Skip to content

Issues: rwth-i6/returnn

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

165 Open 491 Closed
165 Open 491 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

Train proc manager restarts after Bus error crash, still consumes GPU memory, get OutOfMemoryError
#1649 opened Nov 15, 2024 by albertz
Unexpected bus error encountered in worker
#1648 opened Nov 14, 2024 by albertz
2
remove Nose dependency
#1647 opened Nov 14, 2024 by albertz
@NeoLegends
Plan for packed dims
#1645 opened Nov 13, 2024 by albertz
@albertz @NeoLegends
2
RF (PT) meaning of losses with as_error
#1642 opened Nov 8, 2024 by albertz
@albertz @NeoLegends
Tensor Dim, support Dim.capacity > max(Dim.dyn_size_ext) JAX TPU
#1641 opened Nov 6, 2024 by albertz
Potential timeout during data caching in multi-node trainings bug
#1638 opened Oct 23, 2024 by NeoLegends
@NeoLegends
2
RF cross_entropy (matmul, gather) should maybe have allow_broadcast?
#1636 opened Oct 19, 2024 by albertz
Remove outdated Python header attribs?
#1635 opened Oct 18, 2024 by albertz
@albertz @curufinwe
@NeoLegends @JackTemaki
Sharding for multi-GPU training
#1634 opened Oct 15, 2024 by albertz
@NeoLegends
rf.BatchNorm keeps updating statistics when used in eval mode in training returnn-frontend
#1625 opened Sep 12, 2024 by mnghiap
1
TF end layer independent of batch causes error in beam search
#1606 opened Aug 26, 2024 by albertz
rf.RelPosCausalSelfAttention fails with single_step_dim returnn-frontend
#1585 opened Jul 17, 2024 by LucaG1
9
Torch multiple simultaneous gradient_checkpoint_scope
#1583 opened Jul 15, 2024 by albertz
Torch gradient_checkpoint_scope potential memory leak
#1582 opened Jul 12, 2024 by albertz
Torch gradient_checkpoint_scope could trigger segmentation fault?
#1581 opened Jul 12, 2024 by albertz
16
RuntimeError: CUDA error: an illegal memory access was encountered
#1577 opened Jul 9, 2024 by albertz
1
Torch: print model at log verbosity 3
#1575 opened Jul 5, 2024 by NeoLegends
1
@NeoLegends
1
multiprocessing: OSError: AF_UNIX path too long
#1571 opened Jul 3, 2024 by michelwi
@NeoLegends
11
Ignore a single broken gradient
#1568 opened Jul 2, 2024 by JackTemaki
2
Datasets: blocklist in addition to allowlist for segment list file
#1566 opened Jul 2, 2024 by NeoLegends
Hang in training (often with multi GPU training)
#1558 opened Jun 28, 2024 by albertz
1
RF scaled_dot_product_attention
#1555 opened Jun 27, 2024 by albertz
SlowMo (BMUF) support for PyTorch distributed training MultiGPU PyTorch
#1553 opened Jun 26, 2024 by albertz
Tensor deepcopy does not copy raw_tensor
#1541 opened Jun 17, 2024 by albertz
1
ProTip! Follow long discussions with comments:>50.