Modelpublic

AntoineD/camembert_ccnet_classification_tools_classifier-only_fr-p0.2

sourceHugging Faceupdated 3y agoView on Hugging Face

0likes24downloads

Model Card

camembertccnetclassificationtoolsclassifier-only_fr-p0.2

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: 0.2525
—Accuracy: 0.975
—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
—trainbatchsize: 24
—evalbatchsize: 192
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 60

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Rate
No log	1.0	7	1.8405	0.3	0.0010
No log	2.0	14	1.5880	0.425	0.0010
No log	3.0	21	1.2827	0.725	0.0009
No log	4.0	28	1.1343	0.7	0.0009
No log	5.0	35	1.0062	0.7	0.0009
No log	6.0	42	0.8774	0.75	0.0009
No log	7.0	49	0.8032	0.75	0.0009
No log	8.0	56	0.7309	0.8	0.0009
No log	9.0	63	0.6953	0.8	0.0008
No log	10.0	70	0.6711	0.725	0.0008
No log	11.0	77	0.5830	0.9	0.0008
No log	12.0	84	0.6111	0.775	0.0008
No log	13.0	91	0.6019	0.8	0.0008
No log	14.0	98	0.4941	0.9	0.0008
No log	15.0	105	0.4901	0.875	0.0008
No log	16.0	112	0.4450	0.9	0.0007
No log	17.0	119	0.5169	0.775	0.0007
No log	18.0	126	0.4281	0.9	0.0007
No log	19.0	133	0.4314	0.875	0.0007
No log	20.0	140	0.4408	0.85	0.0007
No log	21.0	147	0.3775	0.825	0.0007
No log	22.0	154	0.3641	0.875	0.0006
No log	23.0	161	0.3698	0.925	0.0006
No log	24.0	168	0.3470	0.925	0.0006
No log	25.0	175	0.3649	0.9	0.0006
No log	26.0	182	0.3400	0.95	0.0006
No log	27.0	189	0.3451	0.925	0.0006
No log	28.0	196	0.3841	0.875	0.0005
No log	29.0	203	0.3141	0.95	0.0005
No log	30.0	210	0.3150	0.95	0.0005
No log	31.0	217	0.3493	0.9	0.0005
No log	32.0	224	0.3115	0.95	0.0005
No log	33.0	231	0.3133	0.95	0.0005
No log	34.0	238	0.3169	0.925	0.0004
No log	35.0	245	0.3054	0.95	0.0004
No log	36.0	252	0.2951	0.975	0.0004
No log	37.0	259	0.3018	0.9	0.0004
No log	38.0	266	0.2918	0.9	0.0004
No log	39.0	273	0.2817	0.95	0.0003
No log	40.0	280	0.2723	0.95	0.0003
No log	41.0	287	0.2618	0.95	0.0003
No log	42.0	294	0.2779	0.95	0.0003
No log	43.0	301	0.2806	0.95	0.0003
No log	44.0	308	0.2560	0.95	0.0003
No log	45.0	315	0.2566	0.95	0.0003
No log	46.0	322	0.2543	0.95	0.0002
No log	47.0	329	0.2784	0.975	0.0002
No log	48.0	336	0.2974	0.925	0.0002
No log	49.0	343	0.2755	0.975	0.0002
No log	50.0	350	0.2532	0.95	0.0002
No log	51.0	357	0.2495	0.95	0.0001
No log	52.0	364	0.2700	0.95	0.0001
No log	53.0	371	0.2808	0.95	0.0001
No log	54.0	378	0.2848	0.975	0.0001
No log	55.0	385	0.2728	0.975	0.0001
No log	56.0	392	0.2646	0.975	0.0001
No log	57.0	399	0.2592	0.975	5e-05
No log	58.0	406	0.2561	0.975	0.0000
No log	59.0	413	0.2525	0.975	0.0000
No log	60.0	420	0.2525	0.975	0.0

Framework versions

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