clinico-bsc-bio-ehr-es-finetuned

This model is a fine-tuned version of joheras/bsc-bio-ehr-es-finetuned-clinais on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.9890
Precision: 0.4656
Recall: 0.6294
F1: 0.5352
Accuracy: 0.8600

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	1.1489	0.0399	0.0623	0.0486	0.6528
No log	2.0	50	0.7106	0.1739	0.2070	0.1890	0.8063
No log	3.0	75	0.6518	0.2380	0.2724	0.2541	0.8202
No log	4.0	100	0.6281	0.3515	0.4562	0.3971	0.8288
No log	5.0	125	0.6029	0.3424	0.4372	0.3840	0.8357
No log	6.0	150	0.5911	0.3479	0.4446	0.3904	0.8344
No log	7.0	175	0.5809	0.3363	0.4382	0.3806	0.8429
No log	8.0	200	0.5792	0.3338	0.4593	0.3867	0.8442
No log	9.0	225	0.5980	0.3465	0.4984	0.4088	0.8531
No log	10.0	250	0.6144	0.3700	0.5333	0.4369	0.8511
No log	11.0	275	0.6376	0.3634	0.5280	0.4305	0.8445
No log	12.0	300	0.6668	0.3802	0.5396	0.4461	0.8454
No log	13.0	325	0.6618	0.3957	0.5692	0.4669	0.8506
No log	14.0	350	0.6622	0.3906	0.5713	0.4640	0.8563
No log	15.0	375	0.6637	0.4241	0.5808	0.4902	0.8565
No log	16.0	400	0.6884	0.4251	0.5903	0.4943	0.8594
No log	17.0	425	0.7183	0.4213	0.6051	0.4967	0.8517
No log	18.0	450	0.7387	0.4104	0.5977	0.4867	0.8545
No log	19.0	475	0.7256	0.4261	0.5998	0.4982	0.8563
0.346	20.0	500	0.7427	0.4178	0.6040	0.4940	0.8534
0.346	21.0	525	0.7562	0.4240	0.6008	0.4972	0.8569
0.346	22.0	550	0.7590	0.4038	0.6051	0.4844	0.8517
0.346	23.0	575	0.7677	0.4163	0.6146	0.4964	0.8573
0.346	24.0	600	0.8148	0.4127	0.6040	0.4904	0.8483
0.346	25.0	625	0.7992	0.4230	0.6030	0.4972	0.8533
0.346	26.0	650	0.8156	0.4203	0.6072	0.4968	0.8537
0.346	27.0	675	0.7999	0.4356	0.6103	0.5084	0.8562
0.346	28.0	700	0.8326	0.4379	0.6146	0.5114	0.8508
0.346	29.0	725	0.8394	0.4441	0.6209	0.5178	0.8542
0.346	30.0	750	0.8414	0.4373	0.6072	0.5084	0.8535
0.346	31.0	775	0.8363	0.4394	0.6082	0.5102	0.8566
0.346	32.0	800	0.8442	0.4536	0.6188	0.5234	0.8574
0.346	33.0	825	0.8470	0.4655	0.6199	0.5317	0.8608
0.346	34.0	850	0.8323	0.4647	0.6177	0.5304	0.8587
0.346	35.0	875	0.8590	0.4495	0.6199	0.5211	0.8573
0.346	36.0	900	0.8457	0.4542	0.6230	0.5254	0.8589
0.346	37.0	925	0.8720	0.4543	0.6251	0.5262	0.8552
0.346	38.0	950	0.8736	0.4562	0.6167	0.5245	0.8562
0.346	39.0	975	0.8710	0.4384	0.6199	0.5136	0.8543
0.0257	40.0	1000	0.8805	0.4416	0.6230	0.5169	0.8569
0.0257	41.0	1025	0.8963	0.4634	0.6209	0.5307	0.8536
0.0257	42.0	1050	0.8973	0.4619	0.6146	0.5274	0.8546
0.0257	43.0	1075	0.9123	0.4733	0.6188	0.5364	0.8571
0.0257	44.0	1100	0.9169	0.4570	0.6230	0.5273	0.8532
0.0257	45.0	1125	0.9094	0.4847	0.6357	0.5500	0.8592
0.0257	46.0	1150	0.9096	0.4761	0.6304	0.5425	0.8611
0.0257	47.0	1175	0.9074	0.4622	0.6262	0.5318	0.8590
0.0257	48.0	1200	0.9087	0.4536	0.6199	0.5239	0.8582
0.0257	49.0	1225	0.9412	0.4426	0.6103	0.5131	0.8580
0.0257	50.0	1250	0.9221	0.4435	0.6262	0.5193	0.8587
0.0257	51.0	1275	0.9232	0.4608	0.6199	0.5286	0.8578
0.0257	52.0	1300	0.9313	0.4696	0.6199	0.5344	0.8592
0.0257	53.0	1325	0.9340	0.4529	0.6241	0.5249	0.8603
0.0257	54.0	1350	0.9418	0.4599	0.6241	0.5296	0.8561
0.0257	55.0	1375	0.9428	0.4608	0.6146	0.5267	0.8579
0.0257	56.0	1400	0.9386	0.4728	0.6230	0.5376	0.8608
0.0257	57.0	1425	0.9467	0.4641	0.6209	0.5312	0.8579
0.0257	58.0	1450	0.9402	0.4639	0.6167	0.5295	0.8614
0.0257	59.0	1475	0.9389	0.4667	0.6220	0.5333	0.8601
0.0095	60.0	1500	0.9363	0.4633	0.6262	0.5326	0.8597
0.0095	61.0	1525	0.9302	0.4706	0.6251	0.5370	0.8604
0.0095	62.0	1550	0.9456	0.4707	0.6272	0.5378	0.8609
0.0095	63.0	1575	0.9470	0.4700	0.6283	0.5377	0.8602
0.0095	64.0	1600	0.9706	0.4609	0.6230	0.5299	0.8562
0.0095	65.0	1625	0.9710	0.4785	0.6230	0.5413	0.8567
0.0095	66.0	1650	0.9715	0.4806	0.6283	0.5446	0.8568
0.0095	67.0	1675	0.9638	0.4621	0.6177	0.5287	0.8586
0.0095	68.0	1700	0.9750	0.4754	0.6230	0.5393	0.8568
0.0095	69.0	1725	0.9856	0.4643	0.6251	0.5329	0.8554
0.0095	70.0	1750	0.9855	0.4512	0.6199	0.5222	0.8570
0.0095	71.0	1775	0.9811	0.4756	0.6272	0.5410	0.8563
0.0095	72.0	1800	0.9858	0.4679	0.6167	0.5321	0.8569
0.0095	73.0	1825	0.9794	0.4676	0.6241	0.5346	0.8580
0.0095	74.0	1850	0.9774	0.4772	0.6199	0.5393	0.8572
0.0095	75.0	1875	0.9772	0.4810	0.6272	0.5445	0.8580
0.0095	76.0	1900	0.9805	0.4757	0.6294	0.5418	0.8584
0.0095	77.0	1925	0.9782	0.4782	0.6251	0.5419	0.8585
0.0095	78.0	1950	0.9921	0.4731	0.6315	0.5409	0.8572
0.0095	79.0	1975	0.9797	0.4684	0.6188	0.5332	0.8599
0.0057	80.0	2000	0.9844	0.4747	0.6251	0.5397	0.8585
0.0057	81.0	2025	0.9824	0.4636	0.6251	0.5324	0.8579
0.0057	82.0	2050	0.9803	0.4765	0.6220	0.5396	0.8591
0.0057	83.0	2075	0.9834	0.4742	0.6304	0.5413	0.8607
0.0057	84.0	2100	0.9897	0.4727	0.6315	0.5407	0.8585
0.0057	85.0	2125	0.9835	0.4723	0.6220	0.5369	0.8573
0.0057	86.0	2150	0.9838	0.4773	0.6230	0.5405	0.8580
0.0057	87.0	2175	0.9879	0.4663	0.6220	0.5330	0.8579
0.0057	88.0	2200	0.9844	0.4806	0.6262	0.5438	0.8595
0.0057	89.0	2225	0.9903	0.4767	0.6262	0.5413	0.8588
0.0057	90.0	2250	0.9929	0.4806	0.6283	0.5446	0.8581
0.0057	91.0	2275	0.9947	0.4873	0.6294	0.5493	0.8576
0.0057	92.0	2300	0.9888	0.4713	0.6251	0.5374	0.8581
0.0057	93.0	2325	0.9869	0.4780	0.6315	0.5441	0.8586
0.0057	94.0	2350	0.9866	0.4807	0.6315	0.5459	0.8587
0.0057	95.0	2375	0.9892	0.4850	0.6315	0.5486	0.8587
0.0057	96.0	2400	0.9891	0.4735	0.6325	0.5416	0.8589
0.0057	97.0	2425	0.9883	0.4635	0.6294	0.5338	0.8593
0.0057	98.0	2450	0.9883	0.4670	0.6283	0.5358	0.8601
0.0057	99.0	2475	0.9889	0.4671	0.6304	0.5366	0.8601
0.0045	100.0	2500	0.9890	0.4656	0.6294	0.5352	0.8600

Framework versions

Transformers 4.25.1
Pytorch 1.13.0
Datasets 2.8.0
Tokenizers 0.12.1

joheras/clinico-bsc-bio-ehr-es-finetuned

clinico-bsc-bio-ehr-es-finetuned

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

为此模型做种