Guillaume Genthial

Results 34 comments of Guillaume Genthial

Hi @393301836 , it's probably a tf version error, if you check on the tensorflow documentation, these arguments / methods should be defined. You can either modify the code to...

Hi @cedar33 , Yes, there is an efficient way to predict. Have a look at the `serve.py` script. It provides an example of how to use a predictor on a...

@cedar33 , Glad to know you solved your problem, if you have time to post more details here, I am sure other people will appreciate!

@joshgavinhong thank you! :) See other issue, not a typo, both are ok.

If you're making batchted predictions, sentences can be of different lengths and thus need to be padded before being fed to tensorflow, and as the CRF is a sequence model,...

HI @loretoparisi , This is a good suggestion, will try to provide stuff, but the purpose of this package is different than FLAIR: it's meant as something that people can...

@mungg and @lanzhuzhu , Try using the 840B.300d vectors as well as the BIOES tagging scheme. You should be able to get good results (in average). Note that some runs...

See my answer in the `sequence_tagging` repo (which is the version of the code you're using I assume).

If you use python3 you should not get any issue, but yes, technically specifying the encoding manually should fix it.

@wtkvarox no idea! never done it. If you manage to do it, write a blog post / link it here / add your solution here!