harvardnlp/annotated-transformer issues and pull requests

#131 - Multiplication positional encoding seems to work better than the original division one?

Issue - State: open - Opened by Mightlaus 29 days ago

#130 - [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: Hostname mismatch, certificate is not valid for 'www.quest.dcs.shef.ac.uk'. (_ssl.c:1147)

Issue - State: open - Opened by rhr0411 about 2 months ago - 1 comment

#129 - add docker

Pull Request - State: closed - Opened by michael-lee914 2 months ago

#128 - add docker

Pull Request - State: closed - Opened by michael-lee914 2 months ago

#127 - Synthetic copy task data question

Issue - State: open - Opened by cmmcirvin 2 months ago - 7 comments

#126 - Some `del`s are uneffective

Issue - State: open - Opened by kuraga 2 months ago

#125 - Are all Multi-Head Attention "Masked" really?

Issue - State: open - Opened by kuraga 2 months ago

#124 - Wheel no longer supported by Google Colab

Issue - State: open - Opened by ystoneman 3 months ago - 1 comment

#123 - Fixed minor grammatic mistake.

Pull Request - State: open - Opened by mdarm 8 months ago

#122 - Is there any reason we used a different LayerNorm implementation?

Issue - State: open - Opened by dpheap2222 8 months ago - 1 comment

#121 - A poll on the Embedding class

Issue - State: open - Opened by gitfourteen 8 months ago - 3 comments

#120 - Edit to remove nonessential annotations

Pull Request - State: closed - Opened by dpheap2222 8 months ago

#119 - use biased estimate of std in layernorm as in the original paper

Pull Request - State: closed - Opened by Arunprakash-A 10 months ago

#118 - About transpose processing in `MultiHeadedAttention` class.

Issue - State: open - Opened by Tinghao-NTU about 1 year ago - 2 comments

#117 - note on torch 1.11 vs torch 2.1 compatibility

Issue - State: open - Opened by BStudent about 1 year ago - 1 comment

#116 - The first column of synthetic data in the first example should be set to 0 instead of 1?

Issue - State: open - Opened by alaneuler about 1 year ago - 2 comments

#115 - Update AnnotatedTransformer.ipynb

Pull Request - State: open - Opened by satya400 about 1 year ago - 2 comments

#114 - annotated-transformer

Issue - State: open - Opened by donggegithub over 1 year ago

#113 - dockerfile

Issue - State: open - Opened by sdelahaies over 1 year ago - 1 comment

#112 - Issue with Spacy Dependency Version: issubclass() arg 1 must be a class

Issue - State: open - Opened by Hydway over 1 year ago - 4 comments

#111 - use the greedy_decode two times in check_outputs function

Issue - State: open - Opened by sunxiaoyao-git over 1 year ago

#110 - Incorrect implementaion of SublayerConnection class

Issue - State: closed - Opened by Milad137 over 1 year ago - 2 comments

#109 - label smoothing inf err

Issue - State: open - Opened by jerett over 1 year ago - 3 comments

#108 - How to calculate the BLEU score?

Issue - State: open - Opened by sauravtii over 1 year ago - 1 comment

#107 - How to do the inference?

Issue - State: open - Opened by sauravtii over 1 year ago

#106 - fix len greedy decode

Pull Request - State: open - Opened by vinhtran2611 almost 2 years ago - 2 comments

#105 - No need for a generator in the EncoderDecoder class

Issue - State: open - Opened by mkserge almost 2 years ago - 2 comments

#104 - Visualization issue

Issue - State: open - Opened by jaebak almost 2 years ago

#103 - nbatches vs batch_size

Issue - State: open - Opened by thomasahle about 2 years ago

#102 - TypeError: dropout(): argument 'input' (position 1) must be Tensor, not NoneType

Issue - State: closed - Opened by wujun about 2 years ago - 1 comment

#101 - How long is the training process?

Issue - State: closed - Opened by ccyang123 about 2 years ago - 3 comments

#100 - Some doubts about SublayerConnection

Issue - State: open - Opened by watersounds about 2 years ago - 6 comments

#99 - Added Docker Container

Pull Request - State: closed - Opened by nicholas-dinicola about 2 years ago

#98 - Typo in Multihead-attention:

Issue - State: closed - Opened by Jostarndt over 2 years ago - 4 comments

#97 - Epoch Training: Help

Issue - State: open - Opened by ludwigwittgenstein2 over 2 years ago - 1 comment

#96 - Could not get the file at http://www.quest.dcs.shef.ac.uk/wmt16_files_mmt/training.tar.gz. [RequestException] None.

Issue - State: open - Opened by tiassap over 2 years ago - 8 comments

#95 - MultiHeadedAttention: affine transforms

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

#94 - some questions about MultiHeadAtttention

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

#93 - About Shared Embedding

Issue - State: open - Opened by dawei-yu over 2 years ago - 2 comments

#92 - Should the model LayerNorm before encoding?

Issue - State: closed - Opened by dawei-yu over 2 years ago - 3 comments