autoforge 1.9.6__tar.gz → 2.0.0__tar.gz

{autoforge-1.9.6/src/AutoForge.egg-info → autoforge-2.0.0}/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: AutoForge
-Version: 1.9.6
-Summary: AutoForge is a Python tool for generating 3D printed layered models from an input image.
+Version: 2.0.0
+Summary: AutoForge - 3D color images made easy
 Author-email: Hendric Voss <hvoss@techfak.uni-bielefeld.de>
 License: CC BY-NC-SA 4.0
 Requires-Python: >=3.10
@@ -28,17 +28,22 @@ Requires-Dist: python-multipart>=0.0.12
 Requires-Dist: aiofiles>=24.1.0
 Requires-Dist: websockets>=15.0
 Requires-Dist: pydantic-settings>=2.0.0
+Requires-Dist: posthog>=3.7.0
+Provides-Extra: test
+Requires-Dist: pytest>=7.4; extra == "test"
+Requires-Dist: pytest-cov; extra == "test"
+Requires-Dist: hypothesis; extra == "test"
+Requires-Dist: httpx>=0.27; extra == "test"
 
 # AutoForge
 
 AutoForge is a Python tool for generating 3D printed layered models from an input image. Using a learned optimization strategy with a Gumbel softmax formulation, AutoForge assigns materials per layer and produces both a discretized composite image and a 3D-printable STL file. It also generates swap instructions to guide the printer through material changes during a multi-material print. 
 
-**TLDR:** It uses a picture to generate a 3D layer image that you can print with a 3d printer. Similar to [Hueforge](https://shop.thehueforge.com/), but without the manual work (and without the artistic control).
+**TLDR:** It uses a picture to generate a 3D layer image that you can print with a 3d printer. Similar to [Hueforge](https://shop.thehueforge.com/), but without the manual work.
+
+![2.0_webui.png](images/webui/2.0_webui.png)
 
 
-## You can now run Autoforge for free in your browser thanks to [Huggingface space support](https://huggingface.co/spaces/hvoss-techfak/Autoforge).
-This includes the option to run it locally if you have a powerful pc and don't want to limit yourself to the Huggingface computing limits. \
-For this simply go to the [Huggingface](https://huggingface.co/spaces/hvoss-techfak/Autoforge) space and pull the docker container for this project (upper right corner -> three dots -> "run locally")
 
 ## Example
 All examples use only the 27 BambuLab Basic PLA filaments, currently available in Hueforge 0.9.0, the background color is set to black.
@@ -72,9 +77,41 @@ The pruning is set to a maximum of 8 color and 20 swaps, so each image uses at m
 - **Live Visualization**: (Optional) Displays live composite images during the optimization process.
 - **Hueforge export**: Outputs a project file that can be opened with hueforge.
 
-## Manual Installation
+## Web UI: One-Click Install & Run
+
+
+The easiest way to use AutoForge is the web UI, a local app with drag-and-drop image upload, a filament library, live sliders for adjusting colors after optimization, and pruning. No command-line arguments needed.
+
+1. **Clone this repository** (or download and extract the ZIP from the green "Code" button on GitHub).
+2. **Install**, from the project folder:
+   - Linux/macOS: `./install.sh`
+   - Windows: double-click `install.bat` (or run it from a terminal)
+
+   This installs [`uv`](https://docs.astral.sh/uv/) (a fast Python package manager) if you don't already have it, installs all Python dependencies, and builds the web UI. You'll need [Node.js](https://nodejs.org/) installed for that last step, the installer will tell you if it's missing.
+3. **Run**:
+   - Linux/macOS: `./run_webui.sh`
+   - Windows: double-click `run_webui.bat`
+
+   This starts the server and opens the web UI in your browser automatically (usually at `http://localhost:8000`).
+
+   To make some parts of the picture come out closer than the rest (a face, the eyes, lettering), click **Focus** in the image panel and paint over them; a slider sets how much more they count (2× to 100×, default 10×). After a run, the **Differences** view shows where the print strays furthest from the picture.
+
+   If HueForge is installed on the same computer, the web UI finds your HueForge personal filament library and offers to import it once, on its first start.
 
-To install AutoForge, simply install the current version from PyPI:
+   The web UI sends anonymous usage telemetry to the project via [PostHog](https://posthog.com/) by default, to notify me of problems and any bugs. This includes crash reports (unhandled errors from both the browser frontend and the backend server, with the error type, message, and stack trace) so bugs can get fixed faster. No image data, filament data, or personal information is sent. To disable it, pass `--no-telemetry` (e.g. `./run_webui.sh --no-telemetry` / `run_webui.bat --no-telemetry`), or set `AUTOFORGE_WEBUI_TELEMETRY_ENABLED=false` permanently in your environment.
+
+4. **Update to the latest release** whenever you want, from the project folder:
+   - Linux/macOS: `./update.sh`
+   - Windows: double-click `update.bat`
+
+   This checks GitHub for a newer release and, if one exists, pulls it and reinstalls dependencies for you. Add `--check` to only check without applying it (e.g. `./update.sh --check`).
+
+If you have problems running the code on your GPU, please refer to the [Pytorch Homepage](https://pytorch.org/) for help. \
+CUDA, ROCm, and MPS (Apple Metal) are supported, but you need to install the correct version of pytorch for your system — `install.sh`/`install.bat` install whatever `uv sync` resolves by default, so swap in a GPU-specific PyTorch build afterwards if you need one. (`install.sh` handles older NVIDIA GPUs automatically, see below.)
+
+## Manual Installation (CLI only)
+
+If you just want the command-line tool (no web UI), install the current version from PyPI:
 ```bash
    pip install -U autoforge
 ```
@@ -146,6 +183,8 @@ This will generate separate STL files for each color, allowing you to print face
 - `--nozzle_diameter` Diameter of the printer nozzle in millimeters (default: 0.4).  
   **Note:** Details smaller than half this value will be ignored.
 - `--early_stopping` Number of steps without improvement before stopping (default: 10000).
+- `--priority_mask` *(Optional)* Path to a greyscale image the size of the input that marks the parts that matter most: white areas are matched more closely, black areas still count, just less.
+- `--priority_mask_strength` How many times more a white pixel of `--priority_mask` counts than a black one (default: 10; must be at least 1). Raise it if the marked areas still come out off, lower it for a gentler nudge.
 
 - `--flatforge` Enable FlatForge mode to generate separate STL files for each color (default: False).  
   **Note:** FlatForge creates flat prints where each color is its own STL file, allowing face-down printing for smooth, resin-like finishes. Requires a multi-material printer (AMS, MMU, or tool changer).
@@ -163,7 +202,8 @@ This will generate separate STL files for each color, allowing you to print face
 - `--pruning_max_swaps` Max number of swaps allowed after pruning (default: 100).
 - `--pruning_max_layer` Max number of layers allowed after pruning (default: 75).
 - `--random_seed` Random seed for reproducibility (default: 0 (disabled)).
-- `--mps` Flag to use the Metal Performance Shaders (MPS) backend if available.
+- `--device` Torch device to run on, e.g. `cuda`, `cuda:1`, `mps`, `cpu`. Defaults to auto-detection: CUDA/ROCm first, then Apple Metal (MPS), then CPU. Can also be set with the `AUTOFORGE_DEVICE` environment variable.
+- `--mps` *Deprecated* — Apple Metal is now detected automatically, so this flag is no longer needed. It still works, and forces MPS on a machine that also exposes a CUDA GPU; prefer `--device mps`.
 - `--no-spike-removal` Disable spike removal for the final STL (not recommended).
 
 - `--tensorboard` Flag to enable TensorBoard logging.
@@ -192,6 +232,10 @@ After running, the following files will be created in your specified output fold
   
   *Note:* FlatForge generates multiple STL files that align perfectly when loaded together in your slicer, creating a solid rectangular print with each color as a separate object.
 
+## You can now run Autoforge for free in your browser thanks to [Huggingface space support](https://huggingface.co/spaces/hvoss-techfak/Autoforge).
+This includes the option to run it locally if you have a powerful pc and don't want to limit yourself to the Huggingface computing limits. \
+For this simply go to the [Huggingface](https://huggingface.co/spaces/hvoss-techfak/Autoforge) space and pull the docker container for this project (upper right corner -> three dots -> "run locally")
+
 ## Development
 
 To have a "nightly" version of the repository or have live updating changes during development please do the following: 
@@ -221,6 +265,11 @@ pip install torch torchvision torchaudio --index-url https://download.pytorch.or
 AutoForge © 2025 by Hendric Voss is licensed under [CC BY-NC-SA 4.0](https://creativecommons.org/licenses/by-nc-sa/4.0/).
 The software is provided as-is and comes with no warranty or guarantee of support.
 
+The above license applies to the software itself. How you use the generated files and prints is entirely up to you. If you want to print them for your friends or family, that's great. If you want to sell them, that's fine by me too. 
+
+From a licensing standpoint, this means that the prints you create are entirely subject to the [MIT License](https://opensource.org/licenses/MIT), and you can do whatever you like with them. The only thing the MIT License does not give you is a warranty, and it also frees me from any liability with regard to your 3D prints. Otherwise, do whatever you like.
+
+I would love to see what you have done with the software, so it would be great if you could send me a link to your work (even if it's just for selling). However, this is not necessary if you do not want to.
 
 ## Acknowledgements
 

autoforge-1.9.6/PKG-INFO → autoforge-2.0.0/README.md

@@ -1,44 +1,12 @@
-Metadata-Version: 2.4
-Name: AutoForge
-Version: 1.9.6
-Summary: AutoForge is a Python tool for generating 3D printed layered models from an input image.
-Author-email: Hendric Voss <hvoss@techfak.uni-bielefeld.de>
-License: CC BY-NC-SA 4.0
-Requires-Python: >=3.10
-Description-Content-Type: text/markdown
-Requires-Dist: numpy
-Requires-Dist: torch>=2.9.1
-Requires-Dist: torchvision>=0.21.0
-Requires-Dist: transformers>=4.48.1
-Requires-Dist: opencv-python>=4.2.0.32
-Requires-Dist: matplotlib>=3.3.0
-Requires-Dist: configargparse>=1.2.3
-Requires-Dist: tqdm>=4.50.0
-Requires-Dist: pandas>=1.1.0
-Requires-Dist: scikit-image
-Requires-Dist: scikit-learn
-Requires-Dist: tensorboard
-Requires-Dist: pillow>=11.1.0
-Requires-Dist: shapely>=2.0.7
-Requires-Dist: trimesh>=4.6.4
-Requires-Dist: joblib>=1.4.2
-Requires-Dist: uvicorn[standard]>=0.44.0
-Requires-Dist: fastapi>=0.115.0
-Requires-Dist: python-multipart>=0.0.12
-Requires-Dist: aiofiles>=24.1.0
-Requires-Dist: websockets>=15.0
-Requires-Dist: pydantic-settings>=2.0.0
-
 # AutoForge
 
 AutoForge is a Python tool for generating 3D printed layered models from an input image. Using a learned optimization strategy with a Gumbel softmax formulation, AutoForge assigns materials per layer and produces both a discretized composite image and a 3D-printable STL file. It also generates swap instructions to guide the printer through material changes during a multi-material print. 
 
-**TLDR:** It uses a picture to generate a 3D layer image that you can print with a 3d printer. Similar to [Hueforge](https://shop.thehueforge.com/), but without the manual work (and without the artistic control).
+**TLDR:** It uses a picture to generate a 3D layer image that you can print with a 3d printer. Similar to [Hueforge](https://shop.thehueforge.com/), but without the manual work.
+
+![2.0_webui.png](images/webui/2.0_webui.png)
 
 
-## You can now run Autoforge for free in your browser thanks to [Huggingface space support](https://huggingface.co/spaces/hvoss-techfak/Autoforge).
-This includes the option to run it locally if you have a powerful pc and don't want to limit yourself to the Huggingface computing limits. \
-For this simply go to the [Huggingface](https://huggingface.co/spaces/hvoss-techfak/Autoforge) space and pull the docker container for this project (upper right corner -> three dots -> "run locally")
 
 ## Example
 All examples use only the 27 BambuLab Basic PLA filaments, currently available in Hueforge 0.9.0, the background color is set to black.
@@ -72,9 +40,41 @@ The pruning is set to a maximum of 8 color and 20 swaps, so each image uses at m
 - **Live Visualization**: (Optional) Displays live composite images during the optimization process.
 - **Hueforge export**: Outputs a project file that can be opened with hueforge.
 
-## Manual Installation
+## Web UI: One-Click Install & Run
+
+
+The easiest way to use AutoForge is the web UI, a local app with drag-and-drop image upload, a filament library, live sliders for adjusting colors after optimization, and pruning. No command-line arguments needed.
+
+1. **Clone this repository** (or download and extract the ZIP from the green "Code" button on GitHub).
+2. **Install**, from the project folder:
+   - Linux/macOS: `./install.sh`
+   - Windows: double-click `install.bat` (or run it from a terminal)
+
+   This installs [`uv`](https://docs.astral.sh/uv/) (a fast Python package manager) if you don't already have it, installs all Python dependencies, and builds the web UI. You'll need [Node.js](https://nodejs.org/) installed for that last step, the installer will tell you if it's missing.
+3. **Run**:
+   - Linux/macOS: `./run_webui.sh`
+   - Windows: double-click `run_webui.bat`
+
+   This starts the server and opens the web UI in your browser automatically (usually at `http://localhost:8000`).
+
+   To make some parts of the picture come out closer than the rest (a face, the eyes, lettering), click **Focus** in the image panel and paint over them; a slider sets how much more they count (2× to 100×, default 10×). After a run, the **Differences** view shows where the print strays furthest from the picture.
+
+   If HueForge is installed on the same computer, the web UI finds your HueForge personal filament library and offers to import it once, on its first start.
 
-To install AutoForge, simply install the current version from PyPI:
+   The web UI sends anonymous usage telemetry to the project via [PostHog](https://posthog.com/) by default, to notify me of problems and any bugs. This includes crash reports (unhandled errors from both the browser frontend and the backend server, with the error type, message, and stack trace) so bugs can get fixed faster. No image data, filament data, or personal information is sent. To disable it, pass `--no-telemetry` (e.g. `./run_webui.sh --no-telemetry` / `run_webui.bat --no-telemetry`), or set `AUTOFORGE_WEBUI_TELEMETRY_ENABLED=false` permanently in your environment.
+
+4. **Update to the latest release** whenever you want, from the project folder:
+   - Linux/macOS: `./update.sh`
+   - Windows: double-click `update.bat`
+
+   This checks GitHub for a newer release and, if one exists, pulls it and reinstalls dependencies for you. Add `--check` to only check without applying it (e.g. `./update.sh --check`).
+
+If you have problems running the code on your GPU, please refer to the [Pytorch Homepage](https://pytorch.org/) for help. \
+CUDA, ROCm, and MPS (Apple Metal) are supported, but you need to install the correct version of pytorch for your system — `install.sh`/`install.bat` install whatever `uv sync` resolves by default, so swap in a GPU-specific PyTorch build afterwards if you need one. (`install.sh` handles older NVIDIA GPUs automatically, see below.)
+
+## Manual Installation (CLI only)
+
+If you just want the command-line tool (no web UI), install the current version from PyPI:
 ```bash
    pip install -U autoforge
 ```
@@ -146,6 +146,8 @@ This will generate separate STL files for each color, allowing you to print face
 - `--nozzle_diameter` Diameter of the printer nozzle in millimeters (default: 0.4).  
   **Note:** Details smaller than half this value will be ignored.
 - `--early_stopping` Number of steps without improvement before stopping (default: 10000).
+- `--priority_mask` *(Optional)* Path to a greyscale image the size of the input that marks the parts that matter most: white areas are matched more closely, black areas still count, just less.
+- `--priority_mask_strength` How many times more a white pixel of `--priority_mask` counts than a black one (default: 10; must be at least 1). Raise it if the marked areas still come out off, lower it for a gentler nudge.
 
 - `--flatforge` Enable FlatForge mode to generate separate STL files for each color (default: False).  
   **Note:** FlatForge creates flat prints where each color is its own STL file, allowing face-down printing for smooth, resin-like finishes. Requires a multi-material printer (AMS, MMU, or tool changer).
@@ -163,7 +165,8 @@ This will generate separate STL files for each color, allowing you to print face
 - `--pruning_max_swaps` Max number of swaps allowed after pruning (default: 100).
 - `--pruning_max_layer` Max number of layers allowed after pruning (default: 75).
 - `--random_seed` Random seed for reproducibility (default: 0 (disabled)).
-- `--mps` Flag to use the Metal Performance Shaders (MPS) backend if available.
+- `--device` Torch device to run on, e.g. `cuda`, `cuda:1`, `mps`, `cpu`. Defaults to auto-detection: CUDA/ROCm first, then Apple Metal (MPS), then CPU. Can also be set with the `AUTOFORGE_DEVICE` environment variable.
+- `--mps` *Deprecated* — Apple Metal is now detected automatically, so this flag is no longer needed. It still works, and forces MPS on a machine that also exposes a CUDA GPU; prefer `--device mps`.
 - `--no-spike-removal` Disable spike removal for the final STL (not recommended).
 
 - `--tensorboard` Flag to enable TensorBoard logging.
@@ -192,6 +195,10 @@ After running, the following files will be created in your specified output fold
   
   *Note:* FlatForge generates multiple STL files that align perfectly when loaded together in your slicer, creating a solid rectangular print with each color as a separate object.
 
+## You can now run Autoforge for free in your browser thanks to [Huggingface space support](https://huggingface.co/spaces/hvoss-techfak/Autoforge).
+This includes the option to run it locally if you have a powerful pc and don't want to limit yourself to the Huggingface computing limits. \
+For this simply go to the [Huggingface](https://huggingface.co/spaces/hvoss-techfak/Autoforge) space and pull the docker container for this project (upper right corner -> three dots -> "run locally")
+
 ## Development
 
 To have a "nightly" version of the repository or have live updating changes during development please do the following: 
@@ -221,6 +228,11 @@ pip install torch torchvision torchaudio --index-url https://download.pytorch.or
 AutoForge © 2025 by Hendric Voss is licensed under [CC BY-NC-SA 4.0](https://creativecommons.org/licenses/by-nc-sa/4.0/).
 The software is provided as-is and comes with no warranty or guarantee of support.
 
+The above license applies to the software itself. How you use the generated files and prints is entirely up to you. If you want to print them for your friends or family, that's great. If you want to sell them, that's fine by me too. 
+
+From a licensing standpoint, this means that the prints you create are entirely subject to the [MIT License](https://opensource.org/licenses/MIT), and you can do whatever you like with them. The only thing the MIT License does not give you is a warranty, and it also frees me from any liability with regard to your 3D prints. Otherwise, do whatever you like.
+
+I would love to see what you have done with the software, so it would be great if you could send me a link to your work (even if it's just for selling). However, this is not necessary if you do not want to.
 
 ## Acknowledgements
 

{autoforge-1.9.6 → autoforge-2.0.0}/pyproject.toml

@@ -4,8 +4,8 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "AutoForge"
-version="1.9.6"
-description = "AutoForge is a Python tool for generating 3D printed layered models from an input image."
+version="2.0.0"
+description = "AutoForge - 3D color images made easy"
 authors = [{ name = "Hendric Voss", email = "hvoss@techfak.uni-bielefeld.de" }]
 readme = "README.md"
 requires-python = ">=3.10"
@@ -33,6 +33,18 @@ dependencies = [
   "aiofiles>=24.1.0",
   "websockets>=15.0",
   "pydantic-settings>=2.0.0",
+  "posthog>=3.7.0",
+]
+
+[project.optional-dependencies]
+# httpx isn't a runtime dependency of the app itself, but fastapi's
+# TestClient needs it installed, and tests/webui/test_review_2026_09_23.py
+# uses httpx.AsyncClient/ASGITransport directly to test concurrent requests.
+test = [
+  "pytest>=7.4",
+  "pytest-cov",
+  "hypothesis",
+  "httpx>=0.27",
 ]
 
 [tool.setuptools]

autoforge-1.9.6/README.md → autoforge-2.0.0/src/AutoForge.egg-info/PKG-INFO

@@ -1,13 +1,49 @@
+Metadata-Version: 2.4
+Name: AutoForge
+Version: 2.0.0
+Summary: AutoForge - 3D color images made easy
+Author-email: Hendric Voss <hvoss@techfak.uni-bielefeld.de>
+License: CC BY-NC-SA 4.0
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+Requires-Dist: numpy
+Requires-Dist: torch>=2.9.1
+Requires-Dist: torchvision>=0.21.0
+Requires-Dist: transformers>=4.48.1
+Requires-Dist: opencv-python>=4.2.0.32
+Requires-Dist: matplotlib>=3.3.0
+Requires-Dist: configargparse>=1.2.3
+Requires-Dist: tqdm>=4.50.0
+Requires-Dist: pandas>=1.1.0
+Requires-Dist: scikit-image
+Requires-Dist: scikit-learn
+Requires-Dist: tensorboard
+Requires-Dist: pillow>=11.1.0
+Requires-Dist: shapely>=2.0.7
+Requires-Dist: trimesh>=4.6.4
+Requires-Dist: joblib>=1.4.2
+Requires-Dist: uvicorn[standard]>=0.44.0
+Requires-Dist: fastapi>=0.115.0
+Requires-Dist: python-multipart>=0.0.12
+Requires-Dist: aiofiles>=24.1.0
+Requires-Dist: websockets>=15.0
+Requires-Dist: pydantic-settings>=2.0.0
+Requires-Dist: posthog>=3.7.0
+Provides-Extra: test
+Requires-Dist: pytest>=7.4; extra == "test"
+Requires-Dist: pytest-cov; extra == "test"
+Requires-Dist: hypothesis; extra == "test"
+Requires-Dist: httpx>=0.27; extra == "test"
+
 # AutoForge
 
 AutoForge is a Python tool for generating 3D printed layered models from an input image. Using a learned optimization strategy with a Gumbel softmax formulation, AutoForge assigns materials per layer and produces both a discretized composite image and a 3D-printable STL file. It also generates swap instructions to guide the printer through material changes during a multi-material print. 
 
-**TLDR:** It uses a picture to generate a 3D layer image that you can print with a 3d printer. Similar to [Hueforge](https://shop.thehueforge.com/), but without the manual work (and without the artistic control).
+**TLDR:** It uses a picture to generate a 3D layer image that you can print with a 3d printer. Similar to [Hueforge](https://shop.thehueforge.com/), but without the manual work.
+
+![2.0_webui.png](images/webui/2.0_webui.png)
 
 
-## You can now run Autoforge for free in your browser thanks to [Huggingface space support](https://huggingface.co/spaces/hvoss-techfak/Autoforge).
-This includes the option to run it locally if you have a powerful pc and don't want to limit yourself to the Huggingface computing limits. \
-For this simply go to the [Huggingface](https://huggingface.co/spaces/hvoss-techfak/Autoforge) space and pull the docker container for this project (upper right corner -> three dots -> "run locally")
 
 ## Example
 All examples use only the 27 BambuLab Basic PLA filaments, currently available in Hueforge 0.9.0, the background color is set to black.
@@ -41,9 +77,41 @@ The pruning is set to a maximum of 8 color and 20 swaps, so each image uses at m
 - **Live Visualization**: (Optional) Displays live composite images during the optimization process.
 - **Hueforge export**: Outputs a project file that can be opened with hueforge.
 
-## Manual Installation
+## Web UI: One-Click Install & Run
+
+
+The easiest way to use AutoForge is the web UI, a local app with drag-and-drop image upload, a filament library, live sliders for adjusting colors after optimization, and pruning. No command-line arguments needed.
+
+1. **Clone this repository** (or download and extract the ZIP from the green "Code" button on GitHub).
+2. **Install**, from the project folder:
+   - Linux/macOS: `./install.sh`
+   - Windows: double-click `install.bat` (or run it from a terminal)
+
+   This installs [`uv`](https://docs.astral.sh/uv/) (a fast Python package manager) if you don't already have it, installs all Python dependencies, and builds the web UI. You'll need [Node.js](https://nodejs.org/) installed for that last step, the installer will tell you if it's missing.
+3. **Run**:
+   - Linux/macOS: `./run_webui.sh`
+   - Windows: double-click `run_webui.bat`
+
+   This starts the server and opens the web UI in your browser automatically (usually at `http://localhost:8000`).
+
+   To make some parts of the picture come out closer than the rest (a face, the eyes, lettering), click **Focus** in the image panel and paint over them; a slider sets how much more they count (2× to 100×, default 10×). After a run, the **Differences** view shows where the print strays furthest from the picture.
+
+   If HueForge is installed on the same computer, the web UI finds your HueForge personal filament library and offers to import it once, on its first start.
 
-To install AutoForge, simply install the current version from PyPI:
+   The web UI sends anonymous usage telemetry to the project via [PostHog](https://posthog.com/) by default, to notify me of problems and any bugs. This includes crash reports (unhandled errors from both the browser frontend and the backend server, with the error type, message, and stack trace) so bugs can get fixed faster. No image data, filament data, or personal information is sent. To disable it, pass `--no-telemetry` (e.g. `./run_webui.sh --no-telemetry` / `run_webui.bat --no-telemetry`), or set `AUTOFORGE_WEBUI_TELEMETRY_ENABLED=false` permanently in your environment.
+
+4. **Update to the latest release** whenever you want, from the project folder:
+   - Linux/macOS: `./update.sh`
+   - Windows: double-click `update.bat`
+
+   This checks GitHub for a newer release and, if one exists, pulls it and reinstalls dependencies for you. Add `--check` to only check without applying it (e.g. `./update.sh --check`).
+
+If you have problems running the code on your GPU, please refer to the [Pytorch Homepage](https://pytorch.org/) for help. \
+CUDA, ROCm, and MPS (Apple Metal) are supported, but you need to install the correct version of pytorch for your system — `install.sh`/`install.bat` install whatever `uv sync` resolves by default, so swap in a GPU-specific PyTorch build afterwards if you need one. (`install.sh` handles older NVIDIA GPUs automatically, see below.)
+
+## Manual Installation (CLI only)
+
+If you just want the command-line tool (no web UI), install the current version from PyPI:
 ```bash
    pip install -U autoforge
 ```
@@ -115,6 +183,8 @@ This will generate separate STL files for each color, allowing you to print face
 - `--nozzle_diameter` Diameter of the printer nozzle in millimeters (default: 0.4).  
   **Note:** Details smaller than half this value will be ignored.
 - `--early_stopping` Number of steps without improvement before stopping (default: 10000).
+- `--priority_mask` *(Optional)* Path to a greyscale image the size of the input that marks the parts that matter most: white areas are matched more closely, black areas still count, just less.
+- `--priority_mask_strength` How many times more a white pixel of `--priority_mask` counts than a black one (default: 10; must be at least 1). Raise it if the marked areas still come out off, lower it for a gentler nudge.
 
 - `--flatforge` Enable FlatForge mode to generate separate STL files for each color (default: False).  
   **Note:** FlatForge creates flat prints where each color is its own STL file, allowing face-down printing for smooth, resin-like finishes. Requires a multi-material printer (AMS, MMU, or tool changer).
@@ -132,7 +202,8 @@ This will generate separate STL files for each color, allowing you to print face
 - `--pruning_max_swaps` Max number of swaps allowed after pruning (default: 100).
 - `--pruning_max_layer` Max number of layers allowed after pruning (default: 75).
 - `--random_seed` Random seed for reproducibility (default: 0 (disabled)).
-- `--mps` Flag to use the Metal Performance Shaders (MPS) backend if available.
+- `--device` Torch device to run on, e.g. `cuda`, `cuda:1`, `mps`, `cpu`. Defaults to auto-detection: CUDA/ROCm first, then Apple Metal (MPS), then CPU. Can also be set with the `AUTOFORGE_DEVICE` environment variable.
+- `--mps` *Deprecated* — Apple Metal is now detected automatically, so this flag is no longer needed. It still works, and forces MPS on a machine that also exposes a CUDA GPU; prefer `--device mps`.
 - `--no-spike-removal` Disable spike removal for the final STL (not recommended).
 
 - `--tensorboard` Flag to enable TensorBoard logging.
@@ -161,6 +232,10 @@ After running, the following files will be created in your specified output fold
   
   *Note:* FlatForge generates multiple STL files that align perfectly when loaded together in your slicer, creating a solid rectangular print with each color as a separate object.
 
+## You can now run Autoforge for free in your browser thanks to [Huggingface space support](https://huggingface.co/spaces/hvoss-techfak/Autoforge).
+This includes the option to run it locally if you have a powerful pc and don't want to limit yourself to the Huggingface computing limits. \
+For this simply go to the [Huggingface](https://huggingface.co/spaces/hvoss-techfak/Autoforge) space and pull the docker container for this project (upper right corner -> three dots -> "run locally")
+
 ## Development
 
 To have a "nightly" version of the repository or have live updating changes during development please do the following: 
@@ -190,6 +265,11 @@ pip install torch torchvision torchaudio --index-url https://download.pytorch.or
 AutoForge © 2025 by Hendric Voss is licensed under [CC BY-NC-SA 4.0](https://creativecommons.org/licenses/by-nc-sa/4.0/).
 The software is provided as-is and comes with no warranty or guarantee of support.
 
+The above license applies to the software itself. How you use the generated files and prints is entirely up to you. If you want to print them for your friends or family, that's great. If you want to sell them, that's fine by me too. 
+
+From a licensing standpoint, this means that the prints you create are entirely subject to the [MIT License](https://opensource.org/licenses/MIT), and you can do whatever you like with them. The only thing the MIT License does not give you is a warranty, and it also frees me from any liability with regard to your 3D prints. Otherwise, do whatever you like.
+
+I would love to see what you have done with the software, so it would be great if you could send me a link to your work (even if it's just for selling). However, this is not necessary if you do not want to.
 
 ## Acknowledgements
 

{autoforge-1.9.6 → autoforge-2.0.0}/src/AutoForge.egg-info/SOURCES.txt

@@ -12,6 +12,7 @@ src/autoforge/auto_forge.py
 src/autoforge/priority_mask_editor.py
 src/autoforge/Helper/AmpUtils.py
 src/autoforge/Helper/CAdamW.py
+src/autoforge/Helper/DeviceUtils.py
 src/autoforge/Helper/FilamentHelper.py
 src/autoforge/Helper/ImageHelper.py
 src/autoforge/Helper/OptimizerHelper.py
@@ -54,33 +55,41 @@ src/autoforge/webui/api/system.py
 src/autoforge/webui/api/ws.py
 src/autoforge/webui/helpers/__init__.py
 src/autoforge/webui/helpers/colored_mesh.py
+src/autoforge/webui/helpers/gpu_memory.py
+src/autoforge/webui/helpers/mesh_persist.py
 src/autoforge/webui/helpers/pipeline_runner.py
 src/autoforge/webui/helpers/slider_render.py
 src/autoforge/webui/helpers/sliders.py
+src/autoforge/webui/helpers/telemetry.py
 src/autoforge/webui/services/__init__.py
 src/autoforge/webui/services/filament_service.py
 src/autoforge/webui/services/image_service.py
 src/autoforge/webui/services/optimization_service.py
 src/autoforge/webui/services/project_service.py
-tests/test_amp_autocast_probe.py
+tests/test_amp_utils.py
+tests/test_auto_forge_cli.py
+tests/test_bug_regressions.py
 tests/test_cadamw.py
-tests/test_christofides_height_map.py
 tests/test_christofides_heightmap.py
+tests/test_composite_image.py
 tests/test_depth_adjusted_height_map.py
 tests/test_depth_estimate_height_map.py
+tests/test_device_utils.py
 tests/test_end_to_end_optimizer.py
-tests/test_filament_and_output.py
+tests/test_fasttsp_heightmap.py
 tests/test_filament_helper.py
 tests/test_flatforge.py
-tests/test_heightmap_init.py
 tests/test_image_helper.py
 tests/test_loss_functions.py
-tests/test_optimizer_helper.py
 tests/test_optimizer_module.py
-tests/test_optimizer_smoke.py
+tests/test_optimizer_training.py
+tests/test_other_helper.py
 tests/test_output_helper.py
-tests/test_output_helper_extended.py
-tests/test_perception_and_pruning.py
+tests/test_perception_loss.py
+tests/test_pipeline_smoke.py
 tests/test_property_based.py
+tests/test_prune_repeatability.py
 tests/test_pruning_helper.py
-tests/test_remove_height_spikes.py
+tests/test_remove_height_spikes.py
+tests/test_webui_backend.py
+tests/test_webui_regressions.py

{autoforge-1.9.6 → autoforge-2.0.0}/src/AutoForge.egg-info/requires.txt

@@ -20,3 +20,10 @@ python-multipart>=0.0.12
 aiofiles>=24.1.0
 websockets>=15.0
 pydantic-settings>=2.0.0
+posthog>=3.7.0
+
+[test]
+pytest>=7.4
+pytest-cov
+hypothesis
+httpx>=0.27

{autoforge-1.9.6 → autoforge-2.0.0}/src/autoforge/Helper/AmpUtils.py

@@ -6,6 +6,8 @@ from typing import Optional, Tuple, Dict
 import os
 import torch
 
+from autoforge.Helper.DeviceUtils import is_rocm
+
 # Cache the selection per device key so CPU/CUDA/MPS don't interfere
 _SELECTED_MAP: Dict[str, Tuple[Optional[torch.dtype], str]] = {}
 
@@ -20,11 +22,14 @@ def _device_key(device: torch.device) -> str:
 def _probe_backward(device: torch.device, dtype: torch.dtype) -> bool:
     """Try a tiny forward+backward under autocast for given dtype on the device.
 
-    Returns True if successful, False if a RuntimeError occurs (e.g., oneDNN bf16/f16 backward not supported).
+    Returns True if successful, False if a RuntimeError occurs (e.g., oneDNN
+    bf16/f16 backward not supported, or an op with no reduced-precision Metal
+    kernel). The probe is the whole point: rather than maintaining a table of
+    which backend/driver/CPU-ISA combination supports what, we run the thing
+    and see. That is what makes CUDA, ROCm, Metal and CPU share one code path
+    here.
     """
     try:
-        if device.type == "mps":
-            return False  # no stable autocast on MPS yet
         x = torch.randn(8, 8, device=device, dtype=torch.float32, requires_grad=False)
         w = torch.randn(8, 8, device=device, dtype=torch.float32, requires_grad=True)
         target = torch.randn(8, 8, device=device, dtype=torch.float32)
@@ -40,6 +45,33 @@ def _probe_backward(device: torch.device, dtype: torch.dtype) -> bool:
         return False
 
 
+def _cuda_bf16_supported() -> bool:
+    try:
+        return bool(torch.cuda.is_bf16_supported())
+    except Exception:
+        return False
+
+
+def _cuda_fp16_supported(device: torch.device) -> bool:
+    """fp16 capability check, tolerant of backends without CUDA arch numbers."""
+    if is_rocm():
+        return True
+    try:
+        major, minor = torch.cuda.get_device_capability(device)
+    except Exception:
+        return False
+    return (major * 10 + minor) >= 53  # CC 5.3 or newer
+
+
+def _enable_tf32() -> None:
+    """Turn on TF32 matmuls where the backend has them (Ampere+; no-op on ROCm)."""
+    try:
+        torch.backends.cuda.matmul.allow_tf32 = True
+        torch.backends.cudnn.allow_tf32 = True
+    except Exception:
+        pass
+
+
 def _select_dtype_for_device(device: torch.device) -> Tuple[Optional[torch.dtype], str]:
     """Decide the best autocast dtype for the current device, probing backward support.
 
@@ -59,27 +91,30 @@ def _select_dtype_for_device(device: torch.device) -> Tuple[Optional[torch.dtype
         print("AUTOFORGE_AMP=bf16: disabling bf16 as probe failed")
         return None, "forced bf16 but probe failed"
     if mode in {"fp16", "float16", "half"}:
-        if device.type == "cuda" and _probe_backward(device, torch.float16):
+        if device.type in {"cuda", "mps"} and _probe_backward(device, torch.float16):
             return torch.float16, "forced f16"
-        print("AUTOFORGE_AMP=fp16: disabling fp16 as probe failed or non-CUDA")
-        return None, "forced f16 but probe failed or non-CUDA"
+        print("AUTOFORGE_AMP=fp16: disabling fp16 as probe failed or unsupported device")
+        return None, "forced f16 but probe failed or unsupported device"
 
     # Automatic selection
     if device.type == "cuda":
+        # ROCm reports itself through the torch.cuda namespace, so the whole
+        # branch is shared; only the log label and the compute-capability
+        # lookup (which is CUDA-specific numbering) differ.
+        label = "ROCm" if is_rocm() else "CUDA"
         # Prefer bf16 if it actually works end-to-end
-        if torch.cuda.is_bf16_supported() and _probe_backward(device, torch.bfloat16):
-            print("Using bf16 autocast on CUDA device")
-            return torch.bfloat16, "cuda bf16"
-        # Next try fp16 if the GPU supports it
-        major, minor = torch.cuda.get_device_capability(device)
-        if (major * 10 + minor) >= 53 and _probe_backward(device, torch.float16):
-            print("Using fp16 autocast on CUDA device")
-            return torch.float16, "cuda f16"
+        if _cuda_bf16_supported() and _probe_backward(device, torch.bfloat16):
+            print(f"Using bf16 autocast on {label} device")
+            return torch.bfloat16, f"{label.lower()} bf16"
+        # Next try fp16 if the GPU supports it. Compute capability is a CUDA
+        # concept; on ROCm every supported GPU does fp16, so skip the gate.
+        if _cuda_fp16_supported(device) and _probe_backward(device, torch.float16):
+            print(f"Using fp16 autocast on {label} device")
+            return torch.float16, f"{label.lower()} f16"
         # Fall back to fp32; enable TF32 to speed up matmuls if available
-        torch.backends.cuda.matmul.allow_tf32 = True
-        torch.backends.cudnn.allow_tf32 = True
-        print("Falling back to fp32 autocast on CUDA device with TF32 enabled")
-        return None, "cuda fp32 (tf32 on)"
+        _enable_tf32()
+        print(f"Falling back to fp32 autocast on {label} device with TF32 enabled")
+        return None, f"{label.lower()} fp32 (tf32 on)"
 
     if device.type == "cpu":
         # Some CPUs (AVX512) can do bf16 forward/backward with oneDNN, but many cannot.
@@ -92,7 +127,11 @@ def _select_dtype_for_device(device: torch.device) -> Tuple[Optional[torch.dtype
         return None, "cpu fp32"
 
     if device.type == "mps":
-        print("MPS device detected: autocast not reliably supported, using fp32")
+        # Metal does have an autocast implementation, but its reduced-precision
+        # coverage is uneven across ops and a missing kernel shows up as a
+        # wrong result rather than an error - so fp32 stays the default and
+        # AUTOFORGE_AMP=fp16/bf16 is the opt-in (handled above, probe-gated).
+        print("MPS device detected: defaulting to fp32 (set AUTOFORGE_AMP=fp16 to try half precision)")
         return None, "mps fp32"
 
     # Default: no autocast