framing_classification_longformer_50

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: 0.3739
• Accuracy: 0.9332
• F1: 0.9608
• Precision: 0.9394
• Recall: 0.9832

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: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.8078	1.0	5152	0.8413	0.8323	0.9085	0.8323	1.0
0.7998	2.0	10304	0.8279	0.8323	0.9085	0.8323	1.0
0.9031	3.0	15456	0.9204	0.8323	0.9085	0.8323	1.0
0.7805	4.0	20608	0.8259	0.8323	0.9085	0.8323	1.0
0.8775	5.0	25760	0.8078	0.8323	0.9085	0.8323	1.0
0.7248	6.0	30912	0.7587	0.8323	0.9085	0.8323	1.0
0.8282	7.0	36064	0.7737	0.8323	0.9085	0.8323	1.0
0.774	8.0	41216	0.8283	0.8323	0.9085	0.8323	1.0
0.802	9.0	46368	0.7968	0.8323	0.9085	0.8323	1.0
0.8458	10.0	51520	0.8591	0.8323	0.9085	0.8323	1.0
0.7923	11.0	56672	0.8526	0.8323	0.9085	0.8323	1.0
0.8435	12.0	61824	0.8076	0.8323	0.9085	0.8323	1.0
0.8239	13.0	66976	0.8152	0.8323	0.9085	0.8323	1.0
0.7751	14.0	72128	0.8280	0.8323	0.9085	0.8323	1.0
0.7984	15.0	77280	0.8358	0.8323	0.9085	0.8323	1.0
0.8359	16.0	82432	0.8471	0.8323	0.9085	0.8323	1.0
0.9831	17.0	87584	0.8089	0.8323	0.9085	0.8323	1.0
0.9051	18.0	92736	0.8094	0.8323	0.9085	0.8323	1.0
0.9337	19.0	97888	0.8296	0.8323	0.9085	0.8323	1.0
0.9565	20.0	103040	0.8021	0.8323	0.9085	0.8323	1.0
0.8494	21.0	108192	0.8405	0.8323	0.9085	0.8323	1.0
0.822	22.0	113344	0.8481	0.8323	0.9085	0.8323	1.0
0.856	23.0	118496	0.8194	0.8323	0.9085	0.8323	1.0
0.8892	24.0	123648	0.8394	0.8323	0.9085	0.8323	1.0
0.7816	25.0	128800	0.7035	0.8649	0.9245	0.8639	0.9944
0.6349	26.0	133952	0.6452	0.8773	0.9309	0.8764	0.9925
0.6872	27.0	139104	0.6440	0.8820	0.9331	0.8833	0.9888
0.7452	28.0	144256	0.5578	0.8323	0.9085	0.8323	1.0
0.6425	29.0	149408	0.4712	0.8323	0.9085	0.8323	1.0
0.6705	30.0	154560	0.6447	0.8866	0.9357	0.8865	0.9907
0.5748	31.0	159712	0.4063	0.9239	0.9553	0.9340	0.9776
0.6543	32.0	164864	0.4753	0.9099	0.9482	0.9092	0.9907
0.5376	33.0	170016	0.4782	0.9099	0.9482	0.9092	0.9907
0.6895	34.0	175168	0.4383	0.9177	0.9524	0.9185	0.9888
0.5867	35.0	180320	0.4970	0.9130	0.9497	0.9152	0.9869
0.7092	36.0	185472	0.4719	0.9177	0.9521	0.9229	0.9832
0.6561	37.0	190624	0.4763	0.9146	0.9508	0.9139	0.9907
0.5693	38.0	195776	0.3947	0.9301	0.9591	0.9345	0.9851
0.4321	39.0	200928	0.4632	0.9161	0.9503	0.9382	0.9627
0.5156	40.0	206080	0.4012	0.9301	0.9593	0.9299	0.9907
0.5279	41.0	211232	0.4558	0.9224	0.9550	0.9219	0.9907
0.5489	42.0	216384	0.4438	0.9193	0.9532	0.9201	0.9888
0.5586	43.0	221536	0.4469	0.9177	0.9526	0.9157	0.9925
0.575	44.0	226688	0.4310	0.9270	0.9569	0.9405	0.9739
0.4589	45.0	231840	0.4117	0.9301	0.9591	0.9345	0.9851
0.4012	46.0	236992	0.4501	0.9239	0.9553	0.9356	0.9757
0.5395	47.0	242144	0.3989	0.9317	0.96	0.9362	0.9851
0.5009	48.0	247296	0.3739	0.9332	0.9608	0.9394	0.9832
0.5356	49.0	252448	0.3805	0.9348	0.9617	0.9395	0.9851
0.5729	50.0	257600	0.3833	0.9348	0.9617	0.9395	0.9851

Framework versions

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