You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello, I see in your paper that for datasets like MultiArith/SVAMP, you randomly sampled 500 data points to serve as a validation set, with the rest as the test set. Have you made this split validation and test set public? Or the corresponding index files? I only found the val_index.npy for gsm8k, and it only sampled 200 data points from the training set, which is not quite consistent with what you mentioned in the paper about "sampling 500 data points from the test set to serve as the validation set"?
The text was updated successfully, but these errors were encountered:
Hello, I see in your paper that for datasets like MultiArith/SVAMP, you randomly sampled 500 data points to serve as a validation set, with the rest as the test set. Have you made this split validation and test set public? Or the corresponding index files? I only found the val_index.npy for gsm8k, and it only sampled 200 data points from the training set, which is not quite consistent with what you mentioned in the paper about "sampling 500 data points from the test set to serve as the validation set"?
The text was updated successfully, but these errors were encountered: