xlm-base-CONTINUE-finetuned-ner-chim-v1

This model is a fine-tuned version of quanxuantruong/xlm-base-finetuned-ner-chim-v1 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7630
• Accuracy: 0.8230
• Precision: 0.3548
• Recall: 0.3874
• F1: 0.3628
• Date Precision: 0.1852
• Date Recall: 0.125
• Date F1-score: 0.1493
• Habitat Precision: 0.2424
• Habitat Recall: 0.1684
• Habitat F1-score: 0.1988
• Id Feature Precision: 0.3478
• Id Feature Recall: 0.2727
• Id Feature F1-score: 0.3057
• Location Precision: 0.6286
• Location Recall: 0.9296
• Location F1-score: 0.75
• Organization Precision: 0.0
• Organization Recall: 0.0
• Organization F1-score: 0.0
• Species Precision: 0.7246
• Species Recall: 0.8288
• Species F1-score: 0.7732
• Micro avg Precision: 0.5935
• Micro avg Recall: 0.5803
• Micro avg F1-score: 0.5868
• Macro avg Precision: 0.3548
• Macro avg Recall: 0.3874
• Macro avg F1-score: 0.3628
• Weighted avg Precision: 0.5012
• Weighted avg Recall: 0.5803
• Weighted avg F1-score: 0.5319

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: 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
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Date Precision	Date Recall	Date F1-score	Habitat Precision	Habitat Recall	Habitat F1-score	Id Feature Precision	Id Feature Recall	Id Feature F1-score	Location Precision	Location Recall	Location F1-score	Organization Precision	Organization Recall	Organization F1-score	Species Precision	Species Recall	Species F1-score	Micro avg Precision	Micro avg Recall	Micro avg F1-score	Macro avg Precision	Macro avg Recall	Macro avg F1-score	Weighted avg Precision	Weighted avg Recall	Weighted avg F1-score
No log	1.0	7	1.2386	0.6705	0.0263	0.0157	0.0196	0.0	0.0	0.0	0.0	0.0	0.0	0.1481	0.0909	0.1127	0.0	0.0	0.0	0.0	0.0	0.0	0.0094	0.0034	0.0050	0.0520	0.0125	0.0201	0.0263	0.0157	0.0196	0.0219	0.0125	0.0158
No log	2.0	14	1.0307	0.7273	0.1127	0.1145	0.1122	0.0	0.0	0.0	0.0	0.0	0.0	0.2857	0.2273	0.2532	0.3220	0.4014	0.3574	0.0	0.0	0.0	0.0683	0.0582	0.0628	0.1895	0.1302	0.1544	0.1127	0.1145	0.1122	0.1258	0.1302	0.1266
No log	3.0	21	0.8690	0.7781	0.2275	0.2553	0.2375	0.0	0.0	0.0	0.0	0.0	0.0	0.2875	0.2614	0.2738	0.6181	0.8662	0.7214	0.0	0.0	0.0	0.4591	0.4041	0.4299	0.4740	0.3657	0.4128	0.2275	0.2553	0.2375	0.3423	0.3657	0.3491
No log	4.0	28	0.7881	0.8158	0.2994	0.3409	0.3132	0.0435	0.025	0.0317	0.0909	0.0526	0.0667	0.3239	0.2614	0.2893	0.6161	0.9155	0.7365	0.0	0.0	0.0	0.7219	0.7911	0.7549	0.5735	0.5402	0.5563	0.2994	0.3409	0.3132	0.4670	0.5402	0.4960
No log	5.0	35	0.7630	0.8230	0.3548	0.3874	0.3628	0.1852	0.125	0.1493	0.2424	0.1684	0.1988	0.3478	0.2727	0.3057	0.6286	0.9296	0.75	0.0	0.0	0.0	0.7246	0.8288	0.7732	0.5935	0.5803	0.5868	0.3548	0.3874	0.3628	0.5012	0.5803	0.5319

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1