卢实验结果

游弋白桦林

已于 2022-03-27 15:45:56 修改

阅读量316

点赞数

文章标签： python 开发语言后端

于 2022-03-26 21:53:32 首次发布

本文链接：https://blog.csdn.net/weixin_48759194/article/details/123764110

版权

内部验证

在这里插入图片描述

*************** inference best accuracy model************
 accuracy for best accuracy from my define fn: 0.8119349005424955
 accuracy for best accuracy from pytorch: 0.8119349005424955
 f1 score for best accuracy from pytorch: 0.8807339449541284
val: 100%|██████████| 1/1 [00:00<00:00,  6.30it/s]
*************** inference best f1 score model************
 accuracy for best f1 score from my define fn: 0.8119349005424955
 accuracy for best f1 score from pytorch: 0.8119349005424955
 f1 score for best f1 score from pytorch: 0.8807339449541284
val: 100%|██████████| 1/1 [00:00<00:00,  6.90it/s]
*************** inference finally eopch model************
 accuracy for finally epoch from my define fn: 0.8716094032549728
 accuracy for finally epoch from pytorch: 0.8716094032549728
 f1 score for finally epoch from pytorch: 0.9271794871794871

公开数据集上训练后的测试

data2

val: 100%|██████████| 1/1 [00:00<00:00,  1.89it/s]
*************** inference best accuracy model************
 accuracy for best accuracy from my define fn: 0.6094032549728752
 accuracy for best accuracy from pytorch: 0.6094032549728752
 f1 score for best accuracy from pytorch: 0.7000000000000001
val: 100%|██████████| 1/1 [00:00<00:00,  6.19it/s]
*************** inference best f1 score model************
 accuracy for best f1 score from my define fn: 0.6094032549728752
 accuracy for best f1 score from pytorch: 0.6094032549728752
 f1 score for best f1 score from pytorch: 0.7000000000000001
val: 100%|██████████| 1/1 [00:00<00:00,  6.07it/s]
*************** inference finally eopch model************
 accuracy for finally epoch from my define fn: 0.5750452079566004
 accuracy for finally epoch from pytorch: 0.5750452079566004
 f1 score for finally epoch from pytorch: 0.6638054363376251

data1

val: 100%|██████████| 6/6 [00:03<00:00,  1.88it/s]
*************** inference best accuracy model************
 accuracy for best accuracy from my define fn: 0.6470724369358944
 accuracy for best accuracy from pytorch: 0.6470724369358944
 f1 score for best accuracy from pytorch: 0.7749741773646156
val: 100%|██████████| 6/6 [00:01<00:00,  4.80it/s]
*************** inference best f1 score model************
 accuracy for best f1 score from my define fn: 0.6470724369358944
 accuracy for best f1 score from pytorch: 0.6470724369358944
 f1 score for best f1 score from pytorch: 0.7749741773646156
val: 100%|██████████| 6/6 [00:01<00:00,  5.08it/s]
*************** inference finally eopch model************
 accuracy for finally epoch from my define fn: 0.6651238139319602
 accuracy for finally epoch from pytorch: 0.6651238139319602
 f1 score for finally epoch from pytorch: 0.7890363026680274