update readme

smartin98 · Jan 19, 2024 · b19e041 · b19e041
1 parent 7fea9aa
commit b19e041
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 2 deletions.
diff --git a/.ipynb_checkpoints/README-checkpoint.md b/.ipynb_checkpoints/README-checkpoint.md
@@ -20,7 +20,7 @@ Steps to reproduce SSH mapping workflow:
 
 We also provide checkpoint files for the models used in the paper, in which case steps 4-5 can be skipped. These checkpoint files were too large for GitHub but are stored in a Harvard Dataverse [repo](https://doi.org/10.7910/DVN/H4HQGD) along with the SSH maps.
 
-Also provided in the Dataverse repo is a small sample of pre-processed input files for predicting global maps for 1 day (2019-01-01) using all satellites apart from SARAL/Altika (in line with the data challenge setup), this allows steps 1-6 to be skipped to run inference with a trained network, using the script demo.py. 
+Also provided in the Dataverse repo is a set of pre-processed input files for predicting global maps for 2019 using all satellites apart from SARAL/Altika (in line with the data challenge setup), this allows steps 1-6 to be skipped to run inference with a trained network. 
 
 Minor adaptations to simvp_ddp_training.py would allow any PyTorch model that takes the right input/output dimensions to be used instead. 
 

diff --git a/README.md b/README.md
@@ -20,7 +20,7 @@ Steps to reproduce SSH mapping workflow:
 
 We also provide checkpoint files for the models used in the paper, in which case steps 4-5 can be skipped. These checkpoint files were too large for GitHub but are stored in a Harvard Dataverse [repo](https://doi.org/10.7910/DVN/H4HQGD) along with the SSH maps.
 
-Also provided in the Dataverse repo is a small sample of pre-processed input files for predicting global maps for 1 day (2019-01-01) using all satellites apart from SARAL/Altika (in line with the data challenge setup), this allows steps 1-6 to be skipped to run inference with a trained network, using the script demo.py. 
+Also provided in the Dataverse repo is a set of pre-processed input files for predicting global maps for 2019 using all satellites apart from SARAL/Altika (in line with the data challenge setup), this allows steps 1-6 to be skipped to run inference with a trained network. 
 
 Minor adaptations to simvp_ddp_training.py would allow any PyTorch model that takes the right input/output dimensions to be used instead.