I need help…
I download the latest live dataset through the api
napi.download_dataset(‘numerai_live_data_int8.parquet’, ‘numerai_live_data_int8.parquet’)
Then I make the prediction and upload the results.
Same IDs, same order.
Still I get this error message: Live prediction ids do not match. IDs must match current tournament data exactly, including ordering. Make sure you are using the latest tournament data.
Round 306 contains era1000 (the first 4-digit number), this breakes my legacy pipeline which ignores era1000, and this results in the same error message you got.