Search results

[Enhancement] Adding parallelism info to Accelerator for parallelism logging in W&B

Status: Open.

#3705 In huggingface/accelerate;

· WoosungMyung opened on Aug 3, 2025

UserWarning: Found Missing Adapter keys while loading checkpoints

Status: Open.

#3704 In huggingface/accelerate;

· manitadayon opened on Aug 2, 2025

只使用accelerate数据并行，不用模型并行

Status: Open.

#3699 In huggingface/accelerate;

· Seifon opened on Jul 30, 2025

Package should depend on safetensors with torch extra

Status: Open.

#3696 In huggingface/accelerate;

· xanderlent opened on Jul 27, 2025

Add an external option to toggle offload_to_cpu

Status: Open.

#3694 In huggingface/accelerate;

· scissorstail opened on Jul 25, 2025

logger message "dataset had no length" confusing when <code>drop_last=True</code>

Status: Open.

#3693 In huggingface/accelerate;

· lpjiang97 opened on Jul 23, 2025

Advice with GPU computations in collator/pre-process

Status: Open.

#3691 In huggingface/accelerate;

· saikoneru opened on Jul 22, 2025

[DeepSpeed] GPU VRAM usage increases each validation step

Status: Open.

#3690 In huggingface/accelerate;

· RJExtrac opened on Jul 22, 2025

What will happened if I use mix_precision but don't add with accelerator.autocast():

Status: Open.

#3689 In huggingface/accelerate;

· huangfu170 opened on Jul 22, 2025

dtype issue when using accelerate

Status: Open.

#3685 In huggingface/accelerate;

· ZhangYi1999 opened on Jul 18, 2025

bugs in accelerator.unwrap_model

Status: Open.

#3683 In huggingface/accelerate;

· ThisisBillhe opened on Jul 16, 2025

weight is not tied within <code>init_empty_weights</code>

Status: Open.

#3668 In huggingface/accelerate;

· Ghost-LZW opened on Jul 2, 2025