AMDMIGraphX

AMDMIGraphX copied to clipboard

Description: model.compile() results in string of hiprtc errors when torch is imported after migraphx Steps to Reproduce: 1. pip install -r torch_requirements.txt 2. pip install -r requirements.txt 3. python3 compile_model.py...

richagadgil

Description: tensor.to(device="cuda") results in No HIP GPUs Available when torch imported after migraphx Steps to Reproduce: 1. pip install -r torch_requirements.txt 2. pip install -r requirements.txt 3. python load_model.py [load_error.zip](https://github.com/user-attachments/files/17170333/load_error.zip)...

richagadgil

Fix for Lower unsupported pooling sizes for the CPU to Reference backend #3177

10

Hi @pfultz2,@CharlieL7 I am working on : https://github.com/ROCm/AMDMIGraphX/issues/3177 (this is my first PR) as suggested by @umangyadav that you would help and review my changes. # Problem Description OneDNN has...

aditya-167

Move auto_contiguous before lowering

4

This should fix the perf isssues.

pfultz2

Add GroupQueryAttention with KV-Cache

3

turneram

Modified documentation

5

Changes include: * Updating the landing page * Removing Search and Index links * Updating the What Is using text from "Inference optimization with MIGraphX" * Creating an install and...

spolifroni-amd

Update CI point for Onnxruntime builds - Updated with changes from 2024-09-27 - Auto-generated by [create-pull-request][1] - Update .onnxrt-commit to 6e3163faa598aebcab6709b6da9eeb555b1daec9 [1]: https://github.com/peter-evans/create-pull-request

github-actions[bot]

**### **Discussion regarding ways to reduce the compile time.** SD Clip Analysis:** Name | Time (ms) | Number of times called | Note -- | -- | -- | --...

aarushjain29

Add check for fusable ops to attention fusion

1

Not all pointwise ops are mlir fusable, and so this check is required. Fixes #3479

shivadbhavsar

Update RNN-T Imports

3

Imports causing RNN-T Errors

richagadgil