thesujitroy
commited on
Commit
•
a8fb064
1
Parent(s):
bffd73a
Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,19 @@
|
|
1 |
-
---
|
2 |
-
license: cdla-permissive-2.0
|
3 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: cdla-permissive-2.0
|
3 |
+
---
|
4 |
+
|
5 |
+
Prithvi WxC is a 2.3 billion parameter model trained on 160 different variables from MERRA-2 data. It has been pretrained on both forecasting and masked
|
6 |
+
reconstruction objectives. I.e.~the model is capable of reconstructing atmospheric state from partial information as well as propagating state into the
|
7 |
+
future. The model takes data from two timestamps as input and generates a single, possibly future, timestamp as output. Currently Prithvi WxC comes in two flavors:
|
8 |
+
|
9 |
+
- (This model)`prithvi.wxc.rollout.2300m.v1` has been through further training cycles to be optimzed for autoregressive rollout. Here, we restricted the input delta
|
10 |
+
as well as the lead time to 6 hours. We recommend using `prithvi.wxc.rollout.2300m.v1` for forecasting applications.
|
11 |
+
|
12 |
+
<div style="display: flex; justify-content: center;">
|
13 |
+
|
14 |
+
<b> Zero-Shot Rollout</b>
|
15 |
+
<img src="https://huggingface.co/Prithvi-WxC/prithvi.wxc.rollout.2300m.v1/resolve/bffd73a5b80904a4a1c8637a9f3bb35d32ce3382/2021C4Ida_2021082700_2plots_winds_platecaree_nohurr.gif" alt="Gravity Wave" width="1024"/>
|
16 |
+
<br>
|
17 |
+
<img src="https://huggingface.co/Prithvi-WxC/prithvi.wxc.rollout.2300m.v1/resolve/bffd73a5b80904a4a1c8637a9f3bb35d32ce3382/2021C4Ida_2021082700_2plots_winds_platecaree_CONUS.gif" alt="Gravity Wave" width="1024"/>
|
18 |
+
|
19 |
+
</div>
|