open-mmlab mmengine issues

open-mmlab / mmengine

OpenMMLab Foundational Library for Training Deep Learning Models

https://mmengine.readthedocs.io/

Apache License 2.0

1.18k stars 356 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

[Fix] Speed up "--resume"

#1548 chtzs opened 6 months ago
5
[Feature] 怎样得到每个类别的精确率和召回率等指标？

#1546 anzitong closed 5 months ago
1
[Feature] 训练得到的权重命名问题

#1545 anzitong closed 6 months ago
2
[Feature] about iterabledataset or webdataset

#1543 KyanChen opened 6 months ago
1
[Bug] lr is wrong when resume with an LinearLR

#1542 Cosmo1210 closed 6 months ago
1
FSDPStrategy how to set mixed_precision and other params of pytorch

#1541 apachemycat opened 6 months ago
0
[Bug] 分布式训练代码例子报错，

#1540 apachemycat closed 6 months ago
5
ValueError: torch.utils.checkpoint: please pass in use_reentrant=True or use_reentrant=False explicitly. T

#1539 apachemycat opened 6 months ago
1
[Bug] 中断后恢复训练报错RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu!

#1538 Helen-Ren-yi opened 6 months ago
0
Update is_mlu_available

#1537 ClowDragon closed 5 months ago
4
[Bug] 多卡情况下，训练后eval和离线test的精度不能保证一致

#1536 whlook opened 6 months ago
0
[Feature] Support dataset streaming?

#1535 Ablustrund opened 6 months ago
0
Bump version to 0.10.4

#1534 fanqiNO1 closed 7 months ago
0
[Bug] Unable to save results using pklfile_prefix tag

#1533 abadithela opened 7 months ago
1
fix typo

#1532 hiramf closed 7 months ago
1
[fix] Handle case ClearML already been initialized

#1531 mhtrinh closed 7 months ago
2
[Improve] Perform evaluation upon training completion

#1529 LZHgrla closed 7 months ago
0
[Fix] SLURM distributed training in containers where `scontrol` is not available

#1527 R-Fehler opened 7 months ago
1
[Bug] No module named 'mmengine.models'

#1525 hitbuyi opened 7 months ago
2
[Bug] Error Encountered with mmengine Dependency Involving JSON and Time Modules

#1523 Duguce opened 7 months ago
3
不是大模型使用并行策略的效率大大降低！

#1522 Shen001 closed 5 months ago
2
[Docs] Add OMG-Seg to ecoystem projects

#1521 evdcush closed 3 months ago
0
[Feature] Speed up the resume process of IterBased loop

#1520 YinAoXiong opened 8 months ago
2
[Bug] 性能问题：_get_valid_value函数首次调用torch.Tensor.item()时耗时过长

#1519 BenjaminPang opened 8 months ago
0
DeepSpeed2 不能自动排除冻结的参数[Bug]

#1518 Baboom-l closed 6 months ago
1
[Enhance] Enable `exclude_frozen_parameters` for `DeepSpeedEngine._zero3_consolidated_16bit_state_dict`

#1517 LZHgrla closed 7 months ago
0
Suggested combination of Runner and AmpOptimWrapper does not result in mixed precision training [Docs]

#1515 JMQuehl opened 8 months ago
0
[Bug]

#1514 fsbarros98 opened 8 months ago
0
[Feature] SegVisualizationHook: work in 'iter' but no 'epoch'

#1513 kanousen opened 8 months ago
0
Add the supported pytorch versions in README

#1512 zhouzaida closed 8 months ago
0
Fix config of readthedocs

#1511 zhouzaida closed 8 months ago
1
Is anywhere record the version support matrix between mmengine and pytorch?

#1509 kelvinwang139 closed 8 months ago
2
[Bug] DVCLiveVisBackend initializing DVC workspace in subdirectories

#1508 smarais opened 8 months ago
1
AttributeError: module 'torch.distributed' has no attribute 'ReduceOp'

#1507 AttilaLengyel-TomTom opened 8 months ago
0
[Doc]: Fix docstring of Config

#1506 MambaWong closed 9 months ago
1
[Feature] Support custom `artifact_location` in MLflowVisBackend.

#1505 daavoo closed 8 months ago
2
[Bug] MMDistributedDataParallel distributed training can not help save memory. The total memory usage is twice that of a single card.

#1504 humian321 closed 8 months ago
2
[Feature] Support calculating loss during validation

#1503 fanqiNO1 closed 6 months ago
1
Update `WandbVisBackend`

#1502 soumik12345 opened 9 months ago
0
scattering the data to gpu when using base dataelement

#1501 ajaynitk closed 8 months ago
1
[Bug] load_from pretrained checkpoint fails using FlexibleRunner and DeepSpeed

#1499 pdmct closed 9 months ago
2
[Feature] A Checkpoint hook for saving model checkpoints as Weights & Biases Artifacts

#1498 soumik12345 opened 9 months ago
1
Fix fileio backend ut

#1497 zhouzaida closed 9 months ago
0
Remove codeowners file

#1496 zhouzaida closed 9 months ago
0
Move checkpoint funtions from runner to a new sub-package

#1495 zhouzaida opened 9 months ago
0
[Fix] Fix warning capture

#1494 fanqiNO1 closed 9 months ago
0
[Feature] A Checkpoint hook for saving model checkpoints as Weights & Biases Artifacts

#1492 soumik12345 opened 9 months ago
0
[Feature] Early Stopping, Validation Loss

#1491 1dmesh opened 9 months ago
1
User can choose device in cfg. If not, fall back to current protocol.

#1490 VincentCoulombe opened 9 months ago
2
Fix typos and remove fullwidth unicode chars

#1488 evdcush closed 9 months ago
4

Previous Next