source-affiliation-model

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

• Loss: 2.3321
• F1: 0.5348

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: 5e-05
• train_batch_size: 5
• eval_batch_size: 5
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10.0

Training results

Training Loss	Epoch	Step	Validation Loss	F1
No log	0.12	100	1.4535	0.2435
No log	0.25	200	1.3128	0.3899
No log	0.37	300	1.2888	0.4413
No log	0.49	400	1.1560	0.4614
1.4848	0.62	500	1.0988	0.4477
1.4848	0.74	600	1.1211	0.4583
1.4848	0.86	700	1.1152	0.4693
1.4848	0.99	800	1.0176	0.5018
1.4848	1.11	900	1.0942	0.4774
1.1019	1.23	1000	1.1785	0.5119
1.1019	1.35	1100	1.0751	0.4797
1.1019	1.48	1200	1.0759	0.5206
1.1019	1.6	1300	1.0756	0.5231
1.1019	1.72	1400	1.1329	0.4547
0.9431	1.85	1500	1.0617	0.4852
0.9431	1.97	1600	1.1046	0.5254
0.9431	2.09	1700	1.2489	0.5069
0.9431	2.22	1800	1.2113	0.5363
0.9431	2.34	1900	1.1782	0.5546
0.7589	2.46	2000	1.0453	0.5862
0.7589	2.59	2100	1.0810	0.5223
0.7589	2.71	2200	1.1470	0.5872
0.7589	2.83	2300	1.1522	0.5553
0.7589	2.96	2400	1.0712	0.6273
0.6875	3.08	2500	1.3458	0.5768
0.6875	3.2	2600	1.7052	0.5491
0.6875	3.33	2700	1.5080	0.6582
0.6875	3.45	2800	1.5851	0.5965
0.6875	3.57	2900	1.4771	0.5691
0.5391	3.69	3000	1.6717	0.5350
0.5391	3.82	3100	1.5607	0.5448
0.5391	3.94	3200	1.5464	0.6062
0.5391	4.06	3300	1.7645	0.5755
0.5391	4.19	3400	1.6715	0.5504
0.4928	4.31	3500	1.7604	0.5626
0.4928	4.43	3600	1.8984	0.5142
0.4928	4.56	3700	1.8012	0.5763
0.4928	4.68	3800	1.7107	0.5671
0.4928	4.8	3900	1.7697	0.5598
0.4233	4.93	4000	1.6296	0.6084
0.4233	5.05	4100	2.0418	0.5343
0.4233	5.17	4200	1.8203	0.5526
0.4233	5.3	4300	1.9760	0.5292
0.4233	5.42	4400	2.0136	0.5153
0.2518	5.54	4500	2.0137	0.5121
0.2518	5.67	4600	2.0053	0.5257
0.2518	5.79	4700	1.9539	0.5423
0.2518	5.91	4800	2.0159	0.5686
0.2518	6.03	4900	2.0411	0.5817
0.2234	6.16	5000	2.0025	0.5780
0.2234	6.28	5100	2.1189	0.5413
0.2234	6.4	5200	2.1936	0.5628
0.2234	6.53	5300	2.1825	0.5210
0.2234	6.65	5400	2.0767	0.5471
0.1829	6.77	5500	1.9747	0.5587
0.1829	6.9	5600	2.1182	0.5847
0.1829	7.02	5700	2.1597	0.5437
0.1829	7.14	5800	2.0307	0.5629
0.1829	7.27	5900	2.0912	0.5450
0.1226	7.39	6000	2.2383	0.5379
0.1226	7.51	6100	2.2311	0.5834
0.1226	7.64	6200	2.2456	0.5438
0.1226	7.76	6300	2.2423	0.5860
0.1226	7.88	6400	2.2922	0.5245
0.0883	8.0	6500	2.3304	0.5650
0.0883	8.13	6600	2.3929	0.5288
0.0883	8.25	6700	2.3928	0.5344
0.0883	8.37	6800	2.3854	0.5266
0.0883	8.5	6900	2.4275	0.5339
0.044	8.62	7000	2.3929	0.5380
0.044	8.74	7100	2.3587	0.5339
0.044	8.87	7200	2.3372	0.5423
0.044	8.99	7300	2.3488	0.5424
0.044	9.11	7400	2.3543	0.5818
0.0558	9.24	7500	2.3397	0.5554
0.0558	9.36	7600	2.3255	0.5394
0.0558	9.48	7700	2.3184	0.5557
0.0558	9.61	7800	2.3293	0.5669
0.0558	9.73	7900	2.3358	0.5666
0.0323	9.85	8000	2.3307	0.5344
0.0323	9.98	8100	2.3321	0.5348

Framework versions

• Transformers 4.30.2
• Pytorch 2.0.1+cu117
• Datasets 2.13.1
• Tokenizers 0.13.3