Update README.md
Browse files
README.md
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
---
|
2 |
-
license:
|
3 |
viewer: false
|
4 |
---
|
5 |
# ChaosBench
|
@@ -13,35 +13,36 @@ We establish two tasks: full and sparse dynamics prediction.
|
|
13 |
๐: [https://github.com/leap-stc/](https://github.com/leap-stc/)
|
14 |
|
15 |
๐: [https://arxiv.org/](https://arxiv.org/)
|
16 |
-
|
17 |
-
## Table of Content
|
18 |
-
- [Getting Started](#getting-started)
|
19 |
-
- [Dataset Overview](#dataset-overview)
|
20 |
-
- [Evaluation Metrics](#evaluation-metrics)
|
21 |
-
- [Leaderboard](#leaderboard)
|
22 |
|
23 |
## Getting Started
|
24 |
-
1
|
25 |
-
- Clone the `ChaosBench` Github repository
|
26 |
-
- Create local directory to store your data, e.g., `data/`
|
27 |
-
- Navigate to `chaosbench/config.py` and change the field `DATA_DIR = <YOUR_WORKING_DATA_DIR>`
|
28 |
|
29 |
-
2
|
30 |
-
|
31 |
-
|
32 |
-
|
33 |
-
|
34 |
-
|
35 |
-
|
36 |
-
|
37 |
-
|
38 |
-
|
39 |
-
|
40 |
-
|
41 |
-
|
42 |
-
|
43 |
-
|
44 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
45 |
|
46 |
## Dataset Overview
|
47 |
|
@@ -72,6 +73,7 @@ We establish two tasks: full and sparse dynamics prediction.
|
|
72 |
- [x] ViT/ClimaX
|
73 |
- [x] PanguWeather
|
74 |
- [x] Fourcastnetv2
|
|
|
75 |
|
76 |
## Evaluation Metrics
|
77 |
We divide our metrics into 2 classes: (1) ML-based, which cover evaluation used in conventional computer vision and forecasting tasks, (2) Physics-based, which are aimed to construct a more physically-faithful and explainable data-driven forecast.
|
|
|
1 |
---
|
2 |
+
license: gpl-3.0
|
3 |
viewer: false
|
4 |
---
|
5 |
# ChaosBench
|
|
|
13 |
๐: [https://github.com/leap-stc/](https://github.com/leap-stc/)
|
14 |
|
15 |
๐: [https://arxiv.org/](https://arxiv.org/)
|
|
|
|
|
|
|
|
|
|
|
|
|
16 |
|
17 |
## Getting Started
|
18 |
+
**Step 1**: Clone the [ChaosBench](https://github.com/leap-stc/ChaosBench) Github repository
|
|
|
|
|
|
|
19 |
|
20 |
+
**Step 2**: Create local directory to store your data, e.g.,
|
21 |
+
```
|
22 |
+
cd ChaosBench
|
23 |
+
mkdir data
|
24 |
+
```
|
25 |
+
|
26 |
+
**Step 3**: Navigate to `chaosbench/config.py` and change the field `DATA_DIR = /<YOUR_WORKING_DIR>/ChaosBench/data` (_Provide absolute path_)
|
27 |
+
|
28 |
+
**Step 4**: Initialize the space by running
|
29 |
+
```
|
30 |
+
cd /<YOUR_WORKING_DIR>/ChaosBench/data/
|
31 |
+
wget https://huggingface.co/datasets/juannat7/ChaosBench/blob/main/process.sh
|
32 |
+
chmod +x process.sh
|
33 |
+
```
|
34 |
+
**Step 5**: Download the data
|
35 |
+
|
36 |
+
```
|
37 |
+
# NOTE: you can also run each line one at a time to retrieve individual dataset
|
38 |
+
|
39 |
+
./process.sh era5 # Required: For input ERA5 data
|
40 |
+
./process.sh climatology # Required: For climatology
|
41 |
+
./process.sh ukmo # Optional: For simulation from UKMO
|
42 |
+
./process.sh ncep # Optional: For simulation from NCEP
|
43 |
+
./process.sh cma # Optional: For simulation from CMA
|
44 |
+
./process.sh ecmwf # Optional: For simulation from ECMWF
|
45 |
+
```
|
46 |
|
47 |
## Dataset Overview
|
48 |
|
|
|
73 |
- [x] ViT/ClimaX
|
74 |
- [x] PanguWeather
|
75 |
- [x] Fourcastnetv2
|
76 |
+
- [x] GraphCast
|
77 |
|
78 |
## Evaluation Metrics
|
79 |
We divide our metrics into 2 classes: (1) ML-based, which cover evaluation used in conventional computer vision and forecasting tasks, (2) Physics-based, which are aimed to construct a more physically-faithful and explainable data-driven forecast.
|