Evaluasi: Metrik dan plot di DVC

Pengantar Versioning Data dengan DVC

Ravi Bhadauria

Machine Learning Engineer

Metrik: perubahan di dvc.yaml

Konfigurasi file YAML DVC untuk melacak metrik antar eksperimen
Ubah dari outs

stages:
  train_and_evaluate:
    outs:
    - metrics.json
    - plots.png

$$ $$

Ke metrics

stages:
  train_and_evaluate:
    outs:
    - plots.png
    metrics:
      - metrics.json:
        cache: false

Mencetak metrik DVC

$ dvc metrics show

Path          accuracy    f1_score    precision    recall             
metrics.json  0.947       0.8656      0.988        0.7702

Bandingkan metrik antar run

Ubah satu hiperparameter dan jalankan ulang dvc repro

$ dvc metrics diff

Path          Metric     HEAD    workspace    Change                  
metrics.json  accuracy   0.947   0.9995       0.0525
metrics.json  f1_score   0.8656  0.9989       0.1333
metrics.json  precision  0.988   0.9993       0.0113
metrics.json  recall     0.7702  0.9986       0.2284

Plot: perubahan di dvc.yaml

stages:
  train_and_evaluate:
    ...
    plots:
    - predictions.csv: # Nama file berisi prediksi
        template: confusion # Gaya plot

        x: predicted_label # Nama kolom sumbu X di file csv 
        y: true_label # Nama kolom sumbu Y di file csv
        x_label: 'Predicted label'
        y_label: 'True label'
        title: Confusion matrix

        cache: false # Simpan di Git

¹ https://dvc.org/doc/user-guide/experiment-management/visualizing-plots#plot-templates-data-series-only

Mencetak plot DVC ke file

$ dvc plots show predictions.csv

file:///path/to/index.html

Plot tampilan confusion matrix yang dibuat DVC

Membandingkan plot DVC

# bandingkan plot di predictions.csv dengan branch main
$ dvc plots diff --target predictions.csv <branch name or commit SHA>

Plot perbandingan confusion matrix yang dibuat DVC

Ayo berlatih!

Pengantar Versioning Data dengan DVC