Token Classification
GLiNER
PyTorch
gliner_multi_pii-v1 / gliner_config.json
urchade's picture
better version
de64314 verified
raw
history blame
477 Bytes
{
"size_sup": -1,
"max_types": 25,
"shuffle_types": true,
"random_drop": true,
"max_neg_type_ratio": 1,
"max_len": 384,
"lr_encoder": "1e-6",
"lr_others": "1e-5",
"num_steps": 10000,
"warmup_ratio": 1000,
"train_batch_size": 8,
"eval_every": 2500,
"max_width": 12,
"model_name": "microsoft/mdeberta-v3-base",
"fine_tune": true,
"subtoken_pooling": "first",
"hidden_size": 512,
"span_mode": "markerV0",
"dropout": 0.4,
"name": "correct"
}