Updates to merge latest changes for the RyzenAI-SW from the development branch (PR-117)

.github/workflows/ci.yml

@@ -8,14 +8,14 @@ on:
   pull_request:
     branches: [ "main" ]
     types: [opened, synchronize, reopened, ready_for_review]
-  
+
   # Allows you to run this workflow manually from the Actions tab
   workflow_dispatch:
-  
+
 jobs:
   test-initial:
     runs-on: Ubuntu-22.04
     steps:
       - name: Initial testing
         run: |
-          echo This is testing
+          echo This is testing

.gitignore

@@ -2,3 +2,4 @@
 *.swp
 __pycache__
 *node_modules*
+*.yml

example/transformers/models/llm_onnx/docs/README.md

@@ -64,6 +64,10 @@ Use "prepare_model.py" script to export, optimize and quantize the LLMs. You can
 
 Check script usage
 
+```powershell
+cd %TRANSFORMERS_ROOT%\models\llm_onnx
+python prepare_model.py --help
+
 ```python prepare_model.py --help```
 
 #### Export, Optimize and quantize the model

onnx-benchmark/README.md

@@ -98,6 +98,23 @@ https://github.com/onnx/onnx/issues/6267
 pip install onnx==1.16.1
 ```
 
+In case of PHOENIX device:
+```
+conda env config vars set XCLBINHOME="C:\Program Files\RyzenAI\1.2.0\voe-4.0-win_amd64\xclbins\phoenix"
+conda env config vars set XLNX_VART_FIRMWARE="C:\Program Files\RyzenAI\1.2.0\voe-4.0-win_amd64\xclbins\phoenix\1x4.xclbin"
+conda env config vars set XLNX_TARGET_NAME=AMD_AIE2_Nx4_Overlay
+conda env config vars set VAIP_CONFIG_HOME="C:\Program Files\RyzenAI\1.2.0\voe-4.0-win_amd64"
+conda deactivate
+conda activate ryzen-ai-1.2.0
+conda env config vars list
+```
+
+Downgrade onnx to 1.16.1 to resolve an issue with the 1.16.2 release
+https://github.com/onnx/onnx/issues/6267 
+```
+pip install onnx==1.16.1
+```
+
 
 ---