-
|
Hello, Thanks for putting this dataset together. I was wondering if the checkpoints for the benchmarks are available somewhere? Appreciate it! |
Beta Was this translation helpful? Give feedback.
Replies: 5 comments
-
|
Hello, Thank you for your message! The checkpoints are currently not public, but if there is interest, we could try to provide some of them. Is there a specific checkpoint that you would need (dataset, model, learning rate)? Ruben |
Beta Was this translation helpful? Give feedback.
-
|
Thanks for responding, Providing the benchmark model weights would be really helpful for comparing methods against other metrics and timeframes not reported in the paper. The code is provided, but there’s quite a lot of benchmark models to train independently, 4 for each simulation. Appreciate it! |
Beta Was this translation helpful? Give feedback.
-
|
the checkpoints of the model for each dataset would be really helpful :) |
Beta Was this translation helpful? Give feedback.
-
|
Ok, let me see what we can do! |
Beta Was this translation helpful? Give feedback.
-
|
The checkpoints of the models benchmarked in the original paper have been uploaded to HF. See The Well Benchmark Model Collection. You can now load a specific checkpoint use the following example for the FNO model trained on from the_well.benchmark.models import FNO
model = FNO.from_pretrained("polymathic-ai/FNO-active_matter") |
Beta Was this translation helpful? Give feedback.
The checkpoints of the models benchmarked in the original paper have been uploaded to HF. See The Well Benchmark Model Collection.
You can now load a specific checkpoint use the following example for the FNO model trained on
active_matter