Skip to content
This repository has been archived by the owner on Nov 17, 2023. It is now read-only.

Issues: apache/mxnet

[Development] MXNet 2.0 Update
#18931 opened Aug 15, 2020 by szha
Open 19
[RFC] MXNet 2.0 JVM Language development
#17783 opened Mar 6, 2020 by lanking520
Open 55
[RFC] Future of Apache MXNet
#21206 opened Jun 21, 2023 by szha
Open 5
Clear current search query, filters, and sorts
80 Open 103 Closed
80 Open 103 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

Performance: avoid memory allocation in MXNDArrayAt Feature request Performance
#19245 opened Sep 28, 2020 by kpuatamazon
The amp performance of MxNet is worse than expected. AMP Performance
#19052 opened Aug 31, 2020 by wzzju
5
[RFC] GPU performance improvements in MXNet engine Feature request Performance RFC Post requesting for comments
#18951 opened Aug 17, 2020 by ptrendx
4
Increase inference performance on Jetson Nano + issues with ctx=gpu() Build Edge devices Performance TensorRT
#18893 opened Aug 10, 2020 by benguo2
2
array protocol support for np.asarray Numpy Performance v2.0
#18558 opened Jun 14, 2020 by szha
Possible Performance Regressions Performance
#18511 opened Jun 8, 2020 by josephevans
4
topk operation is very memory intensive Performance
#18336 opened May 15, 2020 by bill10
Speed difference between mxnet.numpy and numpy Performance
#18302 opened May 13, 2020 by acphile
1
[Performance Regression] GPU memory increase for training and inference models Performance
#18280 opened May 11, 2020 by karan6181
@ptrendx
11
MXNet 8 times slower than Numpy in a simple example. Performance
#18052 opened Apr 14, 2020 by MarioGalindoQ
1
[Performance][Numpy] np.einsum can be 500 times slower than torch.einsum Numpy Performance
#18043 opened Apr 13, 2020 by sxjscience
@ptrendx
3
No speedup from using FP16 (4 times slower than PyTorch) Bug Performance
#17665 opened Feb 24, 2020 by oleg-trott
@DickJC123
12
[Performance] CachedOp/Gluon performance thread Performance
#17278 opened Jan 12, 2020 by eric-haibin-lin
1
NDArray.clip() works very slow in imperative execution on GPU. Operator Performance
#16220 opened Sep 20, 2019 by igolan
1
[Discussion] Unified performance tests and dashboard CI Discussion Performance
#15757 opened Aug 5, 2019 by sandeep-krishnamurthy
22
AMP results in higher loss Performance
#15738 opened Aug 2, 2019 by IsaacCheng1
@ptrendx
4
the recall_model->exec->Forward cost most time, how can I reduce the cost time? C++ Related to C++ Performance
#15450 opened Jul 3, 2019 by gxkevin
2
Performance regression on training resnet152 with CIFAR10 on CPU Performance
#15430 opened Jul 2, 2019 by roywei
2
Operator Performance Regression on CPU Operator Performance
#15429 opened Jul 2, 2019 by roywei
6 tasks done
23
Very Large CPU RAM Memory Consumption (>1GB) Memory Performance
#15148 opened Jun 4, 2019 by rvardimon
7
CachedOp performance regression Bug Gluon Performance
#15067 opened May 24, 2019 by lanking520
6
[Discussion] Overhead in MXNet Execution Discussion Performance
#14883 opened May 5, 2019 by szha
12
regression from cudnn upgrade from 7.3.1 to 7.5.0 CUDA Performance
#14838 opened Apr 29, 2019 by roywei
7
Always getting stuck for many hours when run python lstm_ocr_train.py under dir example/ctc Performance
#14821 opened Apr 28, 2019 by PapaMadeleine2022
2
performance degradation in model inference from 1.3.1 to 1.4.0 Performance
#14569 opened Mar 29, 2019 by apeforest
1
ProTip! Add no:assignee to see everything that’s not assigned.