-
- [ ] Support Conv large weights on L1
- [ ] Support sharded input to Conv, to avoid S2I/I2S
- [ ] #14977
- [ ] Transpose to support sharded and bfp8 input
- [ ] Softmax to support sharded input
-…
-
- [ ] change output of all conv2d to ROW_MAJOR -- this will help avoid untilize in halo
- [ ] use block sharding for maxpool2d -- this will enable use of larger #cores
- [ ] upsample is using 1 core…
-
**Submitting author:** @trulsf (Truls Flatberg)
**Repository:** https://github.com/sintefore/TimeStruct.jl
**Branch with paper.md** (empty if default branch): joss_submission
**Version:** 0.9.1
**Edit…
-
**Describe the bug**
I've encountered an issue deploying Yolov8 and Yolov11 models, where its complaining about "Error during layout configuration"
**To Reproduce**
Steps to reproduce the behavio…
-
### Is there an existing issue for this?
- [X] I have searched the existing issues
### Feature Description
Hello,
I would like to propose the addition of a Retail Price Optimization Machine Lear…
-
**Submitting author:** @AlanPearl (Alan Pearl)
**Repository:** https://github.com/AlanPearl/diffopt
**Branch with paper.md** (empty if default branch): paper
**Version:** v1.0.0
**Editor:** @jbytecode…
-
Hello authors, when trying to reproduce the results in scripts/sft/chat_for_13b_v1_5.sh, in the last segment of the code with zero3 optimization, the code fails to retrieve the model weights in zero3 …
-
**Description:** The current Convolutional Neural Network (CNN) model for terrain classification is functional, but there are several areas for improvement to boost performance and efficiency. These o…
-
Agent stage optimization algorithm. FOC
- for a single stage
- looping through a fully specified model (maybe multi-stage)
-
Agent stage optimization algorithm. EGM
- for a single stage
- looping through a fully specified model (maybe multi-stage)
In my view, this would be the right place to generalize the `ConsLab…