33 lines
1.5 KiB
Plaintext
33 lines
1.5 KiB
Plaintext
[jev] device=cuda base=/tmp/base
|
|
Some weights of XLMRobertaForSequenceClassification were not initialized from the model checkpoint at /tmp/base and are newly initialized: ['classifier.dense.bias', 'classifier.dense.weight', 'classifier.out_proj.bias', 'classifier.out_proj.weight']
|
|
You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.
|
|
trainable params: 1,476,097 || all params: 279,520,514 || trainable%: 0.5281
|
|
[jev] train=35848 calib=5291
|
|
[jev] ep0 step100/2240 loss=0.3442
|
|
[jev] ep0 step200/2240 loss=0.1200
|
|
[jev] ep0 step300/2240 loss=0.3088
|
|
[jev] ep0 step400/2240 loss=0.3102
|
|
[jev] ep0 step500/2240 loss=0.0746
|
|
[jev] ep0 step600/2240 loss=0.1947
|
|
[jev] ep0 step700/2240 loss=0.2139
|
|
[jev] ep0 step800/2240 loss=0.2837
|
|
[jev] ep0 step900/2240 loss=0.1628
|
|
[jev] ep0 step1000/2240 loss=0.1508
|
|
[jev] ep0 step1100/2240 loss=0.2874
|
|
[jev] epoch 0 calibration: {'auc': 0.9936, 'ece': 0.0118, 'n': 5291}
|
|
[jev] live-checkpoint 已写 /tmp/jd/model_live
|
|
[jev] ep1 step1200/2240 loss=0.5534
|
|
[jev] ep1 step1300/2240 loss=0.2997
|
|
[jev] ep1 step1400/2240 loss=0.2127
|
|
[jev] ep1 step1500/2240 loss=0.1171
|
|
[jev] ep1 step1600/2240 loss=0.2262
|
|
[jev] ep1 step1700/2240 loss=0.0793
|
|
[jev] ep1 step1800/2240 loss=0.1416
|
|
[jev] ep1 step1900/2240 loss=0.2552
|
|
[jev] ep1 step2000/2240 loss=0.0808
|
|
[jev] ep1 step2100/2240 loss=0.1460
|
|
[jev] ep1 step2200/2240 loss=0.0818
|
|
[jev] epoch 1 calibration: {'auc': 0.9939, 'ece': 0.0138, 'n': 5291}
|
|
[jev] live-checkpoint 已写 /tmp/jd/model_live
|
|
[jev] DONE best={'auc': 0.9939, 'ece': 0.0138, 'epoch': 1}
|