Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / krasserm/fairseq-image-captioning issues and pull requests

#33 - why loss is always 0.000 in epoch 1 ?

Issue - State: closed - Opened by nvnhann over 2 years ago - 2 comments

#32 - Testing on own dataset

Issue - State: open - Opened by arsalananwar11 over 2 years ago

#31 - About tokenizing test-split captions on own dataset

Issue - State: open - Opened by gitraffica about 3 years ago - 1 comment

#30 - meta.csv not found

Issue - State: open - Opened by akshatsingh1718 about 3 years ago

#29 - how can I get these file?

Issue - State: open - Opened by andy9420 about 3 years ago

#27 - FileNotFound Error in demo

Issue - State: open - Opened by Haochen-Luo over 3 years ago

#26 - Training end-to-end on my own dataset

Issue - State: open - Opened by MiriamFarber over 3 years ago

#25 - @krasserm : Can you suggest how to use this repo to inference on our own test dataset?

Issue - State: closed - Opened by Mas-Y almost 4 years ago - 2 comments

#24 - teacher forcing ratio

Issue - State: closed - Opened by goodluck110706112 almost 4 years ago - 5 comments

#23 - WARNING: attempting to recover from OOM in forward/backward pass

Issue - State: open - Opened by pzhren almost 4 years ago - 10 comments

#21 - Can you provide a code for fairseq-video-captioning?

Issue - State: closed - Opened by tuyunbin about 4 years ago - 2 comments

#20 - Performance reproduce

Issue - State: closed - Opened by Beanocean about 4 years ago - 10 comments

#19 - Preprocess issue

Issue - State: closed - Opened by vinod-13 about 4 years ago - 10 comments

#18 - Query regarding COCO dataset

Issue - State: closed - Opened by Ashwin-Ramesh2607 about 4 years ago - 2 comments

#17 - What if using a resnet-152 when extracting features instead of the Faster-RCNN?

Issue - State: open - Opened by Kyubyong about 4 years ago - 4 comments

#16 - Is BPE applied in the caption dataset for sequence-level training

Issue - State: closed - Opened by Beanocean about 4 years ago - 3 comments

#15 - Some bug in caption.py

Issue - State: closed - Opened by hiteshK03 about 4 years ago - 3 comments

#14 - Fix import bug re: Issue #13.

Pull Request - State: closed - Opened by alex-calderwood over 4 years ago

#13 - Issue preprocessing according to the instructions

Issue - State: closed - Opened by alex-calderwood over 4 years ago - 3 comments

#12 - Train new model using custom data

Issue - State: open - Opened by AliOsm over 4 years ago

#11 - Bounding box encoding with sin/cos functions

Issue - State: open - Opened by krasserm over 4 years ago

#11 - Bounding box encoding with sin/cos functions

Issue - State: open - Opened by krasserm over 4 years ago

#10 - what is hyper-parameters for the best checkpoint?

Issue - State: closed - Opened by zplovekq over 4 years ago - 8 comments

#9 - Use Faster-RCNN directly

Issue - State: open - Opened by adrelino over 4 years ago - 7 comments

#8 - Implement Self-critical Sequence Training for Image Captioning

Issue - State: closed - Opened by krasserm over 4 years ago

#7 - Implement Meshed-Memory Transformer

Issue - State: open - Opened by krasserm over 4 years ago

#6 - Error reporting in processing image caption

Issue - State: open - Opened by SmallSmallQiu almost 5 years ago

#5 - Model Evaluation using MS COCO Caption Evaluation

Pull Request - State: closed - Opened by cstub almost 5 years ago - 1 comment

#4 - Option to fine-tune InceptionV3 encoder

Issue - State: open - Opened by krasserm almost 5 years ago - 1 comment

#3 - Support features extracted from detected objects

Pull Request - State: closed - Opened by krasserm almost 5 years ago - 1 comment

#2 - Use Karpathy split for splitting MS-COCO dataset

Pull Request - State: closed - Opened by krasserm almost 5 years ago - 1 comment

#1 - Support variable number of image feature vectors

Pull Request - State: closed - Opened by krasserm almost 5 years ago - 1 comment