You can deploy this model and start earning money today!

Mardiyyah/
CeLLaTe-tapt-pubmedbert-tokenizer-adapted-wwmask-combinedData

Name: CeLLaTe-tapt-pubmedbert-tokenizer-adapted-wwmask-combinedData
Availability: InStock

Views

No views yet

CeLLaTe-tapt-pubmedbert-tokenizer-adapted-wwmask-combinedData

This model is a fine-tuned version of microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract-fulltext on the Mardiyyah/Combined_TAPT_dataset 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: 3e-05
train_batch_size: 32
eval_batch_size: 32
seed: 3407
gradient_accumulation_steps: 2
total_train_batch_size: 64
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-06 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.06
num_epochs: 100
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Accuracy	Perplexity
1.4459	1.0	28	1.3302	0.7452	3.7816
1.3972	2.0	56	1.2823	0.7490	3.6048
1.3563	3.0	84	1.2437	0.7524	3.4685
1.3426	4.0	112	1.2253	0.7506	3.4050
1.3084	5.0	140	1.2217	0.7514	3.3930
1.281	6.0	168	1.2130	0.7539	3.3636
1.2606	7.0	196	1.2227	0.7525	3.3962
1.2452	8.0	224	1.2455	0.7485	3.4745
1.2291	9.0	252	1.2134	0.7532	3.3650
1.2136	10.0	280	1.1771	0.7565	3.2449
1.1945	11.0	308	1.1709	0.7544	3.2248
1.1856	12.0	336	1.2143	0.7526	3.3679
1.1779	13.0	364	1.1815	0.7524	3.2594
1.147	14.0	392	1.1803	0.7535	3.2553
1.1505	15.0	420	1.1819	0.7522	3.2607
1.1372	16.0	448	1.1592	0.7583	3.1875
1.1333	17.0	476	1.1801	0.7528	3.2546
1.1177	18.0	504	1.1837	0.7544	3.2666
1.1193	19.0	532	1.1513	0.7560	3.1623
1.1093	20.0	560	1.1969	0.7518	3.3099
1.0968	21.0	588	1.1824	0.7547	3.2623
1.0809	22.0	616	1.1841	0.7554	3.2676
1.0757	23.0	644	1.1764	0.7533	3.2426
1.0683	24.0	672	1.1730	0.7534	3.2318