-
Hello there,
First, I'd like to express my appreciation for your excellent work on this project.
While experimenting with PPO/RW using this repository, I consistently encounter Out of Memory (OOM) e…
-
Arroyo is a very good library, and we ran into some performance issues when using it, and we found that there were large-scale decoding operations, as shown below.
- aws_batch_compute_environment
### Important Factoids
`image_id` was deprecated in favor of `ec2_configuration.image_id_override`. be sure to also check:
- update…
-
[version](https://github.com/NVIDIA/TensorRT-LLM/tree/31ac30e928a2db795799fdcab6be446bfa3a3998)
When I build model with paged_context_fmha = true and max_num_tokens = 4096, chunked context is enabled…
-
NOTE: Strikethrough ops are deliberately not implemented because the default broadcasting behavior of enzyme batch is enough.
- [ ] StableHLO
- [ ] ~AbsOp~
- [ ] ~AddOp~
- [ ] ~AfterAllOp~
- …
-
### Library name and version
Azure.Messaging.ServiceBus 7.18.1
### Describe the bug
There is some error when estimating batch size.
Some batches (with size that almost hit size limit) were created…
-
**Is your feature request related to a problem? Please describe.**
It appears that the BigQuery DBAPI does not support multi-row INSERT batching for more performant python-based DML transactions. Cu…
-
Is the intent of "Report Partition" to ensure that reports are only aggregated together if they share the same label? If so, you may want to define a [batch mode](https://ietf-wg-ppm.github.io/draft-i…
-
**1. Environment**
- Device: [Pixel 6]
- OS: [Android 13]
- APP: [0.9.6]
**2. Describe the feature request**
_(I may not respond in a timely manner, but you will be notified when this…
-
## Description 任务介绍
Develop backward function for batch_norm operator.
开发batch_norm算子的反向功能。
## Requirements 任务要求
__Interface 接口__
`batch_norm_backward(Tensor grad_out, Tensor input, Tensor weight, Ten…