Ecosyste.ms: Issues

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

GitHub / lucidrains/recurrent-memory-transformer-pytorch issues and pull requests

#25 - Update README.md

Pull Request - State: closed - Opened by avanturist322 8 days ago

#23 - Implement RMT-R (New Paper feature to RMTs)

Issue - State: open - Opened by anoojpatel 8 months ago

#22 - What is the reasoning for no dropout?

Issue - State: closed - Opened by luchris429 10 months ago - 2 comments

#19 - Question: first read memories

Issue - State: open - Opened by pfeatherstone about 1 year ago - 12 comments

#18 - Bug: resiDual implementation

Issue - State: closed - Opened by pfeatherstone about 1 year ago - 3 comments

#17 - Question: Global write tokens or recurrent

Issue - State: closed - Opened by IcarusWizard about 1 year ago - 7 comments

#16 - Feature request: make JIT and ONNX export work

Issue - State: open - Opened by pfeatherstone about 1 year ago - 4 comments

#15 - Question: masks

Issue - State: closed - Opened by pfeatherstone about 1 year ago - 3 comments

#14 - Question: why do we need read_memory_emb

Issue - State: closed - Opened by pfeatherstone over 1 year ago - 6 comments

#12 - Attend : check mask isn't already 4D

Issue - State: closed - Opened by pfeatherstone over 1 year ago - 1 comment

#11 - Question: configuring scaled_dot_product_attention

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

#10 - causal mask assert hit

Issue - State: closed - Opened by pfeatherstone over 1 year ago - 25 comments

#9 - Question: how to adapt this for CTC loss

Issue - State: open - Opened by pfeatherstone over 1 year ago - 2 comments

#8 - Question: How to set seq_len ?

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

#7 - Is rmt compatible with pretrain models like LLaMA？

Issue - State: closed - Opened by yw2278 over 1 year ago - 6 comments

#6 - What happens if texts from the dataset don't have equal lengths

Issue - State: closed - Opened by sentialx over 1 year ago - 4 comments

#5 - have you had a chance to train it yet?

Issue - State: open - Opened by Alignment-Lab-AI over 1 year ago - 2 comments

#4 - bptt depth implementation?

Issue - State: closed - Opened by DaehanKim over 1 year ago - 3 comments

#3 - token_shift

Issue - State: closed - Opened by LanZhenFeng over 1 year ago - 4 comments

#2 - flash attention, and a potentially better improvement

Issue - State: closed - Opened by Alignment-Lab-AI over 1 year ago - 5 comments