You can deploy this model and start earning money today!

Mardiyyah/
CeLLaTe-ner-3class-bioformer16l-tapt-tokenizer-adapted-baseline

Views

No views yet

CeLLaTe-ner-3class-bioformer16l-tapt-tokenizer-adapted-baseline

This model is a fine-tuned version of Mardiyyah/CeLLaTe-tapt-bioformer16l-tokenizer-adapted-baseline on the OTAR3088/CeLLaTe-ner-3class-iob_final dataset. It achieves the following results on the evaluation set:

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 2e-05
train_batch_size: 16
eval_batch_size: 16
seed: 3407
gradient_accumulation_steps: 2
total_train_batch_size: 32
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.01
num_epochs: 20
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	Micro F1	Weighted F1	Macro F1	Accuracy
0.4142	1.0	263	0.1077	0.4542	0.5118	0.4813	0.4745	0.4441	0.9666
0.0758	2.0	526	0.0784	0.7688	0.6721	0.7172	0.7131	0.6973	0.9786
0.0458	3.0	789	0.0742	0.7518	0.7051	0.7277	0.7267	0.7179	0.9788
0.0344	4.0	1052	0.0805	0.7777	0.7222	0.7489	0.7462	0.7362	0.9795
0.0269	5.0	1315	0.0787	0.7134	0.7408	0.7268	0.7282	0.7201	0.9777
0.0213	6.0	1578	0.0857	0.7184	0.7109	0.7146	0.7173	0.7122	0.9781
0.0169	7.0	1841	0.0903	0.7797	0.7210	0.7492	0.7482	0.7376	0.9799
0.0143	8.0	2104	0.0978	0.7694	0.7055	0.7360	0.7363	0.7286	0.9792
0.0121	9.0	2367	0.0963	0.7997	0.7392	0.7683	0.7672	0.7552	0.9810
0.0102	10.0	2630	0.1039	0.7586	0.7194	0.7385	0.7383	0.7302	0.9793
0.009	11.0	2893	0.1008	0.7654	0.7544	0.7598	0.7591	0.7491	0.9802
0.0079	12.0	3156	0.1074	0.7959	0.7322	0.7627	0.7623	0.7533	0.9808
0.0067	13.0	3419	0.1101	0.7630	0.7210	0.7414	0.7412	0.7346	0.9795
0.0063	14.0	3682	0.1086	0.7918	0.7377	0.7638	0.7630	0.7536	0.9805