Update 5.在无标记数据集上进行预训练.md

This commit is contained in:
yuhui 2025-04-14 18:05:40 +08:00 committed by GitHub
parent 60bbaa8712
commit 1ba8e31763
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 2 additions and 2 deletions

View File

@ -328,7 +328,7 @@ neg_avg_log_probas = avg_log_probas * -1
print(neg_avg_log_probas)
```
结算的结果为:`tensor(-10.7940)`。
结算的结果为:`tensor(10.7940)`。
这种将负值 -10.7940 转化为正值 10.7940 的操作在深度学习中称为交叉熵损失。