facebookresearch / SentEval

A python tool for evaluating the quality of sentence embeddings.
Other
2.09k stars 309 forks source link

Incomplete dataset(MR,SUBJ,SICK-E) #63

Closed ChenXi1992 closed 5 years ago

ChenXi1992 commented 5 years ago

Part of the dataset that I download is not complete:

The MR dataset that I download only contains 74 sentences instead of 11K, also SUBJ only contain 5020 instead of 10K, same for SICK-E and SICK-MSE.

I also downloaded the dataset by the link at Readme, The SUBJ dataset only contains 5020 instead of 10K.

Can somebody take a look? Thanks.

Punchwes commented 3 years ago

I wonder how you got this solved? I also got incomplete dataset for a lot of them

ChenXi1992 commented 3 years ago

I wonder how you got this solved? I also got incomplete dataset for a lot of them

It's been a long time.. it's the operation system in my case if I remembered correctly.. I tried different operation system(Linux, Ubuntu ), one of it works.