README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
model-index:
  - name: results
    results: []

results

This model is a fine-tuned version of AfnanTS/ARBERTv2_EntityLinking on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6434

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
0.7546	0.1	100	1.0176
0.7851	0.21	200	0.9869
0.7382	0.31	300	0.9651
0.7615	0.41	400	0.9340
0.8167	0.51	500	0.8955
0.7621	0.62	600	0.8642
0.8508	0.72	700	0.8372
0.838	0.82	800	0.8090
0.7961	0.92	900	0.7915
0.749	1.03	1000	0.7682
0.5584	1.13	1100	0.7613
0.5282	1.23	1200	0.7449
0.5195	1.33	1300	0.7373
0.4969	1.44	1400	0.7220
0.4977	1.54	1500	0.7113
0.4825	1.64	1600	0.6992
0.4938	1.74	1700	0.6922
0.4672	1.85	1800	0.6875
0.4545	1.95	1900	0.6831
0.4034	2.05	2000	0.6712
0.3963	2.15	2100	0.6661
0.4115	2.26	2200	0.6635
0.3953	2.36	2300	0.6593
0.3648	2.46	2400	0.6561
0.3987	2.56	2500	0.6509
0.3677	2.67	2600	0.6489
0.3741	2.77	2700	0.6480
0.3355	2.87	2800	0.6445
0.3603	2.97	2900	0.6434

Framework versions

• Transformers 4.27.1
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.13.3