metadata

library_name: transformers
license: apache-2.0
base_model: google-bert/bert-base-multilingual-cased
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: bert-base-multilingual-cased-finetuned-ner-harem
    results: []

bert-base-multilingual-cased-finetuned-ner-harem

This model is a fine-tuned version of google-bert/bert-base-multilingual-cased on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.2503
Precision: 0.7878
Recall: 0.8008
F1: 0.7942
Accuracy: 0.9658

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: 2
eval_batch_size: 2
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 16
optimizer: Use 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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	0.9978	281	0.2317	0.6004	0.6203	0.6102	0.9475
0.2657	1.9973	562	0.1557	0.7484	0.7282	0.7382	0.9629
0.2657	2.9969	843	0.1438	0.7812	0.7925	0.7868	0.9668
0.0848	3.9964	1124	0.1789	0.7429	0.7614	0.7520	0.9635
0.0848	4.9960	1405	0.2275	0.7769	0.8091	0.7927	0.9655
0.0378	5.9956	1686	0.1942	0.7694	0.8237	0.7956	0.9650
0.0378	6.9951	1967	0.2416	0.7753	0.8091	0.7919	0.9656
0.0192	7.9982	2249	0.2409	0.7831	0.7863	0.7847	0.9655
0.0084	8.9978	2530	0.2450	0.7809	0.7988	0.7897	0.9650
0.0084	9.9938	2810	0.2503	0.7878	0.8008	0.7942	0.9658

Framework versions

Transformers 4.46.3
Pytorch 2.5.1+cu121
Datasets 3.1.0
Tokenizers 0.20.3