diff --git a/.github/workflows/gh-pages-releases.yml b/.github/workflows/gh-pages-releases.yml index b27fbf50..47a4c61e 100644 --- a/.github/workflows/gh-pages-releases.yml +++ b/.github/workflows/gh-pages-releases.yml @@ -10,7 +10,7 @@ jobs: runs-on: ubuntu-latest # Don't run this in everyone's forks. - if: github.repository == 'nod-ai/AMDSHARK' + if: github.repository == 'nod-ai/AMD-SHARK-Studio' steps: - name: Checking out repository @@ -18,7 +18,7 @@ jobs: with: token: ${{ secrets.NODAI_INVOCATION_TOKEN }} - name: Run scrape releases script - run: python ./build_tools/scrape_releases.py nod-ai AMDSHARK > /tmp/index.html + run: python ./build_tools/scrape_releases.py nod-ai AMD-SHARK-Studio > /tmp/index.html shell: bash - run: git fetch --all - run: git switch github-pages diff --git a/.gitmodules b/.gitmodules index 021ed197..7a944f9c 100644 --- a/.gitmodules +++ b/.gitmodules @@ -1,4 +1,4 @@ [submodule "inference/thirdparty/amdshark-runtime"] path = inference/thirdparty/amdshark-runtime url =https://github.com/nod-ai/SRT.git - branch = amdshark-06032022 + branch = shark-06032022 diff --git a/README.md b/README.md index 7441051d..475869d9 100644 --- a/README.md +++ b/README.md @@ -6,7 +6,7 @@ High Performance Machine Learning Distribution *The latest versions of this project are developments towards a refactor on top of IREE-Turbine. Until further notice, make sure you use an .exe release or a checkout of the `AMDSHARK-1.0` branch, for a working AMDSHARK-Studio* -[![Nightly Release](https://github.com/nod-ai/AMDSHARK-Studio/actions/workflows/nightly.yml/badge.svg)](https://github.com/nod-ai/AMDSHARK-Studio/actions/workflows/nightly.yml) +[![Nightly Release](https://github.com/nod-ai/AMD-SHARK-Studio/actions/workflows/nightly.yml/badge.svg)](https://github.com/nod-ai/AMD-SHARK-Studio/actions/workflows/nightly.yml)
Prerequisites - Drivers @@ -27,9 +27,9 @@ Other users please ensure you have your latest vendor drivers and Vulkan SDK fro ### Quick Start for AMDSHARK Stable Diffusion for Windows 10/11 Users -Install the Driver from [Prerequisites](https://github.com/nod-ai/AMDSHARK-Studio#install-your-hardware-drivers) above +Install the Driver from [Prerequisites](https://github.com/nod-ai/AMD-SHARK-Studio#install-your-hardware-drivers) above -Download the [stable release](https://github.com/nod-ai/AMDSHARK-Studio/releases/latest) or the most recent [AMDSHARK 1.0 pre-release](https://github.com/nod-ai/AMDSHARK-Studio/releases). +Download the [stable release](https://github.com/nod-ai/AMD-SHARK-Studio/releases/latest) or the most recent [AMDSHARK 1.0 pre-release](https://github.com/nod-ai/AMD-SHARK-Studio/releases). Double click the .exe, or [run from the command line](#running) (recommended), and you should have the [UI](http://localhost:8080/) in the browser. @@ -67,8 +67,8 @@ Enjoy. ## Check out the code ```shell -git clone https://github.com/nod-ai/AMDSHARK.git -cd AMDSHARK +git clone https://github.com/nod-ai/AMD-SHARK-Studio.git +cd AMD-SHARK-Studio ``` ## Switch to the Correct Branch (IMPORTANT!) @@ -179,12 +179,12 @@ python -m pip install --upgrade pip *macOS Metal* users please install https://sdk.lunarg.com/sdk/download/latest/mac/vulkan-sdk.dmg and enable "System wide install" -### Install AMDSHARK +### Install AMD-SHARK -This step pip installs AMDSHARK and related packages on Linux Python 3.8, 3.10 and 3.11 and macOS / Windows Python 3.11 +This step pip installs AMD-SHARK and related packages on Linux Python 3.8, 3.10 and 3.11 and macOS / Windows Python 3.11 ```shell -pip install nodai-amdshark -f https://nod-ai.github.io/AMDSHARK/package-index/ -f https://llvm.github.io/torch-mlir/package-index/ -f https://nod-ai.github.io/SRT/pip-release-links.html --extra-index-url https://download.pytorch.org/whl/nightly/cpu +pip install nodai-amdshark -f https://nod-ai.github.io/AMD-SHARK-Studio/package-index/ -f https://llvm.github.io/torch-mlir/package-index/ -f https://nod-ai.github.io/SRT/pip-release-links.html --extra-index-url https://download.pytorch.org/whl/nightly/cpu ``` ### Run amdshark tank model tests. @@ -196,7 +196,7 @@ See tank/README.md for a more detailed walkthrough of our pytest suite and CLI. ### Download and run Resnet50 sample ```shell -curl -O https://raw.githubusercontent.com/nod-ai/AMDSHARK/main/amdshark/examples/amdshark_inference/resnet50_script.py +curl -O https://raw.githubusercontent.com/nod-ai/AMD-SHARK-Studio/main/amdshark/examples/amdshark_inference/resnet50_script.py #Install deps for test script pip install --pre torch torchvision torchaudio tqdm pillow gsutil --extra-index-url https://download.pytorch.org/whl/nightly/cpu python ./resnet50_script.py --device="cpu" #use cuda or vulkan or metal @@ -204,7 +204,7 @@ python ./resnet50_script.py --device="cpu" #use cuda or vulkan or metal ### Download and run BERT (MiniLM) sample ```shell -curl -O https://raw.githubusercontent.com/nod-ai/AMDSHARK/main/amdshark/examples/amdshark_inference/minilm_jit.py +curl -O https://raw.githubusercontent.com/nod-ai/AMD-SHARK-Studio/main/amdshark/examples/amdshark_inference/minilm_jit.py #Install deps for test script pip install transformers torch --extra-index-url https://download.pytorch.org/whl/nightly/cpu python ./minilm_jit.py --device="cpu" #use cuda or vulkan or metal @@ -358,12 +358,12 @@ AMDSHARK is maintained to support the latest innovations in ML Models: | Vision Transformer | :green_heart: | :green_heart: | :green_heart: | | ResNet50 | :green_heart: | :green_heart: | :green_heart: | -For a complete list of the models supported in AMDSHARK, please refer to [tank/README.md](https://github.com/nod-ai/AMDSHARK-Studio/blob/main/tank/README.md). +For a complete list of the models supported in AMDSHARK, please refer to [tank/README.md](https://github.com/nod-ai/AMD-SHARK-Studio/blob/main/tank/README.md). ## Communication Channels * [AMDSHARK Discord server](https://discord.gg/RUqY2h2s9u): Real time discussions with the AMDSHARK team and other users -* [GitHub issues](https://github.com/nod-ai/AMDSHARK-Studio/issues): Feature requests, bugs etc +* [GitHub issues](https://github.com/nod-ai/AMD-SHARK-Studio/issues): Feature requests, bugs etc ## Related Projects diff --git a/amdshark/__init__.py b/amdshark/__init__.py index d75e8480..6d3d7c1e 100644 --- a/amdshark/__init__.py +++ b/amdshark/__init__.py @@ -14,7 +14,7 @@ def amdshark(model, inputs, *, options): log.exception( "Unable to import AMDSHARK - High Performance Machine Learning Distribution" "Please install the right version of AMDSHARK that matches the PyTorch version being used. " - "Refer to https://github.com/nod-ai/AMDSHARK-Studio/ for details." + "Refer to https://github.com/nod-ai/AMD-SHARK-Studio/ for details." ) raise return AMDSharkBackend(model, inputs, options) diff --git a/amdshark/examples/amdshark_training/stable_diffusion/README.md b/amdshark/examples/amdshark_training/stable_diffusion/README.md index d182c7a4..5a57e3de 100644 --- a/amdshark/examples/amdshark_training/stable_diffusion/README.md +++ b/amdshark/examples/amdshark_training/stable_diffusion/README.md @@ -23,7 +23,7 @@ pip install accelerate transformers ftfy Please cherry-pick this branch of torch-mlir: https://github.com/vivekkhandelwal1/torch-mlir/tree/sd-ops and build it locally. You can find the instructions for using locally build Torch-MLIR, -here: https://github.com/nod-ai/AMDSHARK-Studio#how-to-use-your-locally-built-iree--torch-mlir-with-amdshark +here: https://github.com/nod-ai/AMD-SHARK-Studio#how-to-use-your-locally-built-iree--torch-mlir-with-amdshark ## Run the Stable diffusion fine tuning diff --git a/amdshark/iree_utils/_common.py b/amdshark/iree_utils/_common.py index 8b8089b5..5b4e1e2c 100644 --- a/amdshark/iree_utils/_common.py +++ b/amdshark/iree_utils/_common.py @@ -157,7 +157,7 @@ def device_driver_info(device): f"Required drivers for {device} not found. {device_driver_err_map[device]['debug']} " f"Please install the required drivers{device_driver_err_map[device]['solution']} " f"For further assistance please reach out to the community on discord [https://discord.com/invite/RUqY2h2s9u]" - f" and/or file a bug at https://github.com/nod-ai/AMDSHARK-Studio/issues" + f" and/or file a bug at https://github.com/nod-ai/AMD-SHARK-Studio/issues" ) return err_msg else: diff --git a/apps/amdshark_studio/api/utils.py b/apps/amdshark_studio/api/utils.py index 58446d88..7a029b6e 100644 --- a/apps/amdshark_studio/api/utils.py +++ b/apps/amdshark_studio/api/utils.py @@ -192,7 +192,7 @@ def get_rocm_target_chip(device_str): if key in device_str: return rocm_chip_map[key] raise AssertionError( - f"Device {device_str} not recognized. Please file an issue at https://github.com/nod-ai/AMDSHARK-Studio/issues." + f"Device {device_str} not recognized. Please file an issue at https://github.com/nod-ai/AMD-SHARK-Studio/issues." ) diff --git a/docs/amdshark_iree_profiling.md b/docs/amdshark_iree_profiling.md index fec041a3..ade9b8b5 100644 --- a/docs/amdshark_iree_profiling.md +++ b/docs/amdshark_iree_profiling.md @@ -1,9 +1,9 @@ # Overview This document is intended to provide a starting point for profiling with AMDSHARK/IREE. At it's core -[AMDSHARK](https://github.com/nod-ai/AMDSHARK-Studio/tree/main/tank) is a python API that links the MLIR lowerings from various +[AMDSHARK](https://github.com/nod-ai/AMD-SHARK-Studio/tree/main/tank) is a python API that links the MLIR lowerings from various frameworks + frontends (e.g. PyTorch -> Torch-MLIR) with the compiler + runtime offered by IREE. More information -on model coverage and framework support can be found [here](https://github.com/nod-ai/AMDSHARK-Studio/tree/main/tank). The intended +on model coverage and framework support can be found [here](https://github.com/nod-ai/AMD-SHARK-Studio/tree/main/tank). The intended use case for AMDSHARK is for compilation and deployment of performant state of the art AI models. ![image](https://user-images.githubusercontent.com/22101546/217151219-9bb184a3-cfb9-4788-bb7e-5b502953525c.png) diff --git a/docs/amdshark_sd_blender.md b/docs/amdshark_sd_blender.md index 871c9dc2..0d1ccdde 100644 --- a/docs/amdshark_sd_blender.md +++ b/docs/amdshark_sd_blender.md @@ -6,7 +6,7 @@ We currently make use of the [AI-Render Plugin](https://github.com/benrugg/AI-Re ## Setup AMDSHARK and prerequisites: - * Download the latest AMDSHARK SD webui .exe from [here](https://github.com/nod-ai/AMDSHARK-Studio/releases) or follow instructions on the [README](https://github.com/nod-ai/AMDSHARK-Studio#readme) + * Download the latest AMDSHARK SD webui .exe from [here](https://github.com/nod-ai/AMD-SHARK-Studio/releases) or follow instructions on the [README](https://github.com/nod-ai/AMD-SHARK-Studio#readme) * Once you have the .exe where you would like AMDSHARK to install, run the .exe from terminal/PowerShell with the `--api` flag: ``` ## Run the .exe in API mode: diff --git a/docs/amdshark_sd_koboldcpp.md b/docs/amdshark_sd_koboldcpp.md index 8f6e28fe..561188f8 100644 --- a/docs/amdshark_sd_koboldcpp.md +++ b/docs/amdshark_sd_koboldcpp.md @@ -20,8 +20,8 @@ This does mean however, that on a brand new fresh install of AMDSHARK that has n ## Setup AMDSHARK and prerequisites: - * Make sure you have suitable drivers for your graphics card installed. See the prerequisties section of the [README](https://github.com/nod-ai/AMDSHARK-Studio#readme). - * Download the latest AMDSHARK studio .exe from [here](https://github.com/nod-ai/AMDSHARK-Studio/releases) or follow the instructions in the [README](https://github.com/nod-ai/AMDSHARK-Studio#readme) for an advanced, Linux or Mac install. + * Make sure you have suitable drivers for your graphics card installed. See the prerequisties section of the [README](https://github.com/nod-ai/AMD-SHARK-Studio#readme). + * Download the latest AMDSHARK studio .exe from [here](https://github.com/nod-ai/AMD-SHARK-Studio/releases) or follow the instructions in the [README](https://github.com/nod-ai/AMD-SHARK-Studio#readme) for an advanced, Linux or Mac install. * Run AMDSHARK from terminal/PowerShell with the `--api` flag. Since koboldcpp also expects both CORS support and the image generator to be running on port `7860` rather than AMDSHARK default of `8080`, also include both the `--api_accept_origin` flag with a suitable origin (use `="*"` to enable all origins) and `--server_port=7860` on the command line. (See the if you want to run AMDSHARK on a different port) ```powershell diff --git a/requirements.txt b/requirements.txt index 4aad4e09..de104555 100644 --- a/requirements.txt +++ b/requirements.txt @@ -8,7 +8,7 @@ wheel torch==2.3.0 iree-turbine @ git+https://github.com/iree-org/iree-turbine.git@main -turbine-models @ git+https://github.com/nod-ai/AMDSHARK-ModelDev.git@main#subdirectory=models +turbine-models @ git+https://github.com/nod-ai/AMD-SHARK-ModelDev.git@main#subdirectory=models diffusers @ git+https://github.com/nod-ai/diffusers@0.29.0.dev0-amdshark brevitas @ git+https://github.com/Xilinx/brevitas.git@6695e8df7f6a2c7715b9ed69c4b78157376bb60b diff --git a/setup.py b/setup.py index 1e040067..631e558b 100644 --- a/setup.py +++ b/setup.py @@ -20,8 +20,8 @@ setup( long_description=long_description, long_description_content_type="text/markdown", project_urls={ - "Code": "https://github.com/nod-ai/AMDSHARK", - "Bug Tracker": "https://github.com/nod-ai/AMDSHARK-Studio/issues", + "Code": "https://github.com/nod-ai/AMD-SHARK-Studio", + "Bug Tracker": "https://github.com/nod-ai/AMD-SHARK-Studio/issues", }, classifiers=[ "Programming Language :: Python :: 3", diff --git a/tank/README.md b/tank/README.md index 8cb66d95..90ca530c 100644 --- a/tank/README.md +++ b/tank/README.md @@ -142,7 +142,7 @@ For more information refer to [MODEL TRACKING SHEET](https://docs.google.com/spr ### Run all model tests on CPU/GPU/VULKAN/Metal -For a list of models included in our pytest model suite, see https://github.com/nod-ai/AMDSHARK-Studio/blob/main/tank/all_models.csv +For a list of models included in our pytest model suite, see https://github.com/nod-ai/AMD-SHARK-Studio/blob/main/tank/all_models.csv ```shell pytest tank/test_models.py diff --git a/tank/all_models.csv b/tank/all_models.csv index 4efcc5b3..0685ef0a 100644 --- a/tank/all_models.csv +++ b/tank/all_models.csv @@ -1,19 +1,19 @@ bert-base-uncased,linalg,torch,1e-2,1e-3,default,None,False,False,False,"","" bert-large-uncased,linalg,torch,1e-2,1e-3,default,None,False,False,False,"","" facebook/deit-small-distilled-patch16-224,linalg,torch,1e-2,1e-3,default,nhcw-nhwc,False,False,False,"Fails during iree-compile.","" -google/vit-base-patch16-224,linalg,torch,1e-2,1e-3,default,nhcw-nhwc,True,True,True,"https://github.com/nod-ai/AMDSHARK/issues/311","" -microsoft/beit-base-patch16-224-pt22k-ft22k,linalg,torch,1e-2,1e-3,default,nhcw-nhwc,False,False,False,"https://github.com/nod-ai/AMDSHARK/issues/390","macos" +google/vit-base-patch16-224,linalg,torch,1e-2,1e-3,default,nhcw-nhwc,True,True,True,"https://github.com/nod-ai/AMD-SHARK-Studio/issues/311","" +microsoft/beit-base-patch16-224-pt22k-ft22k,linalg,torch,1e-2,1e-3,default,nhcw-nhwc,False,False,False,"https://github.com/nod-ai/AMD-SHARK-Studio/issues/390","macos" microsoft/MiniLM-L12-H384-uncased,linalg,torch,1e-2,1e-3,default,None,False,False,False,"","" -google/mobilebert-uncased,linalg,torch,1e-2,1e-3,default,None,False,False,False,"https://github.com/nod-ai/AMDSHARK/issues/344","macos" -mobilenet_v3_small,linalg,torch,1e-1,1e-2,default,nhcw-nhwc,False,False,False,"https://github.com/nod-ai/AMDSHARK/issues/388, https://github.com/nod-ai/AMDSHARK/issues/1487","macos" -nvidia/mit-b0,linalg,torch,1e-2,1e-3,default,None,True,True,True,"https://github.com/nod-ai/AMDSHARK/issues/343,https://github.com/nod-ai/AMDSHARK/issues/1487","macos" +google/mobilebert-uncased,linalg,torch,1e-2,1e-3,default,None,False,False,False,"https://github.com/nod-ai/AMD-SHARK-Studio/issues/344","macos" +mobilenet_v3_small,linalg,torch,1e-1,1e-2,default,nhcw-nhwc,False,False,False,"https://github.com/nod-ai/AMD-SHARK-Studio/issues/388, https://github.com/nod-ai/AMD-SHARK-Studio/issues/1487","macos" +nvidia/mit-b0,linalg,torch,1e-2,1e-3,default,None,True,True,True,"https://github.com/nod-ai/AMD-SHARK-Studio/issues/343,https://github.com/nod-ai/AMD-SHARK-Studio/issues/1487","macos" resnet101,linalg,torch,1e-2,1e-3,default,nhcw-nhwc/img2col,True,True,True,"","macos" resnet18,linalg,torch,1e-2,1e-3,default,None,True,True,True,"","macos" resnet50,linalg,torch,1e-2,1e-3,default,nhcw-nhwc,False,False,False,"","macos" squeezenet1_0,linalg,torch,1e-2,1e-3,default,nhcw-nhwc,False,False,False,"","macos" wide_resnet50_2,linalg,torch,1e-2,1e-3,default,nhcw-nhwc/img2col,True,True,True,"","macos" mnasnet1_0,linalg,torch,1e-2,1e-3,default,nhcw-nhwc,False,False,False,"","macos" -efficientnet_b0,linalg,torch,1e-2,1e-3,default,nhcw-nhwc,True,True,True,"https://github.com/nod-ai/AMDSHARK/issues/1487","macos" -efficientnet_b7,linalg,torch,1e-2,1e-3,default,nhcw-nhwc,True,True,True,"https://github.com/nod-ai/AMDSHARK/issues/1487","macos" +efficientnet_b0,linalg,torch,1e-2,1e-3,default,nhcw-nhwc,True,True,True,"https://github.com/nod-ai/AMD-SHARK-Studio/issues/1487","macos" +efficientnet_b7,linalg,torch,1e-2,1e-3,default,nhcw-nhwc,True,True,True,"https://github.com/nod-ai/AMD-SHARK-Studio/issues/1487","macos" t5-base,linalg,torch,1e-2,1e-3,default,None,True,True,True,"","macos" t5-large,linalg,torch,1e-2,1e-3,default,None,True,True,True,"","macos" diff --git a/tank/generate_amdsharktank.py b/tank/generate_amdsharktank.py index 22ae2175..a776aa32 100644 --- a/tank/generate_amdsharktank.py +++ b/tank/generate_amdsharktank.py @@ -1,5 +1,5 @@ # Lint as: python3 -"""AMDSHARK Tank""" +"""AMD-SHARK Tank""" # python generate_amdsharktank.py, you have to give a csv tile with [model_name, model_download_url] # will generate local amdshark tank folder like this: # /AMDSHARK @@ -191,7 +191,7 @@ if __name__ == "__main__": # type=lambda x: is_valid_file(x), # default="./tank/torch_model_list.csv", # help="""Contains the file with torch_model name and args. - # Please see: https://github.com/nod-ai/AMDSHARK-Studio/blob/main/tank/torch_model_list.csv""", + # Please see: https://github.com/nod-ai/AMD-SHARK-Studio/blob/main/tank/torch_model_list.csv""", # ) # parser.add_argument( # "--ci_tank_dir", diff --git a/tank/test_models.py b/tank/test_models.py index 648b40c9..c45dcb54 100644 --- a/tank/test_models.py +++ b/tank/test_models.py @@ -396,7 +396,7 @@ class AMDSharkModuleTest(unittest.TestCase): and device == "rocm" ): pytest.xfail( - reason="iree-compile buffer limit issue: https://github.com/nod-ai/AMDSHARK-Studio/issues/475" + reason="iree-compile buffer limit issue: https://github.com/nod-ai/AMD-SHARK-Studio/issues/475" ) if ( config["model_name"] @@ -407,7 +407,7 @@ class AMDSharkModuleTest(unittest.TestCase): and device == "rocm" ): pytest.xfail( - reason="Numerics issues: https://github.com/nod-ai/AMDSHARK-Studio/issues/476" + reason="Numerics issues: https://github.com/nod-ai/AMD-SHARK-Studio/issues/476" ) if config["framework"] == "tf" and self.module_tester.batch_size != 1: pytest.xfail(