framing_classification_longformer_30_augmented_multi

This model is a fine-tuned version of allenai/longformer-base-4096 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.3585
• Accuracy: 0.6170
• F1: 0.1988
• Precision: 0.2058
• Recall: 0.2476

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: 1
• eval_batch_size: 1
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Accuracy	F1	Validation Loss	Precision	Recall
1.282	1.0	7043	0.5773	0.1691	1.5099	0.1473	0.2326
2.5135	2.0	14086	0.5455	0.1008	2.2716	0.0779	0.1429
2.384	3.0	21129	0.5455	0.1008	2.6195	0.0779	0.1429
2.5536	4.0	28172	0.5455	0.1008	2.3823	0.0779	0.1429
2.3549	5.0	35215	0.5455	0.1008	2.3964	0.0779	0.1429
2.3181	6.0	42258	0.5455	0.1008	2.4343	0.0779	0.1429
2.4398	7.0	49301	0.5455	0.1008	2.4609	0.0779	0.1429
2.3715	8.0	56344	0.5455	0.1008	2.4317	0.0779	0.1429
2.5554	9.0	63387	0.5455	0.1008	2.3966	0.0779	0.1429
1.3177	10.0	70430	0.5830	0.1707	1.4776	0.1472	0.2352
1.3928	11.0	77473	0.6159	0.1750	1.5114	0.1470	0.2348
1.5202	12.0	84516	0.6159	0.1746	1.4525	0.1465	0.2337
1.4013	13.0	91559	0.5909	0.1625	1.4524	0.1399	0.2113
1.4087	14.0	98602	0.5955	0.1736	1.4572	0.1484	0.2385
2.3755	15.0	105645	0.5727	0.1420	1.9328	0.1193	0.1771
2.2211	16.0	112688	0.5943	0.1596	1.7707	0.1317	0.2043
2.0359	17.0	119731	0.5830	0.1506	1.9399	0.1248	0.19
1.7553	18.0	126774	0.5920	0.1580	1.8171	0.1306	0.2026
1.4321	19.0	133817	0.6125	0.1733	1.4162	0.1462	0.2317
1.4545	20.0	140860	0.6068	0.1728	1.4446	0.1466	0.2324
1.3939	21.0	147903	0.6148	0.1747	1.4451	0.1473	0.2345
1.4333	22.0	154946	0.5841	0.1702	1.4462	0.1474	0.2333
1.3013	23.0	161989	0.6170	0.1757	1.4099	0.1480	0.2363
1.397	24.0	169032	0.6170	0.1766	1.4181	0.1489	0.2385
1.4752	25.0	176075	0.6136	0.1727	1.3997	0.1444	0.2297
1.372	26.0	183118	1.4134	0.6170	0.1748	0.1471	0.2340
1.4563	27.0	190161	1.3920	0.6205	0.1775	0.1492	0.2394
1.3727	28.0	197204	1.3763	0.6125	0.1737	0.1465	0.2328
1.4587	29.0	204247	1.3585	0.6170	0.1988	0.2058	0.2476
1.2723	30.0	211290	1.3586	0.6136	0.1973	0.1967	0.2455

Framework versions

• Transformers 4.32.0.dev0
• Pytorch 2.0.1
• Datasets 2.14.4
• Tokenizers 0.13.3