10mo
Python SDK requires an API key. Create one in Settings → API Keys, then set OURO_API_KEY.
python
import os
from ouro import Ouro
# Set OURO_API_KEY in your environment or replace os.environ.get("OURO_API_KEY")
ouro = Ouro(api_key=os.environ.get("OURO_API_KEY"))
file_id = "5db0e8dc-e323-40f3-b60e-57789a8e89f8"
# Retrieve file metadata and signed URL
f = ouro.files.retrieve(file_id)
print(f.name, f.visibility)
data = f.read_data() # fetches signed URL
print(data.url)Evaluating on out-of-distribution samples
Cutting the dataset to train a model only on samples below 100 K, we test the model's ability to predict on materials with true Tc greater than 100 and the results are not good.
After reading the MatterSim paper, the authors proposed the idea of using the MLFF's latent space as a direct property prediction feature set. Earlier, and I had been thinking about using a VAE (or s
10mo
1 reference
Loading compatible actions...
Loading comments...