SmallSmallQiu

Results 10 comments of SmallSmallQiu

It might be a problem of java.

It was changed, but it didn't work

It might be a problem of java.

![image](https://user-images.githubusercontent.com/44115671/69937247-2eafef80-1515-11ea-942e-f436a554c819.png) You need to find the Java library file in your environment

I just want to tease, why use caffe such a poor frame

It's too hard to install

out of memory if don`t fix VGG16 param

谢谢

'Meteor' object has no attribute 'lock'

![1573559874(1)](https://user-images.githubusercontent.com/44115671/68670320-9e226700-0587-11ea-93d6-a3eb742ad4ea.png)

The problem with the reference to the strengthening module code

I have tried to match a generated description with the same tag in five sentences into the calculation, and the result is similar

The problem with the reference to the strengthening module code

> Do you always get cider score close to 3? Yes, cider can easily score above 3 or even higher. I've tried using different libraries for the same data to...

The problem with the reference to the strengthening module code

> Do you always get cider score close to 3? One scored 1.58, the other 2.93, and the data was made up of 10 copies. The training model and the...

The problem with the reference to the strengthening module code

> Do you always get cider score close to 3? I also made the data score when each batch of training was 400, as shown in the figure below。 One...

The problem with the reference to the strengthening module code

> The ciderD in my package was not changed. For sanity check, use CiderD(df='corpus') to define CiderD_scorer, because that's what is used in nlgeval https://github.com/Maluuba/nlg-eval#important-note Ok, thanks for your reply....