swiftformer-xs-dmae-va-U-80 / train_results.json
Augusto777's picture
End of training
c66ce13 verified
raw
history blame
210 Bytes
{
"epoch": 36.13,
"total_flos": 9.686412043576934e+16,
"train_loss": 0.8973502908434187,
"train_runtime": 243.9108,
"train_samples_per_second": 160.386,
"train_steps_per_second": 1.148
}