        15

[Mon Dec  7 16:58:20 2020]
rule prepare_data:
    input: data/cas_dataset.csv
    output: results/cas_dataset.csv
    jobid: 1

[Mon Dec  7 16:58:29 2020]
Finished job 1.
1 of 15 steps (7%) done

[Mon Dec  7 16:58:29 2020]
rule fit:
    input: results/cas_dataset.csv
    output: results/dummy_model/model.pickle
    jobid: 4
    wildcards: model_type=dummy
    threads: 8


[Mon Dec  7 16:58:29 2020]
rule fit_gbdt:
    input: results/cas_dataset.csv
    output: results/gbdt_model/model.pickle
    jobid: 14
    threads: 8

...

[Mon Dec  7 21:43:56 2020]
Finished job 15.
14 of 15 steps (93%) done

[Mon Dec  7 21:43:56 2020]
localrule all:
    input: results/cas_dataset.csv, results/dummy_model/predictions.csv, results/linear_model/predictions.csv, results/mlp_model/predictions.csv, results/knn_model/predictions.csv, results/radius_model/predictions.csv, results/gbdt_model/predictions.csv, results/summary
    jobid: 0

[Mon Dec  7 21:43:56 2020]
Finished job 0.
15 of 15 steps (100%) done

Worse results?

First release

Model pickle files and predictions. Results from snakemake -j 1 --config N_ITER=1