autoforge 1.7.4__tar.gz → 1.7.6__tar.gz

{autoforge-1.7.4 → autoforge-1.7.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: AutoForge
-Version: 1.7.4
+Version: 1.7.6
 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
@@ -26,7 +26,7 @@ Requires-Dist: joblib>=1.4.2
 
 # 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. \
+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).
 
@@ -97,11 +97,16 @@ The script is run from the command line and accepts several arguments. Below is
 > **Note:** You will need [Hueforge](https://shop.thehueforge.com/) installed to export your filament CSV.  
 > To get your CSV file, simply go to the "Filaments" menu in Hueforge, click the export button, select your filaments, and export them as a CSV file.
 
-
 ```bash
 autoforge --input_image path/to/input_image.jpg --csv_file path/to/materials.csv 
 ```
 
+We also support json files. If you want to use your personal Hueforge library (found in %APPDATA%\HueForge\Filaments\personal_library.json) you can run the command with:
+```bash
+autoforge --input_image path/to/input_image.jpg --json_file %APPDATA%\HueForge\Filaments\personal_library.json
+```
+
+
 > If you want to limit the amount of colors the program can use, you can set these as command line arguments. \
 > For Example: 8 colors and a maximum of 20 swaps:
 
@@ -116,7 +121,9 @@ autoforge --input_image path/to/input_image.jpg --csv_file path/to/materials.csv
 
 
 - `--input_image` **(Required)** Path to the input image.
-- `--csv_file` **(Required)** Path to the CSV file containing material data. The CSV should include columns for the brand, name, color (hex code), and TD values.
+- `--csv_file` Path to the CSV file containing material data. The CSV should include columns for the brand, name, color (hex code), and TD values.
+- `--json_file` Path to the json file containing material data. \
+ **Note:** Either a csv or json file has to be given.
 
 - `--output_folder` Folder where output files will be saved (default: `output`).
 - `--iterations` Number of optimization iterations (default: 2000).

{autoforge-1.7.4 → autoforge-1.7.6}/README.md

@@ -1,6 +1,6 @@
 # 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. \
+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).
 
@@ -71,11 +71,16 @@ The script is run from the command line and accepts several arguments. Below is
 > **Note:** You will need [Hueforge](https://shop.thehueforge.com/) installed to export your filament CSV.  
 > To get your CSV file, simply go to the "Filaments" menu in Hueforge, click the export button, select your filaments, and export them as a CSV file.
 
-
 ```bash
 autoforge --input_image path/to/input_image.jpg --csv_file path/to/materials.csv 
 ```
 
+We also support json files. If you want to use your personal Hueforge library (found in %APPDATA%\HueForge\Filaments\personal_library.json) you can run the command with:
+```bash
+autoforge --input_image path/to/input_image.jpg --json_file %APPDATA%\HueForge\Filaments\personal_library.json
+```
+
+
 > If you want to limit the amount of colors the program can use, you can set these as command line arguments. \
 > For Example: 8 colors and a maximum of 20 swaps:
 
@@ -90,7 +95,9 @@ autoforge --input_image path/to/input_image.jpg --csv_file path/to/materials.csv
 
 
 - `--input_image` **(Required)** Path to the input image.
-- `--csv_file` **(Required)** Path to the CSV file containing material data. The CSV should include columns for the brand, name, color (hex code), and TD values.
+- `--csv_file` Path to the CSV file containing material data. The CSV should include columns for the brand, name, color (hex code), and TD values.
+- `--json_file` Path to the json file containing material data. \
+ **Note:** Either a csv or json file has to be given.
 
 - `--output_folder` Folder where output files will be saved (default: `output`).
 - `--iterations` Number of optimization iterations (default: 2000).

{autoforge-1.7.4 → autoforge-1.7.6}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "AutoForge"
-version="1.7.4"
+version="1.7.6"
 description = "AutoForge is a Python tool for generating 3D printed layered models from an input image."
 authors = [{ name = "Hendric Voss", email = "hvoss@techfak.uni-bielefeld.de" }]
 readme = "README.md"

{autoforge-1.7.4 → autoforge-1.7.6}/src/AutoForge.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: AutoForge
-Version: 1.7.4
+Version: 1.7.6
 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
@@ -26,7 +26,7 @@ Requires-Dist: joblib>=1.4.2
 
 # 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. \
+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).
 
@@ -97,11 +97,16 @@ The script is run from the command line and accepts several arguments. Below is
 > **Note:** You will need [Hueforge](https://shop.thehueforge.com/) installed to export your filament CSV.  
 > To get your CSV file, simply go to the "Filaments" menu in Hueforge, click the export button, select your filaments, and export them as a CSV file.
 
-
 ```bash
 autoforge --input_image path/to/input_image.jpg --csv_file path/to/materials.csv 
 ```
 
+We also support json files. If you want to use your personal Hueforge library (found in %APPDATA%\HueForge\Filaments\personal_library.json) you can run the command with:
+```bash
+autoforge --input_image path/to/input_image.jpg --json_file %APPDATA%\HueForge\Filaments\personal_library.json
+```
+
+
 > If you want to limit the amount of colors the program can use, you can set these as command line arguments. \
 > For Example: 8 colors and a maximum of 20 swaps:
 
@@ -116,7 +121,9 @@ autoforge --input_image path/to/input_image.jpg --csv_file path/to/materials.csv
 
 
 - `--input_image` **(Required)** Path to the input image.
-- `--csv_file` **(Required)** Path to the CSV file containing material data. The CSV should include columns for the brand, name, color (hex code), and TD values.
+- `--csv_file` Path to the CSV file containing material data. The CSV should include columns for the brand, name, color (hex code), and TD values.
+- `--json_file` Path to the json file containing material data. \
+ **Note:** Either a csv or json file has to be given.
 
 - `--output_folder` Folder where output files will be saved (default: `output`).
 - `--iterations` Number of optimization iterations (default: 2000).

{autoforge-1.7.4 → autoforge-1.7.6}/src/autoforge/Helper/FilamentHelper.py

@@ -1,11 +1,14 @@
+import sys
+import traceback
 import uuid
 
 import numpy as np
 import pandas as pd
 import torch
+import json
 
 
-def load_materials(csv_filename):
+def load_materials(args):
     """
     Load material data from a CSV file.
 
@@ -19,13 +22,13 @@ def load_materials(csv_filename):
             - material_names (list): List of material names.
             - colors_list (list): List of color hex strings.
     """
-    df = pd.read_csv(csv_filename)
+    df = load_materials_pandas(args)
     material_names = [
-        brand + " - " + name
-        for brand, name in zip(df["Brand"].tolist(), df[" Name"].tolist())
+        str(brand) + " - " + str(name)
+        for brand, name in zip(df["Brand"].tolist(), df["Name"].tolist())
     ]
-    material_TDs = (df[" TD"].astype(float)).to_numpy()
-    colors_list = df[" Color"].tolist()
+    material_TDs = (df["Transmissivity"].astype(float)).to_numpy()
+    colors_list = df["Color"].tolist()
     # Use float64 for material colors.
     material_colors = np.array(
         [hex_to_rgb(color) for color in colors_list], dtype=np.float64
@@ -33,8 +36,38 @@ def load_materials(csv_filename):
     material_TDs = np.array(material_TDs, dtype=np.float64)
     return material_colors, material_TDs, material_names, colors_list
 
-
-def load_materials_data(csv_filename):
+def load_materials_pandas(args):
+    csv_filename = args.csv_file
+    json_filename = args.json_file
+
+    if csv_filename != "":
+        try:
+            df = pd.read_csv(csv_filename)
+        except Exception as e:
+            traceback.print_exc()
+            print("Error reading filament CSV file:", e)
+            sys.exit(1)
+        # rename all columns that start with a whitespace
+        df.columns = [col.strip() for col in df.columns]
+        # if TD in columns rename to Transmissivity
+        if "TD" in df.columns:
+            df.rename(columns={"TD": "Transmissivity"}, inplace=True)
+    else:
+        # read json
+        with open(json_filename, "r") as f:
+            data = json.load(f)
+        if "Filaments" in data.keys():
+            data = data["Filaments"]
+        else:
+            print(
+                "Warning: No Filaments key found in JSON data. We can't use this json data."
+            )
+            sys.exit(1)
+        # list to dataframe
+        df = pd.DataFrame(data)
+    return df
+
+def load_materials_data(args):
     """
     Load the full material data from the CSV file.
 
@@ -45,7 +78,7 @@ def load_materials_data(csv_filename):
         list: A list of dictionaries (one per material) with keys such as
               "Brand", "Type", "Color", "Name", "TD", "Owned", and "Uuid".
     """
-    df = pd.read_csv(csv_filename)
+    df = load_materials_pandas(args)
     # Use a consistent key naming. For example, convert 'TD' to 'Transmissivity' and 'Uuid' to 'uuid'
     records = df.to_dict(orient="records")
     return records
@@ -79,7 +112,7 @@ def extract_colors_from_swatches(swatch_data):
         out[(brand, name)] = (color, td)
 
     # convert to the same format as the hueforge csv files
-    material_names = [brand + " - " + name for (brand, name) in out.keys()]
+    material_names = [str(brand) + " - " + str(name) for (brand, name) in out.keys()]
     material_colors = np.array(
         [hex_to_rgb("#" + color) for color, _ in out.values()], dtype=np.float64
     )

{autoforge-1.7.4 → autoforge-1.7.6}/src/autoforge/Helper/OutputHelper.py

@@ -81,7 +81,7 @@ def generate_project_file(
     background_layers = int(args.background_height / args.layer_height)
 
     # Load full material data from CSV
-    material_data = load_materials_data(csv_filename)
+    material_data = load_materials_data(args)
 
     # Extract the swap points from the discrete solution
     filament_indices, slider_values = extract_filament_swaps(
@@ -95,17 +95,17 @@ def generate_project_file(
         mat = material_data[idx]
         filament_entry = {
             "Brand": mat["Brand"],
-            "Color": mat[" Color"],
-            "Name": mat[" Name"],
+            "Color": mat["Color"],
+            "Name": mat["Name"],
             # Convert Owned to a boolean (in case it is read as a string)
-            "Owned": str(mat[" Owned"]).strip().lower() == "true"
-            if " Owned" in mat
+            "Owned": str(mat["Owned"]).strip().lower() == "true"
+            if "Owned" in mat
             else False,
-            "Transmissivity": float(mat[" TD"])
-            if not float(mat[" TD"]).is_integer()
-            else int(mat[" TD"]),
-            "Type": mat[" Type"] if " Type" in mat else "PLA",
-            "uuid": mat[" Uuid"] if " Uuid" in mat else str(uuid.uuid4()),
+            "Transmissivity": float(mat["Transmissivity"])
+            if not float(mat["Transmissivity"]).is_integer()
+            else int(mat["Transmissivity"]),
+            "Type": mat["Type"] if "Type" in mat else "PLA",
+            "uuid": mat["Uuid"] if "Uuid" in mat else str(uuid.uuid4()),
         }
         filament_set.append(filament_entry)
 

{autoforge-1.7.4 → autoforge-1.7.6}/src/autoforge/auto_forge.py

@@ -8,6 +8,7 @@ import torch
 import numpy as np
 from tqdm import tqdm
 
+
 from autoforge.Helper.FilamentHelper import hex_to_rgb, load_materials
 from autoforge.Helper.Heightmaps.ChristofidesHeightMap import (
     run_init_threads,
@@ -43,9 +44,15 @@ def main():
     parser.add_argument(
         "--csv_file",
         type=str,
-        required=True,
+        default="",
         help="Path to CSV file with material data",
     )
+    parser.add_argument(
+        "--json_file",
+        type=str,
+        default="",
+        help="Path to json file with material data",
+    )
     parser.add_argument(
         "--output_folder", type=str, default="output", help="Folder to write outputs"
     )
@@ -128,9 +135,17 @@ def main():
     parser.add_argument(
         "--stl_output_size",
         type=int,
-        default=150,
+        default=200,
         help="Size of the longest dimension of the output STL file in mm",
     )
+
+    parser.add_argument(
+        "--processing_size",
+        type=int,
+        default=100,
+        help="Internal image processing size in mm",
+    )
+
     parser.add_argument(
         "--nozzle_diameter",
         type=float,
@@ -266,6 +281,11 @@ def main():
     if args.num_init_cluster_layers == -1:
         args.num_init_cluster_layers = args.max_layers // 2
 
+    # check if csv or json is given
+    if args.csv_file == "" and args.json_file == "":
+        print("Error: No CSV or JSON file given. Please provide one of them.")
+        sys.exit(1)
+
     if torch.cuda.is_available():
         device = torch.device("cuda")
     elif args.mps and torch.backends.mps.is_available():
@@ -288,9 +308,12 @@ def main():
         print(f"Error: Input image '{args.input_image}' not found.", file=sys.stderr)
         sys.exit(1)
 
-    if not os.path.exists(args.csv_file):
+    if args.csv_file != "" and not os.path.exists(args.csv_file):
         print(f"Error: CSV file '{args.csv_file}' not found.", file=sys.stderr)
         sys.exit(1)
+    if args.json_file != "" and not os.path.exists(args.json_file):
+        print(f"Error: Json file '{args.json_file}' not found.", file=sys.stderr)
+        sys.exit(1)
 
     random_seed = args.random_seed
     if random_seed == 0:
@@ -303,9 +326,7 @@ def main():
     background = torch.tensor(bgr_tuple, dtype=torch.float32, device=device)
 
     # Load materials
-    material_colors_np, material_TDs_np, material_names, _ = load_materials(
-        args.csv_file
-    )
+    material_colors_np, material_TDs_np, material_names, _ = load_materials(args)
     material_colors = torch.tensor(
         material_colors_np, dtype=torch.float32, device=device
     )
@@ -314,7 +335,10 @@ def main():
     # Read input image
     img = imread(args.input_image, cv2.IMREAD_UNCHANGED)
     computed_output_size = int(round(args.stl_output_size * 2 / args.nozzle_diameter))
-    print(f"Computed solving pixel size: {computed_output_size}")
+    computed_processing_size = int(
+        round(args.processing_size * 2 / args.nozzle_diameter)
+    )
+    print(f"Computed solving pixel size: {computed_processing_size}")
     alpha = None
     if img.shape[2] == 4:
         # Extract alpha channel
@@ -328,6 +352,8 @@ def main():
 
     # Convert image from BGR to RGB
     img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
+    processing_img_np = resize_image(img, computed_processing_size)
+    processing_img = torch.tensor(processing_img_np, dtype=torch.float32, device=device)
 
     # For the final resolution
     output_img_np = resize_image(img, computed_output_size)
@@ -366,6 +392,10 @@ def main():
     if alpha is not None:
         pixel_height_logits_init[alpha < 128] = -13.815512
 
+    # resize the pixel_height_logits to the processing_size
+    pixel_height_logits_process = resize_image(
+        np.expand_dims(pixel_height_logits_init, -1), computed_processing_size
+    )
     # VGG Perceptual Loss
     # We currently disable this as it is not used in the optimization.
     perception_loss_module = None  # MultiLayerVGGPerceptualLoss().to(device).eval()
@@ -373,8 +403,8 @@ def main():
     # Create an optimizer instance
     optimizer = FilamentOptimizer(
         args=args,
-        target=output_target,
-        pixel_height_logits_init=pixel_height_logits_init,
+        target=processing_img,
+        pixel_height_logits_init=pixel_height_logits_process,
         global_logits_init=global_logits_init,
         material_colors=material_colors,
         material_TDs=material_TDs,
@@ -410,82 +440,99 @@ def main():
 
     post_opt_step = 0
 
-    optimizer.log_to_tensorboard(
-        interval=1, namespace="post_opt", step=(post_opt_step := post_opt_step + 1)
-    )
-
-    if args.perform_pruning:
-        optimizer.prune(
-            max_colors_allowed=args.pruning_max_colors,
-            max_swaps_allowed=args.pruning_max_swaps,
-            min_layers_allowed=args.min_layers,
-            max_layers_allowed=args.pruning_max_layer,
-            search_seed=True,
-            fast_pruning=args.fast_pruning,
-            fast_pruning_percent=args.fast_pruning_percent,
-        )
-        optimizer.log_to_tensorboard(
-            interval=1,
-            namespace="post_opt",
-            step=(post_opt_step := post_opt_step + 1),
-        )
-
-    disc_global, disc_height_image = optimizer.get_discretized_solution(best=True)
-
-    print("Done. Saving outputs...")
-    # Save Image
-    comp_disc = optimizer.get_best_discretized_image()
-    args.max_layers = optimizer.max_layers
+    # change processing pixel heights to large output pixel heights
+    optimizer.pixel_height_logits = torch.from_numpy(pixel_height_logits_init)
+    optimizer.best_params["pixel_height_logits"] = torch.from_numpy(
+        pixel_height_logits_init
+    ).to(device)
+    optimizer.target = output_target
 
     optimizer.log_to_tensorboard(
         interval=1, namespace="post_opt", step=(post_opt_step := post_opt_step + 1)
     )
 
-    comp_disc_np = comp_disc.cpu().numpy().astype(np.uint8)
-    comp_disc_np = cv2.cvtColor(comp_disc_np, cv2.COLOR_RGB2BGR)
-    cv2.imwrite(os.path.join(args.output_folder, "final_model.png"), comp_disc_np)
-
-    stl_filename = os.path.join(args.output_folder, "final_model.stl")
-    height_map_mm = (
-        disc_height_image.cpu().numpy().astype(np.float32)
-    ) * args.layer_height
-    generate_stl(
-        height_map_mm,
-        stl_filename,
-        args.background_height,
-        maximum_x_y_size=args.stl_output_size,
-        alpha_mask=alpha,
-    )
-
-    # Swap instructions
-    background_layers = int(args.background_height // args.layer_height)
-    swap_instructions = generate_swap_instructions(
-        disc_global.cpu().numpy(),
-        disc_height_image.cpu().numpy(),
-        args.layer_height,
-        background_layers,
-        args.background_height,
-        material_names,
-    )
-    with open(os.path.join(args.output_folder, "swap_instructions.txt"), "w") as f:
-        for line in swap_instructions:
-            f.write(line + "\n")
-
-    # Project file
-    project_filename = os.path.join(args.output_folder, "project_file.hfp")
-    generate_project_file(
-        project_filename,
-        args,
-        disc_global.cpu().numpy(),
-        disc_height_image.cpu().numpy(),
-        output_target.shape[1],
-        output_target.shape[0],
-        stl_filename,
-        args.csv_file,
-    )
-
-    print("All done. Outputs in:", args.output_folder)
-    print("Happy Printing!")
+    with torch.no_grad():
+        with torch.autocast(device.type, dtype=torch.bfloat16):
+            if args.perform_pruning:
+                optimizer.prune(
+                    max_colors_allowed=args.pruning_max_colors,
+                    max_swaps_allowed=args.pruning_max_swaps,
+                    min_layers_allowed=args.min_layers,
+                    max_layers_allowed=args.pruning_max_layer,
+                    search_seed=True,
+                    fast_pruning=args.fast_pruning,
+                    fast_pruning_percent=args.fast_pruning_percent,
+                )
+                optimizer.log_to_tensorboard(
+                    interval=1,
+                    namespace="post_opt",
+                    step=(post_opt_step := post_opt_step + 1),
+                )
+
+            disc_global, disc_height_image = optimizer.get_discretized_solution(
+                best=True
+            )
+
+            print("Done. Saving outputs...")
+            # Save Image
+            comp_disc = optimizer.get_best_discretized_image()
+            args.max_layers = optimizer.max_layers
+
+            optimizer.log_to_tensorboard(
+                interval=1,
+                namespace="post_opt",
+                step=(post_opt_step := post_opt_step + 1),
+            )
+
+            comp_disc_np = comp_disc.cpu().numpy().astype(np.uint8)
+            comp_disc_np = cv2.cvtColor(comp_disc_np, cv2.COLOR_RGB2BGR)
+            cv2.imwrite(
+                os.path.join(args.output_folder, "final_model.png"), comp_disc_np
+            )
+
+            stl_filename = os.path.join(args.output_folder, "final_model.stl")
+            height_map_mm = (
+                disc_height_image.cpu().numpy().astype(np.float32)
+            ) * args.layer_height
+            generate_stl(
+                height_map_mm,
+                stl_filename,
+                args.background_height,
+                maximum_x_y_size=args.stl_output_size,
+                alpha_mask=alpha,
+            )
+
+            # Swap instructions
+            background_layers = int(args.background_height // args.layer_height)
+            swap_instructions = generate_swap_instructions(
+                disc_global.cpu().numpy(),
+                disc_height_image.cpu().numpy(),
+                args.layer_height,
+                background_layers,
+                args.background_height,
+                material_names,
+            )
+            with open(
+                os.path.join(args.output_folder, "swap_instructions.txt"), "w"
+            ) as f:
+                for line in swap_instructions:
+                    f.write(line + "\n")
+
+            # Project file
+            project_filename = os.path.join(args.output_folder, "project_file.hfp")
+            generate_project_file(
+                project_filename,
+                args,
+                disc_global.cpu().numpy(),
+                disc_height_image.cpu().numpy(),
+                output_target.shape[1],
+                output_target.shape[0],
+                stl_filename,
+                args.csv_file,
+            )
+
+            print("All done. Outputs in:", args.output_folder)
+            print("Happy Printing!")
 
 
 if __name__ == "__main__":