clinico-roberta-biomedical-finetuned

This model is a fine-tuned version of joheras/roberta-base-biomedical-clinical-es-finetuned-clinais on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.9272
Precision: 0.5095
Recall: 0.6463
F1: 0.5698
Accuracy: 0.8623

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: 32
eval_batch_size: 32
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	25	1.2199	0.0033	0.0053	0.0040	0.5756
No log	2.0	50	0.7306	0.2031	0.2642	0.2296	0.8021
No log	3.0	75	0.6366	0.2967	0.3811	0.3336	0.8235
No log	4.0	100	0.6135	0.3497	0.4653	0.3993	0.8304
No log	5.0	125	0.5845	0.3421	0.4537	0.3900	0.8331
No log	6.0	150	0.5697	0.3307	0.4421	0.3784	0.8390
No log	7.0	175	0.5415	0.3211	0.4495	0.3746	0.8471
No log	8.0	200	0.5430	0.3589	0.5179	0.4240	0.8567
No log	9.0	225	0.5513	0.3342	0.5474	0.4150	0.8604
No log	10.0	250	0.5681	0.3769	0.5768	0.4559	0.8582
No log	11.0	275	0.5813	0.3756	0.5863	0.4579	0.8553
No log	12.0	300	0.6096	0.4181	0.5968	0.4918	0.8574
No log	13.0	325	0.6318	0.3978	0.6042	0.4797	0.8539
No log	14.0	350	0.6309	0.3892	0.5968	0.4711	0.8553
No log	15.0	375	0.6559	0.3987	0.5968	0.4781	0.8565
No log	16.0	400	0.6391	0.4275	0.6021	0.5	0.8560
No log	17.0	425	0.6812	0.4388	0.6074	0.5095	0.8584
No log	18.0	450	0.6901	0.4287	0.6137	0.5048	0.8563
No log	19.0	475	0.6834	0.4572	0.6074	0.5217	0.8581
0.3478	20.0	500	0.7050	0.4397	0.6179	0.5138	0.8573
0.3478	21.0	525	0.7004	0.4462	0.6242	0.5204	0.8591
0.3478	22.0	550	0.7038	0.4264	0.6126	0.5028	0.8599
0.3478	23.0	575	0.7384	0.4416	0.6284	0.5187	0.8576
0.3478	24.0	600	0.7197	0.4479	0.62	0.5201	0.8619
0.3478	25.0	625	0.7412	0.4381	0.6221	0.5141	0.8559
0.3478	26.0	650	0.7535	0.4489	0.6242	0.5222	0.8566
0.3478	27.0	675	0.7534	0.4657	0.6432	0.5402	0.8586
0.3478	28.0	700	0.7672	0.4525	0.6168	0.5220	0.8567
0.3478	29.0	725	0.7680	0.4637	0.6316	0.5348	0.8599
0.3478	30.0	750	0.7590	0.4611	0.6242	0.5304	0.8607
0.3478	31.0	775	0.7671	0.4732	0.6326	0.5414	0.8625
0.3478	32.0	800	0.7921	0.4674	0.6337	0.5380	0.8590
0.3478	33.0	825	0.8037	0.4828	0.6358	0.5488	0.8574
0.3478	34.0	850	0.8376	0.4644	0.6242	0.5326	0.8534
0.3478	35.0	875	0.8346	0.4815	0.6284	0.5452	0.8552
0.3478	36.0	900	0.8249	0.4750	0.6305	0.5418	0.8567
0.3478	37.0	925	0.8420	0.4580	0.6305	0.5306	0.8548
0.3478	38.0	950	0.8341	0.4773	0.6305	0.5433	0.8550
0.3478	39.0	975	0.8085	0.4792	0.6316	0.5450	0.8653
0.0274	40.0	1000	0.7954	0.4992	0.6474	0.5637	0.8651
0.0274	41.0	1025	0.8145	0.4923	0.6421	0.5573	0.8635
0.0274	42.0	1050	0.8290	0.4911	0.6368	0.5545	0.8610
0.0274	43.0	1075	0.8468	0.4821	0.6379	0.5492	0.8571
0.0274	44.0	1100	0.8274	0.4791	0.6389	0.5476	0.8625
0.0274	45.0	1125	0.8583	0.4831	0.6305	0.5470	0.8551
0.0274	46.0	1150	0.8420	0.4726	0.6347	0.5418	0.8589
0.0274	47.0	1175	0.8631	0.5029	0.64	0.5632	0.8564
0.0274	48.0	1200	0.8421	0.4911	0.64	0.5558	0.8617
0.0274	49.0	1225	0.8564	0.5071	0.6411	0.5662	0.8631
0.0274	50.0	1250	0.8659	0.4845	0.6263	0.5464	0.8603
0.0274	51.0	1275	0.8596	0.4860	0.64	0.5525	0.8632
0.0274	52.0	1300	0.8713	0.4856	0.6368	0.5510	0.8593
0.0274	53.0	1325	0.8888	0.4868	0.64	0.5530	0.8585
0.0274	54.0	1350	0.8591	0.4816	0.6337	0.5473	0.8610
0.0274	55.0	1375	0.8755	0.4996	0.64	0.5611	0.8615
0.0274	56.0	1400	0.8749	0.5095	0.6484	0.5706	0.8583
0.0274	57.0	1425	0.8867	0.5025	0.6453	0.5650	0.8580
0.0274	58.0	1450	0.8905	0.4947	0.6337	0.5556	0.8579
0.0274	59.0	1475	0.8911	0.4881	0.6495	0.5574	0.8596
0.0099	60.0	1500	0.9220	0.4914	0.6347	0.5540	0.8570
0.0099	61.0	1525	0.8687	0.4786	0.6368	0.5465	0.8594
0.0099	62.0	1550	0.9080	0.4906	0.6337	0.5531	0.8575
0.0099	63.0	1575	0.9004	0.4831	0.6337	0.5483	0.8583
0.0099	64.0	1600	0.8906	0.4778	0.6337	0.5448	0.8619
0.0099	65.0	1625	0.8870	0.4959	0.6368	0.5576	0.8618
0.0099	66.0	1650	0.8843	0.4851	0.6358	0.5503	0.8611
0.0099	67.0	1675	0.8923	0.4912	0.6453	0.5578	0.8618
0.0099	68.0	1700	0.8864	0.4898	0.6337	0.5525	0.8615
0.0099	69.0	1725	0.8974	0.4943	0.6411	0.5582	0.8615
0.0099	70.0	1750	0.8851	0.4821	0.6379	0.5492	0.8611
0.0099	71.0	1775	0.8958	0.4920	0.6453	0.5583	0.8593
0.0099	72.0	1800	0.8880	0.4988	0.6411	0.5610	0.8618
0.0099	73.0	1825	0.8959	0.4852	0.6379	0.5512	0.8606
0.0099	74.0	1850	0.9036	0.4773	0.6305	0.5433	0.8598
0.0099	75.0	1875	0.9031	0.4864	0.6389	0.5523	0.8615
0.0099	76.0	1900	0.9243	0.4907	0.6368	0.5543	0.8590
0.0099	77.0	1925	0.9285	0.4877	0.6453	0.5555	0.8590
0.0099	78.0	1950	0.9261	0.5074	0.6516	0.5705	0.8598
0.0099	79.0	1975	0.9374	0.5037	0.64	0.5637	0.8580
0.0061	80.0	2000	0.9165	0.5021	0.6316	0.5594	0.8621
0.0061	81.0	2025	0.9307	0.5162	0.6368	0.5702	0.8582
0.0061	82.0	2050	0.9369	0.4911	0.6358	0.5541	0.8574
0.0061	83.0	2075	0.9293	0.5191	0.6421	0.5741	0.8584
0.0061	84.0	2100	0.9187	0.5004	0.6453	0.5637	0.8629
0.0061	85.0	2125	0.9293	0.4927	0.6379	0.5560	0.8623
0.0061	86.0	2150	0.9200	0.5041	0.6453	0.5660	0.8634
0.0061	87.0	2175	0.9273	0.4992	0.6421	0.5617	0.8631
0.0061	88.0	2200	0.9325	0.5021	0.6442	0.5643	0.8623
0.0061	89.0	2225	0.9245	0.4844	0.6389	0.5511	0.8630
0.0061	90.0	2250	0.9291	0.4979	0.6368	0.5589	0.8593
0.0061	91.0	2275	0.9264	0.5083	0.6432	0.5678	0.8622
0.0061	92.0	2300	0.9283	0.5025	0.6411	0.5634	0.8619
0.0061	93.0	2325	0.9264	0.5008	0.6442	0.5635	0.8613
0.0061	94.0	2350	0.9205	0.5079	0.6463	0.5688	0.8626
0.0061	95.0	2375	0.9223	0.5121	0.6484	0.5722	0.8625
0.0061	96.0	2400	0.9244	0.5045	0.6421	0.5651	0.8620
0.0061	97.0	2425	0.9248	0.5062	0.6463	0.5677	0.8622
0.0061	98.0	2450	0.9277	0.5037	0.6453	0.5658	0.8621
0.0061	99.0	2475	0.9272	0.5083	0.6463	0.5690	0.8623
0.0046	100.0	2500	0.9272	0.5095	0.6463	0.5698	0.8623

Framework versions

Transformers 4.27.0.dev0
Pytorch 1.13.0
Datasets 2.8.0
Tokenizers 0.12.1

joheras/clinico-roberta-biomedical-finetuned

clinico-roberta-biomedical-finetuned

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

为此模型做种