50-finetuned-spiderTraining50-200

This model is a fine-tuned version of facebook/convnextv2-tiny-22k-384 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.5289
Accuracy: 0.8759
Precision: 0.8772
Recall: 0.8744
F1: 0.8721

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
distributed_type: multi-GPU
gradient_accumulation_steps: 4
total_train_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
1.9803	1.0	125	1.6169	0.6046	0.6252	0.5927	0.5728
1.1397	2.0	250	0.8750	0.7508	0.7833	0.7512	0.7476
0.9094	3.0	375	0.7509	0.7738	0.7908	0.7691	0.7669
0.9153	4.0	500	0.7770	0.7688	0.8162	0.7669	0.7704
0.9223	5.0	625	0.9739	0.7327	0.7826	0.7285	0.7262
0.8113	6.0	750	0.8510	0.7568	0.7932	0.7511	0.7478
0.7412	7.0	875	0.9463	0.7197	0.7649	0.7221	0.7146
0.6573	8.0	1000	0.8012	0.7778	0.8009	0.7821	0.7738
0.6845	9.0	1125	0.8514	0.7688	0.7995	0.7713	0.7648
0.5381	10.0	1250	0.8984	0.7558	0.7959	0.7533	0.7543
0.474	11.0	1375	0.7790	0.7928	0.8009	0.7959	0.7899
0.5682	12.0	1500	0.7508	0.7868	0.8025	0.7881	0.7841
0.5138	13.0	1625	0.8815	0.7718	0.7932	0.7750	0.7708
0.5068	14.0	1750	0.8265	0.7948	0.8180	0.8000	0.7952
0.4747	15.0	1875	0.7926	0.7948	0.8034	0.7972	0.7905
0.4201	16.0	2000	0.7410	0.8058	0.8192	0.8022	0.8041
0.3609	17.0	2125	0.7136	0.8128	0.8305	0.8112	0.8112
0.3398	18.0	2250	0.7736	0.8018	0.8194	0.8035	0.7983
0.3628	19.0	2375	0.8398	0.8058	0.8252	0.8058	0.8033
0.3083	20.0	2500	0.8905	0.7798	0.8013	0.7771	0.7794
0.3325	21.0	2625	0.7939	0.8068	0.8186	0.8093	0.8024
0.3219	22.0	2750	0.7737	0.8328	0.8402	0.8297	0.8270
0.3661	23.0	2875	0.7341	0.8258	0.8361	0.8264	0.8245
0.298	24.0	3000	0.7641	0.8158	0.8314	0.8183	0.8142
0.2679	25.0	3125	0.7392	0.8158	0.8240	0.8126	0.8089
0.2654	26.0	3250	0.7478	0.8198	0.8320	0.8201	0.8183
0.2818	27.0	3375	0.6509	0.8398	0.8433	0.8426	0.8371
0.2443	28.0	3500	0.7111	0.8378	0.8472	0.8352	0.8348
0.1856	29.0	3625	0.8103	0.8288	0.8394	0.8295	0.8252
0.2489	30.0	3750	0.6829	0.8388	0.8456	0.8424	0.8348
0.2588	31.0	3875	0.6860	0.8358	0.8421	0.8377	0.8334
0.1844	32.0	4000	0.6958	0.8428	0.8497	0.8456	0.8403
0.2136	33.0	4125	0.6840	0.8438	0.8508	0.8425	0.8414
0.1896	34.0	4250	0.6655	0.8589	0.8658	0.8587	0.8566
0.148	35.0	4375	0.7271	0.8529	0.8588	0.8526	0.8488
0.161	36.0	4500	0.7242	0.8498	0.8514	0.8491	0.8448
0.2284	37.0	4625	0.7010	0.8368	0.8422	0.8379	0.8351
0.1886	38.0	4750	0.7341	0.8408	0.8519	0.8437	0.8404
0.1352	39.0	4875	0.6598	0.8539	0.8604	0.8531	0.8520
0.1201	40.0	5000	0.6023	0.8679	0.8734	0.8716	0.8684
0.133	41.0	5125	0.6564	0.8569	0.8651	0.8551	0.8545
0.1489	42.0	5250	0.6510	0.8619	0.8643	0.8645	0.8594
0.1078	43.0	5375	0.6508	0.8619	0.8641	0.8619	0.8591
0.1243	44.0	5500	0.6339	0.8639	0.8648	0.8673	0.8618
0.1208	45.0	5625	0.5840	0.8679	0.8656	0.8677	0.8632
0.1302	46.0	5750	0.5543	0.8699	0.8702	0.8696	0.8666
0.1396	47.0	5875	0.5533	0.8709	0.8727	0.8710	0.8677
0.0995	48.0	6000	0.5555	0.8699	0.8723	0.8697	0.8663
0.0936	49.0	6125	0.5349	0.8749	0.8757	0.8741	0.8710
0.079	50.0	6250	0.5289	0.8759	0.8772	0.8744	0.8721

Framework versions

Transformers 4.33.3
Pytorch 2.0.1+cu117
Datasets 2.14.5
Tokenizers 0.13.3

zkdeng/50-finetuned-spiderTraining50-200

50-finetuned-spiderTraining50-200

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions