autoforge 1.7.1__tar.gz → 1.7.3__tar.gz

{autoforge-1.7.1 → autoforge-1.7.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: AutoForge
-Version: 1.7.1
+Version: 1.7.3
 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
@@ -30,8 +30,11 @@ AutoForge is a Python tool for generating 3D printed layered models from an inpu
 
 **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).
 
+
+## You can now run Autoforge for free in your browser thanks to [Huggingface space support](https://huggingface.co/spaces/hvoss-techfak/Autoforge).
+
 ## Example
-All examples use only the 13 BambuLab Basic filaments, currently available in Hueforge, the background color is set to black.
+All examples use only the 13 BambuLab Basic filaments, currently available in Hueforge 0.8.9, the background color is set to black.
 The pruning is set to a maximum of 8 color and 20 swaps, so each image uses at most 8 different colors and swaps the filament at most 20 times. 
 <div style="display: flex; justify-content: center; gap: 20px;">
   <div style="text-align: center;">

{autoforge-1.7.1 → autoforge-1.7.3}/README.md

@@ -4,8 +4,11 @@ AutoForge is a Python tool for generating 3D printed layered models from an inpu
 
 **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).
 
+
+## You can now run Autoforge for free in your browser thanks to [Huggingface space support](https://huggingface.co/spaces/hvoss-techfak/Autoforge).
+
 ## Example
-All examples use only the 13 BambuLab Basic filaments, currently available in Hueforge, the background color is set to black.
+All examples use only the 13 BambuLab Basic filaments, currently available in Hueforge 0.8.9, the background color is set to black.
 The pruning is set to a maximum of 8 color and 20 swaps, so each image uses at most 8 different colors and swaps the filament at most 20 times. 
 <div style="display: flex; justify-content: center; gap: 20px;">
   <div style="text-align: center;">

{autoforge-1.7.1 → autoforge-1.7.3}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "AutoForge"
-version="1.7.1"
+version="1.7.3"
 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-1.7.1 → autoforge-1.7.3}/src/AutoForge.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: AutoForge
-Version: 1.7.1
+Version: 1.7.3
 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
@@ -30,8 +30,11 @@ AutoForge is a Python tool for generating 3D printed layered models from an inpu
 
 **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).
 
+
+## You can now run Autoforge for free in your browser thanks to [Huggingface space support](https://huggingface.co/spaces/hvoss-techfak/Autoforge).
+
 ## Example
-All examples use only the 13 BambuLab Basic filaments, currently available in Hueforge, the background color is set to black.
+All examples use only the 13 BambuLab Basic filaments, currently available in Hueforge 0.8.9, the background color is set to black.
 The pruning is set to a maximum of 8 color and 20 swaps, so each image uses at most 8 different colors and swaps the filament at most 20 times. 
 <div style="display: flex; justify-content: center; gap: 20px;">
   <div style="text-align: center;">

{autoforge-1.7.1 → autoforge-1.7.3}/src/autoforge/Helper/Heightmaps/ChristofidesHeightMap.py

@@ -532,9 +532,10 @@ def run_init_threads(
     cluster_layers=None,
     material_colors=None,
 ):
+    background_tuple = (np.asarray(background_tuple) * 255).tolist()
     if random_seed is None:
         random_seed = np.random.randint(1e6)
-    lab_space = False
+    lab_space = True
     tasks = [
         delayed(init_height_map)(
             target,

{autoforge-1.7.1 → autoforge-1.7.3}/src/autoforge/Helper/PruningHelper.py

@@ -51,6 +51,7 @@ def prune_num_colors(
     *,
     fast: bool = True,  # enable incremental search
     chunking_percent=0.05,  # percentage of layers to process at once
+    allowed_loss_increase_percent=0.005,  # percentage of loss increase allowed over best loss
 ) -> torch.Tensor:
     """Iteratively merge materials until the number of distinct colors is
     <= max_colors_allowed or until merging would worsen the loss.
@@ -116,7 +117,7 @@ def prune_num_colors(
                     n_jobs=n_jobs, backend="threading", prefer="threads"
                 )(delayed(score_color)(best_dg, *pair) for pair in chunk)
                 merge_loss, merge_dg = min(cand_results, key=lambda x: x[0])
-                if merge_loss < best_loss:
+                if merge_loss < best_loss * (1 + allowed_loss_increase_percent):
                     best_dg, best_loss = merge_dg, merge_loss
                     improved = True
                     break  # move on to next outer iteration
@@ -164,6 +165,7 @@ def prune_num_swaps(
     *,
     fast: bool = True,  # enable incremental search
     chunking_percent=0.05,  # percentage of layers to process at once
+    allowed_loss_increase_percent=0.005,  # percentage of loss increase allowed over best loss
 ) -> torch.Tensor:
     """Reduce the number of color boundaries until it is below or equal to
     *max_swaps_allowed*, stopping earlier if any further merge would worsen the
@@ -231,7 +233,7 @@ def prune_num_swaps(
                     for band_a, band_b, direction in chunk
                 )
                 merge_loss, merge_dg = min(cand_results, key=lambda x: x[0])
-                if merge_loss < best_loss:
+                if merge_loss < best_loss * (1 + allowed_loss_increase_percent):
                     best_dg, best_loss = merge_dg, merge_loss
                     improved = True
                     break
@@ -389,6 +391,7 @@ def prune_redundant_layers(
     *,
     fast: bool = True,  # NEW: enable 10 percent incremental search
     chunking_percent=0.05,  # percentage of layers to process at once
+    allowed_loss_increase_percent=0.005,  # percentage of loss increase allowed over best loss
 ):
     """Iteratively drop layers until the loss cannot be improved.
 
@@ -466,7 +469,7 @@ def prune_redundant_layers(
                 cand_loss, cand_params, cand_max_layers = min(
                     cand_results, key=lambda x: x[0]
                 )
-                if cand_loss <= best_loss:
+                if cand_loss <= best_loss * (1 + allowed_loss_increase_percent):
                     removed_layers += 1
                     best_loss = cand_loss
                     current_max_layers = cand_max_layers

{autoforge-1.7.1 → autoforge-1.7.3}/src/autoforge/Modules/Optimizer.py

@@ -5,12 +5,6 @@ import matplotlib.pyplot as plt
 import numpy as np
 import torch
 from torch.utils.tensorboard import SummaryWriter
-from torchcam.methods import SmoothGradCAMpp
-from torchcam.utils import overlay_mask
-from torchvision import models
-from torchvision.models import ResNet18_Weights
-from torchvision.transforms._functional_tensor import resize, normalize
-from torchvision.transforms.functional import to_pil_image
 from tqdm import tqdm
 
 from autoforge.Helper.CAdamW import CAdamW
@@ -119,48 +113,48 @@ class FilamentOptimizer:
             lr=self.learning_rate,
         )
 
-        # Setup the focus map component
-        # Load a state-of-the-art ResNet50 model with pretrained weights.
-        cam_model = models.resnet18(weights=ResNet18_Weights.DEFAULT).to(
-            self.target.device
-        )
-        cam_model.eval()
-        # Use ScoreCAM from torchcam on the 'layer4' block.
-        cam_extractor = SmoothGradCAMpp(cam_model, target_layer="layer4")
-        # Preprocess the target image for ResNet50.
-        target_preprocessed = normalize(
-            resize(self.target.permute(2, 0, 1), (224, 224)) / 255.0,
-            [0.485, 0.456, 0.406],
-            [0.229, 0.224, 0.225],
-        )
-        focus_maps = []
-        for _ in tqdm(range(1), desc="Creating loss function attention map..."):
-            out = cam_model(target_preprocessed.unsqueeze(0))
-            # ScoreCAM returns a dict with keys corresponding to layers.
-            focus_map = cam_extractor(out.squeeze(0).argmax().item(), out)[0]
-            # Normalize the focus map to [0, 1]
-            focus_map = (focus_map - focus_map.min()) / (
-                focus_map.max() - focus_map.min() + 1e-8
-            )
-            focus_maps.append(focus_map)
-        # get maximum over all maps
-        self.focus_map = torch.stack(focus_maps)
-        self.focus_map = self.focus_map.max(dim=0)[0]
-        # normalize again
-        self.focus_map = (self.focus_map - self.focus_map.min()) / (
-            self.focus_map.max() - self.focus_map.min() + 1e-8
-        )
-
-        # resize back to target size, current dim is [1,h,w]
-        self.focus_map = (
-            torch.nn.functional.interpolate(
-                self.focus_map.unsqueeze(0),
-                (self.H, self.W),
-                mode="bilinear",
-            )
-            .squeeze(0)
-            .squeeze(0)
-        )
+        # # Setup the focus map component
+        # # Load a state-of-the-art ResNet50 model with pretrained weights.
+        # cam_model = models.resnet18(weights=ResNet18_Weights.DEFAULT).to(
+        #     self.target.device
+        # )
+        # cam_model.eval()
+        # # Use ScoreCAM from torchcam on the 'layer4' block.
+        # cam_extractor = SmoothGradCAMpp(cam_model, target_layer="layer4")
+        # # Preprocess the target image for ResNet50.
+        # target_preprocessed = normalize(
+        #     resize(self.target.permute(2, 0, 1), (224, 224)) / 255.0,
+        #     [0.485, 0.456, 0.406],
+        #     [0.229, 0.224, 0.225],
+        # )
+        # focus_maps = []
+        # for _ in tqdm(range(1), desc="Creating loss function attention map..."):
+        #     out = cam_model(target_preprocessed.unsqueeze(0))
+        #     # ScoreCAM returns a dict with keys corresponding to layers.
+        #     focus_map = cam_extractor(out.squeeze(0).argmax().item(), out)[0]
+        #     # Normalize the focus map to [0, 1]
+        #     focus_map = (focus_map - focus_map.min()) / (
+        #         focus_map.max() - focus_map.min() + 1e-8
+        #     )
+        #     focus_maps.append(focus_map)
+        # # get maximum over all maps
+        # self.focus_map = torch.stack(focus_maps)
+        # self.focus_map = self.focus_map.max(dim=0)[0]
+        # # normalize again
+        # self.focus_map = (self.focus_map - self.focus_map.min()) / (
+        #     self.focus_map.max() - self.focus_map.min() + 1e-8
+        # )
+        #
+        # # resize back to target size, current dim is [1,h,w]
+        # self.focus_map = (
+        #     torch.nn.functional.interpolate(
+        #         self.focus_map.unsqueeze(0),
+        #         (self.H, self.W),
+        #         mode="bilinear",
+        #     )
+        #     .squeeze(0)
+        #     .squeeze(0)
+        # )
 
         # Setup best discrete solution tracking
         self.best_discrete_loss = float("inf")
@@ -202,17 +196,14 @@ class FilamentOptimizer:
             )
             self.ax[1, 1].set_title("Height Map Changes")
 
-            target_vis = self.target.cpu().detach().numpy()
-            focus_vis = self.focus_map.cpu().detach().numpy()
-
-            overlay = overlay_mask(
-                to_pil_image(target_vis),
-                to_pil_image(focus_vis, mode="F"),
-                alpha=0.5,
-            )
-
-            self.focus_map_ax = self.ax[1, 2].imshow(overlay)
-            self.ax[1, 2].set_title("Image Focus Map (currently unused)")
+            # target_vis = self.target.cpu().detach().numpy()
+            # focus_vis = self.focus_map.cpu().detach().numpy()
+            #
+            # overlay = overlay_mask(
+            #     to_pil_image(target_vis),
+            #     to_pil_image(focus_vis, mode="F"),
+            #     alpha=0.5,
+            # )
 
             # Compute and store the initial height map for later difference computation.
             with torch.no_grad():
@@ -286,7 +277,7 @@ class FilamentOptimizer:
             material_TDs=self.material_TDs,
             background=self.background,
             add_penalty_loss=10.0,
-            focus_map=self.focus_map,
+            focus_map=None,
             focus_strength=0.0,
         )
 
@@ -609,9 +600,6 @@ class FilamentOptimizer:
                     rng_seed=seed,
                 )
 
-            # 3) Compute the "discrete" loss function
-            #    If you have a separate function for that, just call it.
-            #    Otherwise, reuse the standard compute_loss with add_penalty_loss=False, etc.
             current_disc_loss = compute_loss(
                 comp=comp_disc,
                 target=self.target,

{autoforge-1.7.1 → autoforge-1.7.3}/src/autoforge/auto_forge.py

@@ -390,7 +390,7 @@ def main():
         for i in tbar:
             loss_val = optimizer.step(record_best=i % 8 == 0)
 
-            optimizer.visualize(interval=25)
+            optimizer.visualize(interval=100)
             optimizer.log_to_tensorboard(interval=100)
 
             if (i + 1) % 100 == 0: