31 lines
1.4 KiB
Plaintext
31 lines
1.4 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=37182 calib=5291
|
|
[jev] ep0 step100/2322 loss=0.2847
|
|
[jev] ep0 step200/2322 loss=0.1989
|
|
[jev] ep0 step300/2322 loss=0.2504
|
|
[jev] ep0 step400/2322 loss=0.3098
|
|
[jev] ep0 step500/2322 loss=0.1724
|
|
[jev] ep0 step600/2322 loss=0.2330
|
|
[jev] ep0 step700/2322 loss=0.2886
|
|
[jev] ep0 step800/2322 loss=0.2370
|
|
[jev] ep0 step900/2322 loss=0.3038
|
|
[jev] ep0 step1000/2322 loss=0.2497
|
|
[jev] ep0 step1100/2322 loss=0.1760
|
|
[jev] epoch 0 calibration: {'auc': 0.9926, 'ece': 0.0072, 'n': 5291}
|
|
[jev] live-checkpoint 已写 /tmp/jd/model_live
|
|
[jev] ep1 step1200/2322 loss=0.2885
|
|
[jev] ep1 step1300/2322 loss=0.3465
|
|
[jev] ep1 step1400/2322 loss=0.2921
|
|
[jev] ep1 step1500/2322 loss=0.1571
|
|
[jev] ep1 step1600/2322 loss=0.1510
|
|
[jev] ep1 step1700/2322 loss=0.1976
|
|
[jev] ep1 step1800/2322 loss=0.0938
|
|
[jev] ep1 step1900/2322 loss=0.1129
|
|
[jev] ep1 step2000/2322 loss=0.1629
|
|
[jev] ep1 step2100/2322 loss=0.4172
|
|
[jev] ep1 step2200/2322 loss=0.2923
|
|
[jev] ep1 step2300/2322 loss=0.2418
|