camembert-base-articles-ner-v4

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

• Loss: 0.6720
• F1: 0.6484

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: 5e-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	F1
2.0241	1.0	5	1.7392	0.0
1.706	2.0	10	1.5603	0.0
1.5411	3.0	15	1.5006	0.0
1.4327	4.0	20	1.4519	0.0
1.4006	5.0	25	1.3972	0.0
1.3422	6.0	30	1.3435	0.0
1.329	7.0	35	1.2744	0.0979
1.1985	8.0	40	1.2313	0.0982
1.1433	9.0	45	1.1661	0.1798
1.14	10.0	50	1.1327	0.1379
1.0554	11.0	55	1.1010	0.1310
0.9679	12.0	60	1.0609	0.0769
0.9281	13.0	65	1.0207	0.1520
0.8697	14.0	70	0.9726	0.1743
0.8641	15.0	75	0.9455	0.4355
0.7648	16.0	80	0.8995	0.5138
0.7321	17.0	85	0.8895	0.5929
0.7009	18.0	90	0.8631	0.5594
0.6914	19.0	95	0.8692	0.5926
0.6339	20.0	100	0.8335	0.5637
0.6184	21.0	105	0.8327	0.6136
0.6095	22.0	110	0.8194	0.5735
0.5999	23.0	115	0.7862	0.5789
0.5695	24.0	120	0.7849	0.5962
0.5535	25.0	125	0.7711	0.6084
0.5525	26.0	130	0.7802	0.5977
0.5295	27.0	135	0.7753	0.6038
0.5148	28.0	140	0.7450	0.6332
0.5177	29.0	145	0.7274	0.6561
0.4961	30.0	150	0.7235	0.6561
0.4948	31.0	155	0.7263	0.6510
0.4774	32.0	160	0.7259	0.6484
0.4766	33.0	165	0.7168	0.6434
0.4623	34.0	170	0.7145	0.6459
0.4821	35.0	175	0.7094	0.6535
0.4522	36.0	180	0.7030	0.664
0.4555	37.0	185	0.6975	0.6640
0.4409	38.0	190	0.6963	0.6667
0.4386	39.0	195	0.6947	0.6745
0.4387	40.0	200	0.6810	0.6825
0.4321	41.0	205	0.6722	0.664
0.4358	42.0	210	0.6722	0.6693
0.4201	43.0	215	0.6802	0.6719
0.4223	44.0	220	0.6856	0.6667
0.4227	45.0	225	0.6821	0.6693
0.4226	46.0	230	0.6757	0.6563
0.4142	47.0	235	0.6740	0.6484
0.417	48.0	240	0.6725	0.6459
0.4144	49.0	245	0.6723	0.6459
0.4144	50.0	250	0.6720	0.6484

Framework versions

• Transformers 4.37.2
• Pytorch 2.2.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1