bert-base-greek-uncased-v1-finetuned-polylex

This model is a fine-tuned version of nlpaueb/bert-base-greek-uncased-v1 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.1624

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss
4.1637	1.0	12	2.6649
3.0581	2.0	24	2.5475
2.648	3.0	36	2.1624
2.5983	4.0	48	2.3285
2.7524	5.0	60	2.5745
2.4923	6.0	72	2.8096
2.5336	7.0	84	2.9470
2.3271	8.0	96	2.5497
2.4018	9.0	108	2.3413
2.544	10.0	120	2.4170
1.9144	11.0	132	2.5254
2.0996	12.0	144	2.4147
1.8733	13.0	156	2.5462
1.8261	14.0	168	2.2045
2.0033	15.0	180	1.9549
1.9967	16.0	192	2.1614
1.8515	17.0	204	2.8167
1.8583	18.0	216	2.8441
1.7512	19.0	228	2.4536
1.5746	20.0	240	2.6204
1.5267	21.0	252	2.9290
1.7248	22.0	264	2.0433
1.5692	23.0	276	2.4710
1.6093	24.0	288	2.4340
1.619	25.0	300	2.2689
1.4406	26.0	312	3.6729
1.5452	27.0	324	3.2225
1.4575	28.0	336	1.8853
1.5534	29.0	348	2.2135
1.4872	30.0	360	2.7540
1.3923	31.0	372	2.2408
1.3682	32.0	384	2.5181
1.2623	33.0	396	2.1360
1.1888	34.0	408	2.3912
1.3427	35.0	420	2.4600
1.1969	36.0	432	2.6388
1.3367	37.0	444	2.5489
1.226	38.0	456	1.5805
1.1808	39.0	468	2.7466
1.1694	40.0	480	2.4887
1.2736	41.0	492	2.5735
1.2292	42.0	504	2.2357
1.2556	43.0	516	2.9244
1.0155	44.0	528	1.8348
1.2425	45.0	540	2.4494
1.2665	46.0	552	2.4866
1.3439	47.0	564	2.3430
1.4468	48.0	576	1.7801
1.1772	49.0	588	2.5785
1.0618	50.0	600	2.9959

Framework versions

• Transformers 4.30.2
• Pytorch 2.0.1+cu118
• Datasets 2.13.1
• Tokenizers 0.13.3