camembert_ccnet_classification_tools_classifier-only_fr_lr1e-3_V3

This model is a fine-tuned version of camembert/camembert-base-ccnet on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1260
• Accuracy: 0.9524
• Learning Rate: 0.0

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: 0.001
• train_batch_size: 24
• eval_batch_size: 192
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 60

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Rate
1.7785	1.0	14	1.5198	0.4643	0.0010
1.3586	2.0	28	1.0475	0.7381	0.0010
1.0682	3.0	42	0.7517	0.7976	0.0009
0.7986	4.0	56	0.7405	0.7262	0.0009
0.6711	5.0	70	0.6039	0.7976	0.0009
0.6062	6.0	84	0.4750	0.8333	0.0009
0.5263	7.0	98	0.3627	0.8929	0.0009
0.4188	8.0	112	0.3923	0.8452	0.0009
0.4206	9.0	126	0.3147	0.9048	0.0008
0.5178	10.0	140	0.3345	0.8571	0.0008
0.3435	11.0	154	0.3869	0.8095	0.0008
0.3486	12.0	168	0.2324	0.9405	0.0008
0.3507	13.0	182	0.2324	0.9286	0.0008
0.379	14.0	196	0.2336	0.9048	0.0008
0.3516	15.0	210	0.3526	0.8571	0.0008
0.3349	16.0	224	0.2204	0.9286	0.0007
0.2979	17.0	238	0.2769	0.9167	0.0007
0.2981	18.0	252	0.2374	0.9048	0.0007
0.2902	19.0	266	0.2410	0.9405	0.0007
0.3779	20.0	280	0.2106	0.9167	0.0007
0.2486	21.0	294	0.2172	0.9405	0.0007
0.2773	22.0	308	0.1927	0.9286	0.0006
0.2685	23.0	322	0.1876	0.9524	0.0006
0.2416	24.0	336	0.1924	0.9286	0.0006
0.2369	25.0	350	0.1686	0.9405	0.0006
0.2334	26.0	364	0.2043	0.9048	0.0006
0.223	27.0	378	0.1836	0.9405	0.0006
0.3389	28.0	392	0.2298	0.9167	0.0005
0.2863	29.0	406	0.2005	0.9167	0.0005
0.2573	30.0	420	0.1696	0.9405	0.0005
0.2192	31.0	434	0.1853	0.9286	0.0005
0.2388	32.0	448	0.1546	0.9286	0.0005
0.2461	33.0	462	0.1649	0.9286	0.0005
0.303	34.0	476	0.1588	0.9405	0.0004
0.2262	35.0	490	0.1524	0.9405	0.0004
0.3037	36.0	504	0.1469	0.9405	0.0004
0.2268	37.0	518	0.1387	0.9524	0.0004
0.2315	38.0	532	0.1896	0.9405	0.0004
0.2247	39.0	546	0.1572	0.9524	0.0003
0.1841	40.0	560	0.1512	0.9524	0.0003
0.2357	41.0	574	0.1501	0.9405	0.0003
0.2186	42.0	588	0.1642	0.9286	0.0003
0.2437	43.0	602	0.1438	0.9405	0.0003
0.2399	44.0	616	0.1835	0.9405	0.0003
0.2589	45.0	630	0.1565	0.9524	0.0003
0.2306	46.0	644	0.1868	0.9286	0.0002
0.2159	47.0	658	0.1369	0.9524	0.0002
0.212	48.0	672	0.1238	0.9524	0.0002
0.1755	49.0	686	0.1439	0.9524	0.0002
0.2242	50.0	700	0.1324	0.9524	0.0002
0.2211	51.0	714	0.1277	0.9524	0.0001
0.1589	52.0	728	0.1268	0.9405	0.0001
0.2339	53.0	742	0.1248	0.9524	0.0001
0.1963	54.0	756	0.1332	0.9524	0.0001
0.2195	55.0	770	0.1350	0.9524	0.0001
0.1619	56.0	784	0.1246	0.9524	0.0001
0.2054	57.0	798	0.1282	0.9524	5e-05
0.206	58.0	812	0.1243	0.9524	0.0000
0.188	59.0	826	0.1260	0.9524	0.0000
0.1891	60.0	840	0.1260	0.9524	0.0

Framework versions

• Transformers 4.34.0
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.14.1