feat(exe): add specs and launch scripts for Windows EXE bundle (#305)

api/entry.py

@@ -0,0 +1,29 @@
+import os
+
+def script_method(fn, _rcb=None):
+    return fn
+
+def script(obj, optimize=True, _frames_up=0, _rcb=None):
+    return obj
+
+import torch.jit
+torch.jit.script_method = script_method
+torch.jit.script = script
+
+import multiprocessing
+
+if __name__ == '__main__':
+    multiprocessing.freeze_support()
+    try:
+        from onnx_web.main import main
+        app, pool = main()
+        print("starting workers")
+        pool.start()
+        print("starting flask")
+        app.run("0.0.0.0", 5000, debug=False)
+        input("press the any key")
+        pool.join()
+    except Exception as e:
+        print(e)
+    finally:
+        os.system("pause")

docs/setup-guide.md

@@ -0,0 +1,51 @@
+# Setup Guide
+
+This guide covers the setup process for onnx-web, including downloading the Windows bundle.
+
+## Contents
+
+- [Setup Guide](#setup-guide)
+  - [Contents](#contents)
+  - [Windows all-in-one bundle](#windows-all-in-one-bundle)
+  - [Windows Python installer](#windows-python-installer)
+  - [Windows Store Python](#windows-store-python)
+
+## Windows all-in-one bundle
+
+1. Download the latest ZIP file from TODO
+2. Find the ZIP file and `Extract All` to a memorable folder
+3. Open the folder where you extracted the files
+   1. You can add models to the `models` folder
+   2. Your images will be in the `outputs` folder, along with files containing the parameters used to generate them
+4. Run the local server using one of the `onnx-web-*.bat` scripts
+   1. Run `onnx-web-half.bat` if you are using a GPU and you have < 12GB of VRAM
+      - `-half` mode is compatible with both AMD and Nvidia GPUs
+      - `-half` mode is not compatible with CPU mode
+   2. Run `onnx-web-full.bat` if you are using CPU mode or if you have >= 16GB of VRAM
+      - Try the `onnx-web-half.bat` script if you encounter out-of-memory errors or generating images is very slow
+5. Wait for the models to be downloaded and converted
+   1. Most models are distributed in PyTorch format and need to be converted into ONNX format
+   2. This only happens once for each model and takes a few minutes
+6. Open one of the URLs shown in the logs in your browser
+   1. This will typically be http://127.0.0.1:5000?api=http://127.0.0.1:5000
+   2. If you running the server on a different PC and not accessing it from a browser on the same system, use that PC's
+      IP address instead of 127.0.0.1
+   3. Any modern browser should work, including Chrome, Edge, and Firefox
+   4. Mobile browsers also work, but have stricter mixed-content policies
+
+## Windows Python installer
+
+1. Install Git
+2. Install Python 3.10
+3. Open command prompt
+4. Run one of the `setup-*.bat` scripts
+   1. Run `setup-amd.bat` if you are using an AMD GPU and DirectML
+   2. Run `setup-nvidia.bat` if you are using an Nvidia GPU and CUDA
+   3. Run `setup-cpu.bat` if you are planning on only using CPU mode
+5. In the future, run `launch.bat`
+   1. You should only need to run the setup script once
+   2. If you encounter any errors with Python imports, run the setup script again
+
+## Windows Store Python
+
+TODO

exe/README.txt

@@ -0,0 +1,16 @@
+# onnx-web Windows bundle
+
+This is the Windows all-in-one bundle for onnx-web, a tool for running Stable Diffusion using ONNX hardware
+acceleration: https://github.com/ssube/onnx-web.
+
+Please check the setup guide for the latest instructions, this may be an older version: https://github.com/ssube/onnx-web/blob/main/docs/setup-guide.md#windows-all-in-one-bundle
+
+## Running onnx-web
+
+You can run the local server using one of the setup scripts, onnx-web-full.bat or onnx-web-half.bat. Use the
+onnx-web-half.bat script if you are using a GPU and have < 12GB of VRAM. Use the onnx-web-full.bat script if
+you are using CPU mode or if you have >= 16GB of VRAM.
+
+The user interface should be available in your browser at http://127.0.0.1:5000?api=http://127.0.0.1:5000. If
+your PC uses a different IP address or you are running the server on one PC and using it from another, use that IP
+address instead.

exe/onnx-web-full.bat

@@ -0,0 +1,7 @@
+set ONNX_WEB_BASE_PATH=%~dp0
+set ONNX_WEB_BUNDLE_PATH=%ONNX_WEB_BASE_PATH%\client
+set ONNX_WEB_MODEL_PATH=%ONNX_WEB_BASE_PATH%\models
+set ONNX_WEB_OUTPUT_PATH=%ONNX_WEB_BASE_PATH%\outputs
+
+@echo Launching onnx-web in full-precision mode...
+server\onnx-web.exe --diffusion --correction --upscaling

exe/onnx-web-half.bat

@@ -0,0 +1,9 @@
+set ONNX_WEB_BASE_PATH=%~dp0
+set ONNX_WEB_BUNDLE_PATH=%ONNX_WEB_BASE_PATH%\client
+set ONNX_WEB_MODEL_PATH=%ONNX_WEB_BASE_PATH%\models
+set ONNX_WEB_OUTPUT_PATH=%ONNX_WEB_BASE_PATH%\outputs
+
+set ONNX_WEB_BLOCK_PLATFORMS=cpu
+
+@echo Launching onnx-web in half-precision mode...
+server\onnx-web.exe --diffusion --correction --upscaling --half

exe/win10.directml.dir.spec

@@ -0,0 +1,99 @@
+# -*- mode: python ; coding: utf-8 -*-
+
+from PyInstaller.utils.hooks import collect_data_files, copy_metadata
+
+import sys
+
+sys.setrecursionlimit(sys.getrecursionlimit() * 5)
+
+
+block_cipher = None
+
+dynamic_packages = [
+    "filelock",
+    "numpy",
+    "packaging",
+    "onnxruntime",
+    "onnxruntime-directml",
+    "regex",
+    "requests",
+    "tokenizers",
+    "tqdm",
+]
+metadatas = [copy_metadata(pkg) for pkg in dynamic_packages]
+metadatas = sum(metadatas, [])
+
+datas = [
+    collect_data_files("basicsr", include_py_files=True, includes=[
+        "archs/**",
+        "data/**",
+        "losses/**",
+        "models/**",
+        "utils/**",
+    ]),
+    collect_data_files("realesrgan", include_py_files=True, includes=[
+        "archs/**",
+        "data/**",
+        "losses/**",
+        "models/**",
+        "utils/**",
+    ]),
+    collect_data_files("onnxruntime", include_py_files=True, includes=[
+        "transformers/**",
+    ]),
+    collect_data_files("transformers", include_py_files=True, includes=[
+        "**",
+    ]),
+]
+datas = sum(datas, [])
+
+a = Analysis(
+    ['../api/entry.py'],
+    pathex=[],
+    binaries=[],
+    datas=[
+        *metadatas,
+        *datas,
+    ],
+    hiddenimports=['onnxruntime', 'onnxruntime-directml', 'tqdm'],
+    hookspath=[],
+    hooksconfig={},
+    runtime_hooks=[],
+    excludes=[],
+    win_no_prefer_redirects=False,
+    win_private_assemblies=False,
+    cipher=block_cipher,
+    noarchive=False,
+)
+pyz = PYZ(a.pure, a.zipped_data, cipher=block_cipher)
+
+exe = EXE(
+    pyz,
+    a.scripts,
+    [],
+    name='onnx-web',
+    debug=False,
+    exclude_binaries=True,
+    bootloader_ignore_signals=False,
+    strip=False,
+    upx=True,
+    upx_exclude=[],
+    runtime_tmpdir=None,
+    console=True,
+    disable_windowed_traceback=False,
+    argv_emulation=False,
+    target_arch=None,
+    codesign_identity=None,
+    entitlements_file=None,
+)
+
+dir = COLLECT(
+    exe,
+    a.binaries,
+    a.zipfiles,
+    a.datas,
+    strip=False,
+    upx=True,
+    upx_exclude=[],
+    name='server',
+)

exe/win10.directml.file.spec

@@ -0,0 +1,80 @@
+# -*- mode: python ; coding: utf-8 -*-
+
+from PyInstaller.utils.hooks import collect_data_files, copy_metadata
+
+import sys
+
+sys.setrecursionlimit(sys.getrecursionlimit() * 5)
+
+
+block_cipher = None
+
+dynamic_packages = [
+    "filelock",
+    "numpy",
+    "packaging",
+    "onnxruntime",
+    "onnxruntime-directml",
+    "regex",
+    "requests",
+    "tokenizers",
+    "tqdm",
+]
+metadatas = [copy_metadata(pkg) for pkg in dynamic_packages]
+metadatas = sum(metadatas, [])
+
+datas = [
+    collect_data_files("basicsr", include_py_files=True, includes=[
+        "archs/**",
+        "data/**",
+        "losses/**",
+        "models/**",
+        "utils/**",
+    ]),
+    collect_data_files("transformers", include_py_files=True, includes=[
+        "**",
+    ]),
+]
+datas = sum(datas, [])
+
+a = Analysis(
+    ['../api/entry.py'],
+    pathex=[],
+    binaries=[],
+    datas=[
+        *metadatas,
+        *datas,
+    ],
+    hiddenimports=['onnxruntime', 'onnxruntime-directml', 'tqdm'],
+    hookspath=[],
+    hooksconfig={},
+    runtime_hooks=[],
+    excludes=[],
+    win_no_prefer_redirects=False,
+    win_private_assemblies=False,
+    cipher=block_cipher,
+    noarchive=False,
+)
+pyz = PYZ(a.pure, a.zipped_data, cipher=block_cipher)
+
+exe = EXE(
+    pyz,
+    a.scripts,
+    a.binaries,
+    a.zipfiles,
+    a.datas,
+    [],
+    name='onnx-web',
+    debug=False,
+    bootloader_ignore_signals=False,
+    strip=False,
+    upx=True,
+    upx_exclude=[],
+    runtime_tmpdir=None,
+    console=True,
+    disable_windowed_traceback=False,
+    argv_emulation=False,
+    target_arch=None,
+    codesign_identity=None,
+    entitlements_file=None,
+)