simba-uw-tf-dev 5.2.4__tar.gz → 5.2.6__tar.gz

{simba_uw_tf_dev-5.2.4 → simba_uw_tf_dev-5.2.6}/.gitignore

@@ -39,3 +39,6 @@ docs/_build
 
 # ignore recent projects file anywhere
 **/.recent_projects.txt
+
+# Claude Code
+simba/.claude/

{simba_uw_tf_dev-5.2.4 → simba_uw_tf_dev-5.2.6}/MANIFEST.in

@@ -9,6 +9,8 @@ prune */pose_configurations_archive
 recursive-exclude */pose_configurations_archive *
 global-exclude pose_configurations_archive/*
 global-exclude */pose_configurations_archive/*
+prune simba/.claude
+recursive-exclude simba/.claude *
 prune docs
 prune misc
 prune images

{simba_uw_tf_dev-5.2.4 → simba_uw_tf_dev-5.2.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: simba_uw_tf_dev
-Version: 5.2.4
+Version: 5.2.6
 Summary: Toolkit for computer classification and analysis of behaviors in experimental animals
 Home-page: https://github.com/sgoldenlab/simba
 Author: Simon Nilsson, Jia Jie Choong, Sophia Hwang

{simba_uw_tf_dev-5.2.4 → simba_uw_tf_dev-5.2.6}/setup.py

@@ -29,7 +29,7 @@ with open(GPU_REQUIREMENTS_PATH, "r") as f:
 # Setup configuration
 setuptools.setup(
     name="simba_uw_tf_dev",
-    version="5.2.4",
+    version="5.2.6",
     author="Simon Nilsson, Jia Jie Choong, Sophia Hwang",
     author_email="sronilsson@gmail.com",
     description="Toolkit for computer classification and analysis of behaviors in experimental animals",

{simba_uw_tf_dev-5.2.4 → simba_uw_tf_dev-5.2.6}/simba/data_processors/smoothing.py

@@ -110,9 +110,10 @@ class Smoothing(ConfigReader):
                 else:
                     self.video_info_df = self.read_video_info_csv(file_path=self.video_info_path)
                     video_info = read_video_info(vid_info_df=self.video_info_df,video_name=video_name, raise_error=False)
-                    if video_info is None:
-                        raise NoFilesFoundError(msg=f"To perform smoothing, SimBA needs to read the video FPS. SimBA could not find the video {video_name} in represented in the {self.video_dir} directory or in {self.video_info_path} file. Please import the video and/or include it in the video_logs.csv file so SimBA can know the video FPS", source=self.__class__.__name__)
-                    fps = video_info[2]
+                    if video_info[2] is None:
+                        raise NoFilesFoundError(msg=f"To perform smoothing, SimBA needs to read the video FPS. SimBA could not find the video {video_name} represented in the {self.video_dir} directory or in {self.video_info_path} file. Please import the video and/or include it in the video_logs.csv file so SimBA can know the video FPS", source=self.__class__.__name__)
+                    else:
+                        fps = video_info[2]
             else:
                 fps = get_video_meta_data(video_path=video_path)['fps']
             if self.method == 'savitzky-golay':

{simba_uw_tf_dev-5.2.4 → simba_uw_tf_dev-5.2.6}/simba/labelling/extract_labelled_frames.py

@@ -74,6 +74,7 @@ class AnnotationFrameExtractor(ConfigReader):
             video_path = find_video_of_file(video_dir=self.video_dir, filename=video_name, raise_error=True)
             self.video_lk[video_name] = video_path
 
+
     def run(self):
         create_directory(paths=os.path.join(self.annotated_frm_dir, 'images'), overwrite=False, verbose=False)
         for file_cnt, data_path in enumerate(self.data_paths):
@@ -91,9 +92,9 @@ class AnnotationFrameExtractor(ConfigReader):
                     continue
                 for frm_cnt, frm in enumerate(clf_annot_idx):
                     img = read_frm_of_video(video_path=cap, frame_index=frm, size=size, greyscale=self.img_greyscale)
-                    img_save_path = os.path.join(self.annotated_frm_dir, 'imgs', f'{video_name}_{clf}_{frm_cnt}.{self.img_format}')
+                    img_save_path = os.path.join(self.annotated_frm_dir, 'images', f'{video_name}_{clf}_{frm_cnt}.{self.img_format}')
                     cv2.imwrite(img_save_path, img)
-                    stdout_information(f"Saved {clf} annotated image for classifier {clf} ({str(frm_cnt)}/{str(len(clf_annot_idx))}), video: {file_cnt+1}/{len(self.data_paths)}, ({video_name}) ...")
+                    stdout_information(f"Saved {clf} annotated image for classifier {clf} at location {img_save_path} ({str(frm_cnt)}/{str(len(clf_annot_idx))}), video: {file_cnt+1}/{len(self.data_paths)}, ({video_name}) ...")
         self.timer.stop_timer()
         stdout_success(msg=f"Annotated frames saved in {self.annotated_frm_dir} directory", elapsed_time=self.timer.elapsed_time_str)
 

{simba_uw_tf_dev-5.2.4 → simba_uw_tf_dev-5.2.6}/simba/mixins/geometry_mixin.py

@@ -4150,7 +4150,8 @@ class GeometryMixin(object):
         return results
 
     @staticmethod
-    def smooth_geometry_bspline(data: Union[np.ndarray, Polygon, List[Polygon]], smooth_factor: float = 1.0,
+    def smooth_geometry_bspline(data: Union[np.ndarray, Polygon, List[Polygon]],
+                                smooth_factor: float = 1.0,
                                 points: int = 50) -> List[Polygon]:
         """
         Smooths the geometry of polygons or coordinate arrays using B-spline interpolation.

{simba_uw_tf_dev-5.2.4 → simba_uw_tf_dev-5.2.6}/simba/model/yolo_fit.py

@@ -181,17 +181,17 @@ if __name__ == "__main__" and not hasattr(sys, 'ps1'):
     yolo_fitter.run()
 
 
-
-# fitter = FitYolo(weights_path=r"E:\litpose_yolo\yolo_from_sam3\yolo26m-seg.pt",
-#                  model_yaml=r"E:\litpose_yolo\yolo_from_sam3\map.yaml",
-#                  save_path=r'E:\litpose_yolo\yolo_from_sam3\mdl',
+#
+# fitter = FitYolo(weights_path=r"C:\Users\sroni\Downloads\yolo26x-seg.pt",
+#                  model_yaml=r"E:\open_video\open_field_2\yolo_seg_project\map.yaml",
+#                  save_path=r'E:\open_video\open_field_2\yolo_seg_project\mdl',
 #                  epochs=1500,
-#                  batch=16,
+#                  batch=8,
 #                  format=None,
 #                  device=0,
 #                  imgsz=256)
 # fitter.run()
-#
+
 
 
 

{simba_uw_tf_dev-5.2.4 → simba_uw_tf_dev-5.2.6}/simba/model/yolo_inference.py

@@ -31,8 +31,7 @@ from simba.utils.enums import Formats, Options
 from simba.utils.errors import (InvalidVideoFileError, SimBAGPUError,
                                 SimBAPAckageVersionError)
 from simba.utils.printing import SimbaTimer, stdout_success
-from simba.utils.read_write import (find_core_cnt,
-                                    find_files_of_filetypes_in_directory,
+from simba.utils.read_write import (find_files_of_filetypes_in_directory,
                                     get_video_meta_data)
 from simba.utils.yolo import (_get_undetected_obs, apply_fixed_bbox_size,
                               check_valid_device, load_yolo_model,
@@ -217,27 +216,28 @@ class YoloInference():
             video_timer.stop_timer()
             fps_speeds.append(len(results[video_name]) / video_timer.elapsed_time)
 
-        timer.stop_timer()
-        if not self.save_dir:
-            if self.verbose:
-                stdout_success(f'YOLO results created', elapsed_time=timer.elapsed_time_str)
-            return results
-        else:
+        if self.save_dir:
             for k, v in results.items():
                 save_path = os.path.join(self.save_dir, f'{k}.csv')
                 v.to_csv(save_path)
-            if self.verbose:
+        timer.stop_timer()
+        if self.verbose:
+            if self.save_dir:
                 stdout_success(f'YOLO results for {len(self.video_path)} video(s) saved in {self.save_dir} directory', elapsed_time=timer.elapsed_time_str)
+            else:
+                stdout_success(f'YOLO results created for {len(self.video_path)} video(s)', elapsed_time=timer.elapsed_time_str)
+        if not self.save_dir:
+            return results
 
 
 
 
 
 
-#
-# VIDEO_PATH = r"Z:\home\simon\lp_300126\videos\6.01.001_2026_03_11_23_25_00_000_2\6.01.001_2026_03_11_23_25_00_000_2_cam1.mp4"
-# WEIGHTS_PATH = r"E:\litpose_yolo\bbox\mdl\train3\weights\best.pt"
-# SAVE_DIR = r"E:\litpose_yolo\bbox\out_pose"
+
+# VIDEO_PATH = r"E:\open_video\open_field_2\sample\clips"
+# WEIGHTS_PATH = r'E:\open_video\open_field_2\yolo_bbox_project\mdl\train2\weights\best.pt'
+# SAVE_DIR = r'E:\open_video\open_field_2\yolo_bbox_project\results'
 # i = YoloInference(weights=WEIGHTS_PATH,
 #                   video_path=VIDEO_PATH,
 #                   save_dir=SAVE_DIR,
@@ -246,9 +246,9 @@ class YoloInference():
 #                   verbose=True,
 #                   core_cnt=18,
 #                   imgsz=256,
-#                   bbox_size=(512, 512),
+#                   bbox_size=None,#(512, 512),
 #                   interpolate=True,
-#                   batch_size=100)
+#                   batch_size=8)
 # i.run()
 #