framing_classification_longformer_30_augmented

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: 0.6496
• Accuracy: 0.8751
• F1: 0.9011
• Precision: 0.8225
• Recall: 0.9963

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	Validation Loss	Accuracy	F1	Precision	Recall
0.8842	1.0	7499	0.6496	0.8751	0.9011	0.8225	0.9963
0.8986	2.0	14998	1.3833	0.5710	0.7269	0.5710	1.0
1.4876	3.0	22497	1.5014	0.5710	0.7269	0.5710	1.0
1.4259	4.0	29996	1.4886	0.5710	0.7269	0.5710	1.0
1.5043	5.0	37495	1.6020	0.5710	0.7269	0.5710	1.0
1.5677	6.0	44994	1.5306	0.5710	0.7269	0.5710	1.0
1.4929	7.0	52493	1.4485	0.5710	0.7269	0.5710	1.0
1.5105	8.0	59992	1.5439	0.5710	0.7269	0.5710	1.0
1.3803	9.0	67491	1.4443	0.5710	0.7269	0.5710	1.0
1.4626	10.0	74990	1.5080	0.5710	0.7269	0.5710	1.0
1.4786	11.0	82489	1.5953	0.5710	0.7269	0.5710	1.0
1.5471	12.0	89988	1.4525	0.5710	0.7269	0.5710	1.0
1.5419	13.0	97487	1.5372	0.5710	0.7269	0.5710	1.0
1.3997	14.0	104986	1.3026	0.5710	0.7269	0.5710	1.0
1.4623	15.0	112485	1.4700	0.5710	0.7269	0.5710	1.0
1.4559	16.0	119984	1.5842	0.5710	0.7269	0.5710	1.0
1.462	17.0	127483	1.3627	0.5710	0.7269	0.5710	1.0
1.4793	18.0	134982	1.4688	0.5710	0.7269	0.5710	1.0
1.5473	19.0	142481	1.5292	0.5710	0.7269	0.5710	1.0
1.4102	20.0	149980	1.4355	0.5710	0.7269	0.5710	1.0
1.399	21.0	157479	1.4642	0.5710	0.7269	0.5710	1.0
1.4259	22.0	164978	1.3940	0.5710	0.7269	0.5710	1.0
1.4668	23.0	172477	1.4560	0.5710	0.7269	0.5710	1.0
1.2382	24.0	179976	1.2598	0.6094	0.6599	0.6562	0.6636
1.3404	25.0	187475	1.4411	0.5656	0.4919	0.7406	0.3682
1.4606	26.0	194974	1.2831	0.6009	0.5844	0.7205	0.4916
0.6338	27.0	202473	1.8519	0.5774	0.7258	0.5765	0.9794
1.4405	28.0	209972	1.5227	0.5816	0.7308	0.5776	0.9944
0.6593	29.0	217471	1.6163	0.5507	0.7087	0.5626	0.9570
0.6664	30.0	224970	1.7090	0.5699	0.7249	0.5710	0.9925

Framework versions

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