XPostsClassificationModel

This model is a fine-tuned version of FacebookAI/roberta-large on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.5237
• Accuracy: 0.9195

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: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.1653	1.0526	20	1.0058	0.4228
0.6361	2.1053	40	0.5189	0.7383
0.2347	3.1579	60	0.2813	0.8792
0.2686	4.2105	80	0.2233	0.8993
0.0763	5.2632	100	0.5500	0.8389
0.0284	6.3158	120	0.6369	0.8389
0.0084	7.3684	140	0.4523	0.8591
0.0038	8.4211	160	0.8007	0.8456
0.0289	9.4737	180	0.5008	0.9060
0.0009	10.5263	200	0.7118	0.8792
0.0068	11.5789	220	0.4647	0.9128
0.0131	12.6316	240	0.3912	0.9060
0.0007	13.6842	260	0.5074	0.9195
0.0003	14.7368	280	0.6823	0.9060
0.0007	15.7895	300	0.3905	0.9195
0.0018	16.8421	320	0.4539	0.9195
0.031	17.8947	340	0.7965	0.8993
0.0336	18.9474	360	0.5107	0.8926
0.0008	20.0	380	0.5105	0.8859
0.0002	21.0526	400	0.4367	0.9060
0.0002	22.1053	420	0.4576	0.9128
0.0004	23.1579	440	0.4095	0.9195
0.0003	24.2105	460	0.4776	0.9128
0.0002	25.2632	480	0.4460	0.9128
0.0001	26.3158	500	0.4092	0.9195
0.0001	27.3684	520	0.5776	0.8859
0.0053	28.4211	540	0.4901	0.9060
0.0124	29.4737	560	0.5235	0.9128
0.0003	30.5263	580	0.4890	0.9262
0.0033	31.5789	600	0.4679	0.8993
0.0001	32.6316	620	0.5067	0.9060
0.0001	33.6842	640	0.5433	0.9128
0.0001	34.7368	660	0.5524	0.9128
0.0001	35.7895	680	0.5494	0.9128
0.0001	36.8421	700	0.4944	0.9060
0.0001	37.8947	720	0.4902	0.9060
0.0001	38.9474	740	0.4910	0.9060
0.0001	40.0	760	0.5675	0.9128
0.0001	41.0526	780	0.6314	0.8926
0.0001	42.1053	800	0.6384	0.8926
0.0001	43.1579	820	0.6317	0.8926
0.0001	44.2105	840	0.5489	0.9195
0.0001	45.2632	860	0.5316	0.9195
0.0001	46.3158	880	0.5325	0.9195
0.0001	47.3684	900	0.5332	0.9195
0.0001	48.4211	920	0.5244	0.9195
0.0001	49.4737	940	0.5237	0.9195

Framework versions

• Transformers 4.57.1
• Pytorch 2.6.0+cu124
• Datasets 4.4.1
• Tokenizers 0.22.1