Skip to content

Pull requests: microsoft/DeepSpeed

New pull request New
131 Open 1,531 Closed
131 Open 1,531 Closed
Author
Filter by author
Label
Filter by label
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Milestones
Filter by milestone
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Fix for Stable Diffusion
#3218 opened Apr 13, 2023 by mrwyattii Loading…
1
[update] reference in cifar-10
#3212 opened Apr 13, 2023 by dogukanutuna Loading…
zero3 checkpoint frozen params
#3205 opened Apr 12, 2023 by tjruwase Loading…
1
3
Additional changes to support MI200
#3204 opened Apr 12, 2023 by loadams • Draft
Make deepspeed.zero.Init() idempotent
#3203 opened Apr 12, 2023 by eisene Loading…
3
Update automatic-tensor-parallelism.md
#3198 opened Apr 12, 2023 by sywangyi Loading…
1
Add HE support for the rest of model containers
#3191 opened Apr 12, 2023 by RezaYazdaniAminabadi • Draft
4
improving int4 asymmetric quantization accuracy
#3190 opened Apr 11, 2023 by HeyangQin Loading…
Documentation for DeepSpeed Accelerator Abstraction Interface
#3184 opened Apr 11, 2023 by delock Loading…
AMD Kernel Compatibility Fixes
#3180 opened Apr 11, 2023 by cmikeh2 • Draft
Enable auto TP policy for llama model
#3170 opened Apr 10, 2023 by jianan-gu Loading…
5
[Fix] _conv_flops_compute when padding is a str and stride=1
#3169 opened Apr 9, 2023 by zhiruiluo Loading…
1
stage_1_and_2.py: do gradient scale only for fp16
#3166 opened Apr 9, 2023 by guoyejun Loading…
1
Fix handling of (CUDA,ROCR)_VISIBLE_DEVICES
#3165 opened Apr 8, 2023 by jglaser Loading…
Update torch version check in building sparse_attn
#3152 opened Apr 5, 2023 by loadams Loading…
1
2
Dev/fp32
#3149 opened Apr 5, 2023 by ShijieZZZZ • Draft
Update megatron-lm to support torch 2
#3148 opened Apr 5, 2023 by loadams • Draft
CPP PYBIND Module for Tensor Map
#3129 opened Apr 3, 2023 by DannyBruno Loading…
1
Disable ZeRO loading when load_module_only=True
#3116 opened Mar 30, 2023 by ploshkin Loading…
Token length type fixing
#3110 opened Mar 29, 2023 by tgergo1 Loading…
add sharded checkpoint loading for AutoTP path to reduce the peak mem…
#3102 opened Mar 27, 2023 by sywangyi Loading…
2
add bf16 cuda kernel support
#3092 opened Mar 24, 2023 by dc3671 Loading…
13
remove torch.cuda.is_available() check when compiling ops
#3085 opened Mar 23, 2023 by jinzhen-lin Loading…
5
fix mpich launcher issue in multi-node
#3078 opened Mar 22, 2023 by sywangyi Loading…
3
fix pop off grad mistakenly
#3064 opened Mar 21, 2023 by HuangLK Loading…
1
ProTip! What’s not been updated in a month: updated:<2023-03-13.