Closed gaozhihan closed 1 year ago
We need to also exchange the metric names in https://github.com/amazon-research/earth-forecasting-transformer/blob/61cab363c8c15d98ecd40643d628303d0234096b/scripts/cuboid_transformer/enso/train_cuboid_enso.py#L533-L534
valid_acc
--> valid_corr_nino3.4
valid_weighted_acc
--> valid_corr_nino3.4_weighted
.Also, change accordingly the test_acc / test_weighted_acc
We need to also exchange the metric names in
valid_acc
-->valid_corr_nino3.4
valid_weighted_acc
-->valid_corr_nino3.4_weighted
.
Done
Also, we need to add test cases...
Issue #5
Fix the cfg.yaml for training on ENSO.
total_batch_size: 32
tototal_batch_size: 64
to reproduce the reported results in our paper.total_batch_size: 32
results in suboptimal performance, whiletotal_batch_size: 64
works correctly.data_seq_len