about the ensemble model

peteanderson80 / Up-Down-Captioner

Automatic image captioning model based on Caffe, using features from bottom-up attention.

http://www.panderson.me/up-down-attention/

MIT License

243 stars 69 forks source link

about the ensemble model #18

Open BetterZhouXu opened 5 years ago

BetterZhouXu commented 5 years ago

hello, can you tell me how to do ensembling? Try different initialize method or cross-validation? I can't replicate your results in mscoco test server, and I realized that I didn't do ensembling. Thanks.

gorgeousyouth commented 5 years ago

hello, do you solve this problem about ensembling?