hosting-lexical-10k

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

• Loss: 0.2278
• Accuracy: 0.9332
• F1: 0.9278

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: 0.0005
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use 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: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
13.1909	1.0	613	0.3549	0.9171	0.9177
4.0777	2.0	1226	0.2093	0.9394	0.9310
2.6943	3.0	1839	0.2833	0.9140	0.9158
2.2544	4.0	2452	0.2122	0.9302	0.9244
1.488	5.0	3065	0.2241	0.9340	0.9276
0.8386	6.0	3678	0.2202	0.9332	0.9271
1.4058	7.0	4291	0.2439	0.9233	0.9238
1.1671	8.0	4904	0.2386	0.9294	0.9253
1.2749	9.0	5517	0.2307	0.9355	0.9262
0.7176	10.0	6130	0.2207	0.9348	0.9273
0.7662	11.0	6743	0.2334	0.9317	0.9261
0.6094	12.0	7356	0.2609	0.9248	0.9205
1.2271	13.0	7969	0.2310	0.9363	0.9267
0.4984	14.0	8582	0.2321	0.9325	0.9273
0.8758	15.0	9195	0.2618	0.9363	0.9250
0.9482	16.0	9808	0.2416	0.9325	0.9244
0.7253	17.0	10421	0.2317	0.9302	0.9265
0.6299	18.0	11034	0.2376	0.9332	0.9256
0.686	19.0	11647	0.2213	0.9371	0.9264
0.799	20.0	12260	0.2387	0.9248	0.9224
0.7195	21.0	12873	0.2219	0.9355	0.9285
0.5301	22.0	13486	0.2250	0.9355	0.9270
0.6929	23.0	14099	0.2334	0.9340	0.9276
0.5645	24.0	14712	0.2219	0.9332	0.9271
0.6048	25.0	15325	0.2255	0.9378	0.9277
0.4626	26.0	15938	0.2201	0.9371	0.9295
0.7271	27.0	16551	0.2294	0.9348	0.9266
0.462	28.0	17164	0.2237	0.9332	0.9271
0.5708	29.0	17777	0.2253	0.9355	0.9285
0.3946	30.0	18390	0.2278	0.9332	0.9278

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0