bart-large-mnli-aitools-6n

This model is a fine-tuned version of facebook/bart-large-mnli on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2748
• Accuracy: 0.9444

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: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.09	50	0.0885	0.9762
No log	0.18	100	0.4805	0.8571
No log	0.26	150	0.2582	0.9524
No log	0.35	200	0.2742	0.9286
No log	0.44	250	0.1553	0.9683
No log	0.53	300	0.2574	0.9603
No log	0.62	350	0.3690	0.9444
No log	0.7	400	0.3113	0.9365
No log	0.79	450	0.3474	0.9206
0.3671	0.88	500	0.2385	0.9206
0.3671	0.97	550	0.2947	0.9365
0.3671	1.05	600	0.2834	0.9444
0.3671	1.14	650	0.2425	0.9524
0.3671	1.23	700	0.2494	0.9524
0.3671	1.32	750	0.3040	0.9444
0.3671	1.41	800	0.2974	0.9444
0.3671	1.49	850	0.2268	0.9683
0.3671	1.58	900	0.3889	0.9365
0.3671	1.67	950	0.3333	0.8968
0.1777	1.76	1000	0.2748	0.9444
0.1777	1.85	1050	0.3463	0.9206
0.1777	1.93	1100	0.2951	0.9444
0.1777	2.02	1150	0.2726	0.9524
0.1777	2.11	1200	0.3241	0.9444
0.1777	2.2	1250	0.3543	0.9365
0.1777	2.28	1300	0.4440	0.9444

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu116
• Datasets 2.9.0
• Tokenizers 0.13.2