Update training/test/val splits based off #28 critique

The correlation between days of the week might mean that there is leakage between the training and val/test sets. This is now split up so that 2020 is used for training, and 2021 is used for val/testing. This does mean that there is only data for the first half of 2021 in the test/val set for now, but that does cover winter/spring/part of summer so far, with more as time goes on.
openclimatefix · Jun 30, 2021 · 85f7cc6 · 85f7cc6
1 parent 425dded
commit 85f7cc6
Show file tree

Hide file tree

Showing 4 changed files with 19 additions and 19 deletions.
diff --git a/satflow/configs/datamodule/metnet_datamodule.yaml b/satflow/configs/datamodule/metnet_datamodule.yaml
@@ -1,22 +1,22 @@
 # @package _group_
 _target_: satflow.data.datamodules.SatFlowDataModule
 
-batch_size: 4
+batch_size: 1
 data_dir: ${data_dir} # data_dir is specified in config.yaml
 shuffle: 0
 sources:
-  train: "satflow-flow-144-tiled-{00001..00149..2}.tar"
-  val: "satflow-flow-144-tiled-{00002..00149..4}.tar"
-  test: "satflow-flow-144-tiled-{00004..00149..4}.tar"
-num_workers: 4
+  train: "satflow-flow-144-tiled-{00001..00105}.tar"
+  val: "satflow-flow-144-tiled-{00106..00129}.tar"
+  test: "satflow-flow-144-tiled-{00130..00149}.tar"
+num_workers: 12
 pin_memory: True
 config:
   visualize: False
-  num_timesteps: 12
-  skip_timesteps: 1
-  forecast_times: 48
-  output_shape: 256
-  output_target: 64
+  num_timesteps: 6
+  skip_timesteps: 3
+  forecast_times: 24
+  output_shape: 64
+  output_target: 16
   target_type: "cloudmask"
   num_crops: 10
   use_topo: True

diff --git a/satflow/configs/datamodule/satflow_dataloaders.yaml b/satflow/configs/datamodule/satflow_dataloaders.yaml
@@ -18,9 +18,9 @@ pin_memory: True
 time_as_channels: False
 data_dir: ${data_dir} # data_dir is specified in config.yaml
 sources:
-  train: "satflow-flow-36-{00001..00105..2}.tar"
-  val: "satflow-flow-36-{00002..00105..4}.tar"
-  test: "satflow-flow-36-{000106..00150}.tar"
+  train: "satflow-flow-144-tiled-{00001..00105}.tar"
+  val: "satflow-flow-144-tiled-{00106..00129}.tar"
+  test: "satflow-flow-144-tiled-{00130..00149}.tar"
 bands:
   [
     "HRV",

diff --git a/satflow/configs/datamodule/satflow_datamodule.yaml b/satflow/configs/datamodule/satflow_datamodule.yaml
@@ -5,9 +5,9 @@ batch_size: 1
 data_dir: ${data_dir} # data_dir is specified in config.yaml
 shuffle: 0
 sources:
-  train: "satflow-flow-144-tiled-{00001..00149..2}.tar"
-  val: "satflow-flow-144-tiled-{00002..00149..4}.tar"
-  test: "satflow-flow-144-tiled-{000014..00149..4}.tar"
+  train: "satflow-flow-144-tiled-{00001..00105}.tar"
+  val: "satflow-flow-144-tiled-{00106..00129}.tar"
+  test: "satflow-flow-144-tiled-{00130..00149}.tar"
 num_workers: 1
 pin_memory: True
 config:

diff --git a/satflow/configs/datamodule/unet_dataloaders.yaml b/satflow/configs/datamodule/unet_dataloaders.yaml
@@ -17,9 +17,9 @@ use_image: False
 pin_memory: True
 time_as_channels: True
 sources:
-  train: "/run/media/bieker/data/EUMETSAT/2020/satflow-flow-36-{00001..00105..2}.tar"
-  val: "/run/media/bieker/data/EUMETSAT/2020/satflow-flow-36-{00002..00105..4}.tar"
-  test: "/run/media/bieker/data/EUMETSAT/2021/satflow-flow-36-{00106..00150}.tar"
+  train: "satflow-flow-144-tiled-{00001..00105}.tar"
+  val: "satflow-flow-144-tiled-{00106..00129}.tar"
+  test: "satflow-flow-144-tiled-{00130..00149}.tar"
 bands:
   [
     "HRV",