batch processing v1.2.7

Author: Koldim2001

README.md

@@ -121,6 +121,7 @@ Class implementing cropping and passing crops through a neural network for detec
 | resize_initial_size   | bool                   | False        | Whether to resize the results to the original input image size (ps: slow operation).                           |
 | memory_optimize       | bool                   | True         | Memory optimization option for segmentation (less accurate results when enabled).                              |
 | inference_extra_args  | dict                   | None         | Dictionary with extra ultralytics [inference parameters](https://docs.ultralytics.com/modes/predict/#inference-arguments) (possible keys: half, device, max_det, augment, agnostic_nms and retina_masks) |
+| batch_inference       | bool                   | False        | Batch inference of image crops through a neural network instead of sequential passes of crops (ps: faster inference, higher gpu memory use).                           |
 
 
 **CombineDetections**

patched_yolo_infer/README.md

@@ -9,7 +9,7 @@ This library facilitates various visualizations of inference results from ultral
 You can install the library via pip:
 
 ```bash
-pip install patched_yolo_infer
+pip install patched-yolo-infer
 ```
 
 Note: If CUDA support is available, it's recommended to pre-install PyTorch with CUDA support before installing the library. Otherwise, the CPU version will be installed by default.
@@ -99,6 +99,7 @@ Class implementing cropping and passing crops through a neural network for detec
 - **resize_initial_size** (*bool*): Whether to resize the results to the original image size (ps: slow operation).
 - **memory_optimize** (*bool*): Memory optimization option for segmentation (less accurate results when enabled).
 - **inference_extra_args** (*dict*): Dictionary with extra ultralytics [inference parameters](https://docs.ultralytics.com/modes/predict/#inference-arguments) (possible keys: half, device, max_det, augment, agnostic_nms and retina_masks)
+- **batch_inference** (*bool*): Batch inference of image crops through a neural network instead of sequential passes of crops (ps: faster inference, higher gpu memory use)
 
 **CombineDetections**
 Class implementing combining masks/boxes from multiple crops + NMS (Non-Maximum Suppression).\

requirements.txt

@@ -1,5 +1,5 @@
+numpy<2.0
 torch
-numpy
 opencv-python
 matplotlib
 ultralytics

setup.py

@@ -8,8 +8,8 @@
     long_description = "\n" + fh.read()
 
 
-VERSION = '1.2.6'
-DESCRIPTION = '''YOLO-Patch-Based-Inference for detection/segmentation of small objects in images.'''
+VERSION = '1.2.7'
+DESCRIPTION = '''Patch-Based-Inference for detection/segmentation of small objects in images.'''
 
 setup(
     name="patched_yolo_infer",
@@ -23,7 +23,7 @@
     packages=find_packages(),
     python_requires=">=3.8",
     install_requires=[
-        'numpy',
+        'numpy<2.0',
         'opencv-python',
         'matplotlib',
         'torch',
@@ -33,8 +33,10 @@
         "python",
         "yolov8",
         "yolov9",
+        "yolov10",
         "rtdetr",
-        "sam",
+        "fastsam",
+        "sahi",
         "object detection",
         "instance segmentation",
         "patch-based inference",