evalml/config/forecasters-co1e.yaml at 91983fe5fe3bd46f6f8b9bd55cfe12492bbfaa8e · MeteoSwiss/evalml · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
# yaml-language-server: $schema=../workflow/tools/config.schema.json
description: |
  Experiment with COSMO-1E emulators finetuned on COSMO-1E analysis
  (KENDA-1) at 1km resolution.

dates:
  start: 2020-01-01T12:00
  end: 2020-01-10T00:00
  frequency: 54h

runs:
  - forecaster:
      checkpoint: https://mlflow.ecmwf.int/#/experiments/367/runs/2174c939c8844555a52843b71219d425
      label: Cosmo 1km + era5 N320, finetuned on cerra checkpoint, lam resolution 11
      config: resources/inference/configs/sgm-forecaster-regional_fromtraining.yaml
      steps: 0/120/6
      inference_resources:
        gpu: 4
        tasks: 4
      extra_requirements:
        - git+https://github.com/ecmwf/anemoi-inference.git@0.6.3

  - baseline:
      label: COSMO-1E
      root: /store_new/mch/msopr/ml/COSMO-1E
      steps: 0/33/6

truth:
  label: COSMO KENDA
  root: /scratch/mch/fzanetta/data/anemoi/datasets/mch-co1e-an-archive-0p01-2019-2024-1h-v1-pl13.zarr

stratification:
  regions:
    - jura
    - mittelland
    - voralpen
    - alpennordhang
    - innerealpentaeler
    - alpensuedseite
  root: /scratch/mch/bhendj/regions/Prognoseregionen_LV95_20220517

thresholds:
  TOT_PREC:
    gt: [0.0, 0.001, 0.005]
  U_10M:
    gt: [2.5, 5.0, 10.0]
  V_10M:
    gt: [2.5, 5.0, 10.0]
  T_2M:
    lt: [273.15]
    gt: [288.15, 298.15]

dashboard:
  stratification:
    # - init_hour
    # - region
    - season

locations:
  output_root: output/

profile:
  executor: slurm
  global_resources:
    gpus: 16
  default_resources:
    slurm_partition: "postproc"
    cpus_per_task: 1
    mem_mb_per_cpu: 1800
    runtime: "1h"
    gpus: 0
  jobs: 50
  batch_rules:
    plot_forecast_frame: 32

score_maps:
  params:
    - T_2M
    # - TD_2M
    # - U_10M
    # - V_10M
    # - SP_10M
    # - PS
    # - PMSL
    # - TOT_PREC
  leadtimes: [6, 24]
  # Or, to compute every available leadtime from runs+baselines:
  # leadtimes: "all"
  scores:
    - BIAS
    # - RMSE
    # - MAE
  regions:
    - switzerland
    # - centraleurope
  seasons:
    - all
    # - DJF
    # - MAM
    # - JJA
    # - SON
  init_hours:
    - all
    # - "00"
    # - "06"
    # - "12"
    # - "18"