File size: 2,623 Bytes
b72b9a2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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

data:
  seq_len: 12
  pred_horz: 12
  batch_size: 128
  shuffle: False
  dataSplit: chronologicalSplit        # Chronological or Stratified Refer to  ~/Projects/PhD/Ionosphere/Python/Data Scripts/Preprocessing/Preprocess_Data_v3.ipynb
  selectedOMNI: [1,4,5,6,9,10,11,12,13,14,16,17,19,27,28,29,30] # Check data.py
  # ['F10.7', 'Dst', 'ap', 'AE', 'pc', 'SW Temp', 'SW Density', 'SW Speed', 'SW Long Angle', 'SW Lat Angle', 'Flow Pressure', 'E Field', 'Scalar B', 'BZ (GSM)', 'Year', 'Day of Year', 'Hour']
  # [  1,        4,      5,    6,   9,    10,           11,           12,         13,                 14,              16,          17,         19,        27,        28,      29,          30]

model:
  type: "SimVP"
  input_shape: [12, 18, 72, 72]  # (Time, Channels, Height, Width)
  hid_S: 16 
  hid_T: 256
  N_S: 4
  N_T: 4
  model_type: "gSTA"            
  mlp_ratio: 8.0
  drop: 0.0
  drop_path: 0.0
  spatio_kernel_enc: 3
  spatio_kernel_dec: 3
  conv_in_channels: 18
  conv_out_channels: 1
  conv_kernel_size: [1, 1, 1]
  conv_stride: [1, 1, 1]


train:
  model: "SimVP"                    # Model to train
  sessionName: "SimVP_chronologicalSplit"   # Name of the session
  lr: 8e-5                          # Learning rate for Optimizer (LRrangetest minima at [LR=0.0001] Global average loss = 5.77232401785838e-06)
  lrEnd: 5e-6                       # Learning rate end for ExponentialLR
  batchSize: 128                     # Batch size for training
  numEpochs: 50                     # Number of epochs to train
  resumeEpoch: 0                    # Resume training from epoch
  earlyStop: 10                     # Early stopping patience
  saveResults: True                 # Save results
  inputNames: []                    # Names of the input features
  criterion: "MSELoss"              # Loss functions: MSELoss.
  optimizer: "Adam"                 # Optimizer: Adam.
  scheduler: "ReduceLROnPlateau"    # Scheduler: ExponentialLR, ReduceLROnPlateau
  mixedPrecision: False             # Mixed precision training

test:
  model: "SimVP"
  sessionName: "SimVP_chronologicalSplit"     # Name of the session
  batchSize: 128
  testDates: []                       # Dates to test will be set in the script
  maxTEC: "210"                   # Set inside the main script
  minTEC: "0"                     # Set inside the main script
  lat: "72"                      # Set inside the main script
  lon: "72"                      # Set inside the main script
  numOMNI: "18"                  # Set inside the main script
  inputNames: []                # Set inside the main script
  saveResults: True