21 lines
797 B
Plaintext
21 lines
797 B
Plaintext
[jev] Phase 1: env setup
|
|
[jev] laya.common OK
|
|
[jev] data OK (cached)
|
|
[jev] Phase 2: download mmBERT (ModelScope)
|
|
[jev] config.json: 0MB
|
|
[jev] tokenizer.json: 16MB
|
|
[jev] tokenizer_config.json: 0MB
|
|
[jev] special_tokens_map.json: 0MB
|
|
[jev] pytorch_model.bin: 1174MB
|
|
[jev] device=cuda tiny=False max_len=1024 head_len=160 win=800 aug_p=0.4
|
|
[jev] Phase 3: build DecisionModel
|
|
[jev] encoder load: ckpt=138 matched=134 unexpected=4 core_missing=0
|
|
[jev] model on cuda, trainable 17,267,203
|
|
[jev] train=20626 calib=5291
|
|
[jev] ep-1 pretrain calib {'auc': 0.4745, 'ece': 0.0399, 'n': 5291}
|
|
[jev] saved best (auc=0.9521)
|
|
[jev] TRAINING-DONE best_auc=0.9521
|
|
[jev] CALIBRATION T=1.35 raw(ece=0.0143) calibrated(ece=0.0101) auc=0.9937
|
|
[jev] Phase 5: push back
|
|
[jev] adapter params=17,267,206
|