albert-base-spanish-finetuned-ner-finetuned-ner

This model is a fine-tuned version of dccuchile/albert-base-spanish-finetuned-ner on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3012
• Precision: 0.8356
• Recall: 0.8356
• F1: 0.8356
• Accuracy: 0.9385

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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	13	1.8849	0.0	0.0	0.0	0.5939
No log	2.0	26	1.4600	0.0	0.0	0.0	0.6687
No log	3.0	39	1.1449	0.0	0.0	0.0	0.6832
No log	4.0	52	0.9138	0.2857	0.2329	0.2566	0.8056
No log	5.0	65	0.7441	0.4504	0.4041	0.4260	0.8399
No log	6.0	78	0.6292	0.5310	0.5274	0.5292	0.875
No log	7.0	91	0.5406	0.6786	0.6507	0.6643	0.9041
No log	8.0	104	0.4747	0.7397	0.7397	0.7397	0.9259
No log	9.0	117	0.4228	0.7945	0.7945	0.7945	0.9306
No log	10.0	130	0.3900	0.8333	0.8219	0.8276	0.9332
No log	11.0	143	0.3685	0.8392	0.8219	0.8304	0.9339
No log	12.0	156	0.3487	0.8333	0.8219	0.8276	0.9339
No log	13.0	169	0.3325	0.8219	0.8219	0.8219	0.9339
No log	14.0	182	0.3227	0.8472	0.8356	0.8414	0.9339
No log	15.0	195	0.3150	0.8531	0.8356	0.8443	0.9358
No log	16.0	208	0.3094	0.8345	0.8288	0.8316	0.9358
No log	17.0	221	0.3047	0.8414	0.8356	0.8385	0.9378
No log	18.0	234	0.3027	0.8356	0.8356	0.8356	0.9385
No log	19.0	247	0.3017	0.8414	0.8356	0.8385	0.9385
No log	20.0	260	0.3012	0.8356	0.8356	0.8356	0.9385

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu118
• Datasets 2.21.0
• Tokenizers 0.19.1