simba-uw-tf-dev 5.5.2__tar.gz → 5.5.3__tar.gz

{simba_uw_tf_dev-5.5.2 → simba_uw_tf_dev-5.5.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: simba_uw_tf_dev
-Version: 5.5.2
+Version: 5.5.3
 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-5.5.2 → simba_uw_tf_dev-5.5.3}/setup.py

@@ -30,7 +30,7 @@ with open(GPU_REQUIREMENTS_PATH, "r") as f:
 # Setup configuration
 setuptools.setup(
     name="simba_uw_tf_dev",
-    version="5.5.2",
+    version="5.5.3",
     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-5.5.2 → simba_uw_tf_dev-5.5.3}/simba/plotting/interactive_probability_grapher.py

@@ -15,9 +15,18 @@ from simba.utils.checks import (check_file_exist_and_readable,
                                 check_valid_boolean, check_valid_dataframe)
 from simba.utils.enums import OS, Formats, Paths
 from simba.utils.errors import ColumnNotFoundError, InvalidInputError
+from simba.utils.lookups import get_monitor_info
 from simba.utils.read_write import get_fn_ext, get_video_meta_data, read_df
 from simba.utils.warnings import FrameRangeWarning
 
+WINDOW_PADDING = 10
+PLOT_DPI = 100
+MIN_PLOT_W = 500
+MAX_PLOT_W = 1200
+PLOT_H = 600
+VIDEO_IMG_SCREEN_RATIO = (0.45, 0.7)
+VIDEO_WINDOW_CHROME_H = 220   # NOTE: approximate height of the video window navigation buttons and title bar, plus room for the taskbar - keeps the window on-screen on small displays.
+
 ICON_WINDOWS = os.path.join(os.path.dirname(simba.__file__), Paths.LOGO_ICON_WINDOWS_PATH.value)
 ICON_DARWIN = os.path.join(os.path.dirname(simba.__file__), Paths.LOGO_ICON_DARWIN_PATH.value)
 
@@ -89,7 +98,9 @@ class InteractiveProbabilityGrapher(ConfigReader):
         if self.video_meta_data['frame_count'] != len(self.data_df):
             FrameRangeWarning(msg=f'The video {current_video_file_path} contains {self.video_meta_data["frame_count"]} frames, while the data file {self.data_path} contains {len(self.data_df)} frames.', source=self.__class__.__name__)
 
-        self.video_frm = InteractiveVideoPlotterWindow(video_path=current_video_file_path, p_arr=self.p_arr, data_df=self.data_df, config_path=self.config_path, show_names= True if self.animal_cnt >1 else False)
+        _, (self.screen_w, self.screen_h) = get_monitor_info()
+        max_img_size = (int(self.screen_w * VIDEO_IMG_SCREEN_RATIO[0]), min(int(self.screen_h * VIDEO_IMG_SCREEN_RATIO[1]), self.screen_h - VIDEO_WINDOW_CHROME_H))
+        self.video_frm = InteractiveVideoPlotterWindow(video_path=current_video_file_path, p_arr=self.p_arr, data_df=self.data_df, config_path=self.config_path, show_names= True if self.animal_cnt >1 else False, max_img_size=max_img_size, data_path=self.data_path)
         self.video_frm.main_frm.protocol("WM_DELETE_WINDOW", self._close_windows)
 
     @staticmethod
@@ -119,7 +130,11 @@ class InteractiveProbabilityGrapher(ConfigReader):
         plt.rcParams['keymap.back'] = [k for k in plt.rcParams['keymap.back'] if k != 'left']
         plt.rcParams['keymap.forward'] = [k for k in plt.rcParams['keymap.forward'] if k != 'right']
 
-        fig, ax = plt.subplots(figsize=(12, 6), dpi=100)
+        self.video_frm.main_frm.update_idletasks()
+        video_w = self.video_frm.main_frm.winfo_width()
+        plot_w = min(MAX_PLOT_W, max(MIN_PLOT_W, self.screen_w - video_w - (WINDOW_PADDING * 2)))
+        plot_h = min(PLOT_H, int(self.screen_h * 0.75))
+        fig, ax = plt.subplots(figsize=(plot_w / PLOT_DPI, plot_h / PLOT_DPI), dpi=PLOT_DPI)
         fig.patch.set_facecolor('white')
         ax.spines['top'].set_visible(False)
         ax.spines['right'].set_visible(False)
@@ -159,6 +174,7 @@ class InteractiveProbabilityGrapher(ConfigReader):
         fig.canvas.mpl_connect("button_press_event", lambda event: self.__click_event(event))  # ADD THIS - it's missing!
         fig.canvas.mpl_connect("key_press_event", self.__key_press_event)
         plt.show(block=False)
+        self._place_windows_side_by_side(fig=fig)
         fig.canvas.get_tk_widget().focus_force()
 
         while plt.fignum_exists(fig.number):
@@ -186,6 +202,25 @@ class InteractiveProbabilityGrapher(ConfigReader):
 
         self.video_frm.main_frm.destroy()
 
+    def _place_windows_side_by_side(self, fig) -> None:
+        """Place the video window at the top-left of the display, and the probability plot window immediately to the right of it."""
+        plot_frm = fig.canvas.manager.window
+        if not hasattr(plot_frm, 'wm_geometry'):
+            return
+        self.video_frm.main_frm.update_idletasks()
+        plot_frm.update_idletasks()
+        video_w = self.video_frm.main_frm.winfo_width()
+        plot_w = plot_frm.winfo_width()
+        if plot_w <= 1:
+            plot_w = int(fig.get_size_inches()[0] * fig.dpi)
+        plot_x = min(video_w + WINDOW_PADDING, max(0, self.screen_w - plot_w))
+        self.video_frm.main_frm.wm_geometry('+0+0')
+        plot_frm.wm_geometry(f'+{plot_x}+0')
+        self.video_frm.main_frm.update()   # NOTE: full update (not update_idletasks) - the window manager requires processed events for the new positions to stick.
+        plot_frm.update()
+        self.video_frm.main_frm.lift()
+        plot_frm.lift()
+
     def _close_windows(self):
         try:
             self.video_frm.main_frm.destroy()

{simba_uw_tf_dev-5.5.2 → simba_uw_tf_dev-5.5.3}/simba/plotting/tools/tkinter_tools.py

@@ -1,208 +1,248 @@
-__author__ = "Simon Nilsson; sronilsson@gmail.com"
-
-import os
-from tkinter import *
-from typing import Union, Optional
-
-import cv2
-import numpy as np
-import pandas as pd
-from PIL import Image as Img
-from PIL import ImageTk
-from copy import deepcopy
-
-from simba.mixins.config_reader import ConfigReader
-from simba.mixins.plotting_mixin import PlottingMixin
-from simba.utils.checks import check_file_exist_and_readable, check_valid_array, check_valid_boolean, check_valid_dataframe
-from simba.utils.enums import Formats
-from simba.utils.errors import InvalidInputError
-from simba.utils.lookups import get_icons_paths, get_simba_font_name_and_path, get_color_dict
-from simba.utils.enums import TextOptions
-from simba.utils.data import create_color_palettes
-from simba.utils.read_write import get_video_meta_data, read_frm_of_video, str_2_bool
-from simba.utils.warnings import FrameRangeWarning
-from simba.ui.tkinter_functions import SimBALabel, SimBAScaleBar, SimbaButton, Entry_Box, CreateLabelFrameWithIcon, SimBADropDown
-
-PADDING = 5
-MAX_SIZE = (1080, 650)
-
-
-class InteractiveVideoPlotterWindow(ConfigReader, PlottingMixin):
-    def __init__(self,
-                 video_path: Union[str, os.PathLike],
-                 p_arr: np.array,
-                 config_path: Optional[Union[str, os.PathLike]] = None,
-                 data_df: Optional[pd.DataFrame] = None,
-                 show_names: bool = False,
-                 show_pose: bool = False):
-
-        check_file_exist_and_readable(file_path=video_path, raise_error=True)
-        check_valid_array(data=p_arr, source=f'{self.__class__.__name__} p_arr', accepted_dtypes=Formats.NUMERIC_DTYPES.value)
-        check_valid_boolean(value=show_names, source=f'{self.__class__.__name__} show_names', raise_error=True)
-        check_valid_boolean(value=show_pose, source=f'{self.__class__.__name__} show_pose', raise_error=True)
-        if data_df is not None:
-            check_valid_dataframe(df=data_df, source=f'{self.__class__.__name__} data_df', valid_dtypes=Formats.NUMERIC_DTYPES.value)
-        if config_path is not None:
-            check_file_exist_and_readable(file_path=config_path, raise_error=True)
-            ConfigReader.__init__(self, config_path=config_path, read_video_info=False, create_logger=False)
-        if (show_names or show_pose) and (not isinstance(data_df, pd.DataFrame) or config_path is None):
-            raise InvalidInputError(msg='If showing names and/or pose, please pass config_path and data_df')
-        self.video_meta_data = get_video_meta_data(video_path=video_path)
-        if show_names or show_pose:
-            self.color_palettes = create_color_palettes(no_animals=len(self.animal_bp_dict.keys()), map_size=int(len(self.body_parts_lst)/len(self.animal_bp_dict.keys())))
-            self.color_palettes = [[[int(c) for c in clr] for clr in sublist] for sublist in self.color_palettes]
-            _, self.font_path = get_simba_font_name_and_path(font=TextOptions.DEFAULT_FONT.value)
-            self.font_size, _, _ = PlottingMixin().get_optimal_font_size_ttf(text=list(self.animal_bp_dict.keys()), font_path=self.font_path, accepted_px_width=int(self.video_meta_data['width']/10), accepted_px_height=int(self.video_meta_data['height']/10))
-            self.circle_size = PlottingMixin().get_optimal_circle_size(frame_size=(self.video_meta_data['width'], self.video_meta_data['height']), circle_frame_ratio=100)
-        else:
-            self.color_palettes, self.font_size, self.circle_size = None, None, None
-        self.show_names, self.show_pose, self.data_df, self.config_path = show_names, show_pose, data_df, config_path
-        self.main_frm = Toplevel()
-        self.btn_icons = get_icons_paths()
-        for k in self.btn_icons.keys():
-            self.btn_icons[k]["img"] = ImageTk.PhotoImage(image=Img.open(os.path.join(os.path.dirname("__file__"), self.btn_icons[k]["icon_path"])))
-        self.main_frm.iconphoto(False, self.btn_icons['frames']['img'])
-        self.current_frm_number, self.jump_size = 0, 0
-        self.img_frm = Frame(self.main_frm)
-        self.clr_dict = get_color_dict()
-        self.text_bg_clr, self.name_bg_opacity = 'Black', 0.4
-        self.img_frm.grid(row=0, column=1, sticky=NW)
-        self.button_frame = Frame(self.main_frm, bd=2, width=700, height=300)
-        self.button_frame.grid(row=1, column=0)
-        self.main_frm.wm_title(self.video_meta_data['video_name'])
-        self.cap = cv2.VideoCapture(video_path)
-        self.max_frm = np.argmax(p_arr)
-        self.frame_id_lbl = SimBALabel(parent=self.button_frame, txt="FRAME NUMBER", justify='center')
-        self.frame_id_lbl.grid(row=0, column=0, pady=(0, PADDING))
-
-        self.nav_frm = Frame(self.button_frame)
-        self.nav_frm.grid(row=1, column=0)
-        self.back_first_frm = SimbaButton(parent=self.nav_frm, txt="<<", cmd=self.load_new_frame, cmd_kwargs={'frm_cnt': lambda: 0})
-        self.back_first_frm.grid(row=0, column=0)
-        self.back_one_frm_btn = SimbaButton(parent=self.nav_frm, txt="<", cmd=self.load_new_frame, cmd_kwargs={'frm_cnt': lambda: self.current_frm_number - 1})
-        self.back_one_frm_btn.grid(row=0, column=1, padx=(0, PADDING))
-        self.frame_entry_box = Entry_Box(parent=self.nav_frm, value=self.current_frm_number, allow_blank=False, justify='center', padx=PADDING, validation='numeric')
-        self.frame_entry_box.grid(row=0, column=2)
-        self.forward_next_frm_btn = SimbaButton(parent=self.nav_frm, txt=">", cmd=self.load_new_frame, cmd_kwargs={'frm_cnt': lambda: self.current_frm_number + 1})
-        self.forward_next_frm_btn.grid(row=0, column=3, padx=(PADDING, 0))
-        self.forward_last_frm_btn = SimbaButton(parent=self.nav_frm, txt=">>", cmd=self.load_new_frame, cmd_kwargs={'frm_cnt': lambda: self.video_meta_data["frame_count"] - 1})
-        self.forward_last_frm_btn.grid(row=0, column=4)
-        self.select_frm_btn = SimbaButton(parent=self.button_frame, txt="Jump to selected frame", cmd=self.load_new_frame, cmd_kwargs={'frm_cnt': lambda: int(self.frame_entry_box.entry_get)}, img='jump')
-        self.select_frm_btn.grid(row=2, column=0, pady=(PADDING, 0))
-
-        self.jump_frm = Frame(self.button_frame)
-        self.jump_frm.grid(row=3, column=0, pady=(PADDING, 0))
-
-        self.jump_size_scale = SimBAScaleBar(parent=self.jump_frm, label="Jump Size:", from_=0, to=100, length=200, value=0)
-        self.jump_size_scale.grid(row=0, column=0)
-        self.jump_back_btn = SimbaButton(parent=self.jump_frm, txt="<<", cmd=self.load_new_frame, cmd_kwargs={'frm_cnt': lambda: self.current_frm_number - self.jump_size_scale.get()})
-        self.jump_back_btn.grid(row=0, column=1, padx=(PADDING, 0))
-        self.jump_forward_btn = SimbaButton(parent=self.jump_frm, txt=">>", cmd=self.load_new_frame, cmd_kwargs={'frm_cnt': lambda: self.current_frm_number + self.jump_size_scale.get()})
-        self.jump_forward_btn.grid(row=0, column=2)
-
-        self.load_new_frame(frm_cnt=self.current_frm_number)
-
-        instructions_frm = Frame(self.main_frm, width=100, height=100)
-        instructions_frm.grid(row=0, column=2, sticky=N)
-
-        key_presses = SimBALabel(parent=instructions_frm, txt="\n\n Keyboard shortcuts for frame navigation: \n Right Arrow = +1 frame"
-                                                "\n Left Arrow = -1 frame"
-                                                "\n Ctrl + l = Last frame"
-                                                "\n Ctrl + o = First frame", font=Formats.FONT_REGULAR_BOLD.value)
-        key_presses.grid(row=0, column=0, sticky=S)
-        self.move_to_highest_p_btn = SimbaButton(parent=instructions_frm, txt="SHOW HIGHEST \n PROBABILITY FRAME", cmd=self.load_new_frame, cmd_kwargs={'frm_cnt': lambda: self.max_frm}, img='pct')
-        self.move_to_highest_p_btn.grid(row=1, column=0, sticky=N, pady=(PADDING, 0))
-
-        self.menu = Menu(self.main_frm, tearoff=0)
-        self.file_menu = Menu(self.menu, tearoff=0)
-        self.menu.add_cascade(label="File...", menu=self.file_menu)
-        self.file_menu.add_command(label="Preferences", compound="left", image=self.btn_icons["settings"]["img"], command=lambda: self.preferences_pop_up())
-        self.main_frm.config(menu=self.menu)
-
-        self._bind_keys()
-
-    def _show_names(self, img: np.ndarray, frm_cnt: int) -> np.ndarray:
-        for cnt, animal_name in enumerate(self.animal_bp_dict.keys()):
-            name_cols = [self.animal_bp_dict[animal_name]['X_bps'][0], self.animal_bp_dict[animal_name]['Y_bps'][0]]
-            name_loc = self.data_df.loc[frm_cnt, name_cols].values.astype(np.int32)
-            img = PlottingMixin().put_text(img=img, text=animal_name, pos=(int(name_loc[0]), int(name_loc[1])), font_size=self.font_size, font_path=self.font_path, text_color=tuple(self.color_palettes[cnt][0]), text_bg_alpha=self.name_bg_opacity, text_color_bg=self.clr_dict[self.text_bg_clr])
-        return img
-
-    def _show_pose(self, img: np.ndarray, frm_cnt: int) -> np.ndarray:
-        for cnt, animal_name in enumerate(self.animal_bp_dict.keys()):
-            animal_cols = [self.animal_bp_dict[animal_name]['X_bps'], self.animal_bp_dict[animal_name]['Y_bps']]
-            animal_cols = [item for pair in zip(animal_cols[0], animal_cols[1]) for item in pair]
-            animal_pose_data = self.data_df.loc[frm_cnt, animal_cols].values.astype(np.int32).reshape(-1, 2)
-            for point_idx in range(animal_pose_data.shape[0]):
-                img = cv2.circle(img=img, center=animal_pose_data[point_idx], radius=self.circle_size, color=self.color_palettes[cnt][point_idx], thickness=-1)
-        return img
-
-    def preferences_pop_up(self):
-        if hasattr(self, 'preferences_frm'):
-            self.preferences_frm.destroy()
-        self.preferences_frm = Toplevel()
-        self.preferences_frm.minsize(400, 300)
-        self.preferences_frm.wm_title("PREFERENCES")
-        self.preferences_frm.iconphoto(False, self.btn_icons['settings']["img"])
-
-        pref_frm_panel = CreateLabelFrameWithIcon(parent=self.preferences_frm, header="PREFERENCES", icon_name='settings', padx=5, pady=5)
-        self.show_names_dd = SimBADropDown(parent=pref_frm_panel, dropdown_options=['TRUE', 'FALSE'], label="SHOW ANIMAL NAMES: ", label_width=35, dropdown_width=35, value='TRUE' if self.show_names else 'FALSE', tooltip_key='ROI_POLYGON_TOLERANCE', img='label')
-        self.show_pose_dd = SimBADropDown(parent=pref_frm_panel, dropdown_options=['TRUE', 'FALSE'], label="SHOW ANIMAL POSE: ", label_width=35, dropdown_width=35, value='TRUE' if self.show_pose else 'FALSE', tooltip_key='ROI_KEYBOARD_SENSITIVITY', img='pose')
-        self.name_bg_clr_dd = SimBADropDown(parent=pref_frm_panel, dropdown_options=list(self.clr_dict.keys()), label="NAME BACKGROUND COLOR: ", label_width=35, dropdown_width=35, value=self.text_bg_clr, tooltip_key='ROI_KEYBOARD_SENSITIVITY', img='paint')
-        self.name_bg_opacity_dd = SimBADropDown(parent=pref_frm_panel, dropdown_options=[f"{x / 10:.1f}" for x in range(0, 11)], label="NAME BACKGROUND OPACITY: ", label_width=35, dropdown_width=35, value=self.name_bg_opacity, tooltip_key='ROI_KEYBOARD_SENSITIVITY', img='opacity')
-
-        pref_save_btn = SimbaButton(parent=pref_frm_panel, txt="SAVE", img='save_large', font=Formats.FONT_REGULAR.value, cmd=self._set_preferences)
-        pref_frm_panel.grid(row=0, column=0)
-        self.show_names_dd.grid(row=0, column=0, sticky=NW, pady=5)
-        self.show_pose_dd.grid(row=1, column=0, sticky=NW, pady=5)
-        self.name_bg_clr_dd.grid(row=2, column=0, sticky=NW, pady=5)
-        self.name_bg_opacity_dd.grid(row=3, column=0, sticky=NW, pady=5)
-        pref_save_btn.grid(row=4, column=0, sticky=NW, pady=5)
-        self.status_bar = SimBALabel(parent=self.preferences_frm, txt='', txt_clr='black', bg_clr=None, font=Formats.FONT_REGULAR.value, relief='sunken')
-        self.status_bar.grid(row=1, column=0, sticky='we')
-        self.preferences_frm.grid_rowconfigure(1, weight=0)
-
-    def _set_preferences(self):
-        show_pose_selection = str_2_bool(self.show_pose_dd.get_value())
-        show_name_selection = str_2_bool(self.show_names_dd.get_value())
-        if (show_pose_selection or show_name_selection) and (not isinstance(self.data_df, pd.DataFrame) or self.config_path is None):
-            self.show_names_dd.set_value(value='FALSE')
-            self.show_pose_dd.set_value(value='FALSE')
-            self.status_bar.configure(text='If showing names and/or pose, please pass config_path and data_df', fg='red')
-            self.status_bar.update_idletasks()
-            raise InvalidInputError(msg='If showing names and/or pose, please pass config_path and data_df')
-        else:
-            self.show_pose = deepcopy(show_pose_selection)
-            self.show_names = deepcopy(show_name_selection)
-            self.text_bg_clr = self.name_bg_clr_dd.get_value()
-            self.name_bg_opacity = float(self.name_bg_opacity_dd.get_value())
-            self.load_new_frame(frm_cnt=self.current_frm_number)
-            self.status_bar.configure(text='Updated interactive probability grapher visualization settings', fg='blue')
-            self.status_bar.update_idletasks()
-
-
-
-    def _bind_keys(self):
-        self.main_frm.bind("<Right>", lambda x: self.load_new_frame(frm_cnt=self.current_frm_number + 1))
-        self.main_frm.bind("<Left>", lambda x: self.load_new_frame(frm_cnt=self.current_frm_number - 1))
-        self.main_frm.bind("<Control-l>", lambda x: self.load_new_frame(frm_cnt=self.video_meta_data["frame_count"] - 1))
-        self.main_frm.bind("<Control-o>", lambda x: self.load_new_frame(frm_cnt=0))
-
-    def load_new_frame(self, frm_cnt: int):
-        if (frm_cnt > self.video_meta_data["frame_count"] - 1) or (frm_cnt < 0):
-            FrameRangeWarning(msg=f'Frame {str(frm_cnt)} is outside of the video frame range: (0-{self.video_meta_data["frame_count"]-1}).')
-        else:
-            self.new_frm = read_frm_of_video(video_path=self.cap, frame_index=self.current_frm_number)
-            if self.show_pose:
-                self.new_frm = self._show_pose(img=self.new_frm, frm_cnt=frm_cnt)
-            if self.show_names:
-                self.new_frm = self._show_names(img=self.new_frm, frm_cnt=frm_cnt)
-            self.new_frm = cv2.cvtColor(self.new_frm, cv2.COLOR_RGB2BGR)
-            self.new_frm = Img.fromarray(self.new_frm)
-            self.new_frm.thumbnail(MAX_SIZE, Img.LANCZOS)
-            self.new_frm = ImageTk.PhotoImage(master=self.main_frm, image=self.new_frm)
-            self.img_frm = Label(self.main_frm, image=self.new_frm)
-            self.img_frm.image = self.new_frm
-            self.img_frm.grid(row=0, column=0)
-            self.current_frm_number = frm_cnt
-            self.frame_entry_box.entry_set(val=self.current_frm_number)
+__author__ = "Simon Nilsson; sronilsson@gmail.com"
+
+import os
+from copy import deepcopy
+from tkinter import *
+from typing import List, Optional, Tuple, Union
+
+import cv2
+import numpy as np
+import pandas as pd
+from PIL import Image as Img
+from PIL import ImageTk
+
+from simba.mixins.config_reader import ConfigReader
+from simba.mixins.plotting_mixin import PlottingMixin
+from simba.ui.tkinter_functions import (CreateLabelFrameWithIcon, Entry_Box,
+                                        SimbaButton, SimBADropDown, SimBALabel,
+                                        SimBAScaleBar)
+from simba.utils.checks import (check_file_exist_and_readable,
+                                check_valid_array, check_valid_boolean,
+                                check_valid_dataframe, check_valid_tuple)
+from simba.utils.enums import Formats, TextOptions
+from simba.utils.errors import ColumnNotFoundError, InvalidInputError
+from simba.utils.lookups import (get_color_dict, get_icons_paths,
+                                 get_simba_font_name_and_path)
+from simba.utils.read_write import (get_video_meta_data, read_frm_of_video,
+                                    str_2_bool)
+from simba.utils.warnings import FrameRangeWarning
+
+PADDING = 5
+MAX_SIZE = (1080, 650)
+
+
+class InteractiveVideoPlotterWindow(ConfigReader, PlottingMixin):
+    def __init__(self,
+                 video_path: Union[str, os.PathLike],
+                 p_arr: np.array,
+                 config_path: Optional[Union[str, os.PathLike]] = None,
+                 data_df: Optional[pd.DataFrame] = None,
+                 data_path: Optional[Union[str, os.PathLike]] = None,
+                 show_names: bool = False,
+                 show_pose: bool = False,
+                 max_img_size: Optional[Tuple[int, int]] = None):
+
+        check_file_exist_and_readable(file_path=video_path, raise_error=True)
+        check_valid_array(data=p_arr, source=f'{self.__class__.__name__} p_arr', accepted_dtypes=Formats.NUMERIC_DTYPES.value)
+        check_valid_boolean(value=show_names, source=f'{self.__class__.__name__} show_names', raise_error=True)
+        check_valid_boolean(value=show_pose, source=f'{self.__class__.__name__} show_pose', raise_error=True)
+        if data_path is not None:
+            check_file_exist_and_readable(file_path=data_path)
+        self.data_path = data_path
+        if max_img_size is not None:
+            check_valid_tuple(x=max_img_size, source=f'{self.__class__.__name__} max_img_size', accepted_lengths=(2,), valid_dtypes=(int,), min_integer=1)
+        self.max_img_size = MAX_SIZE if max_img_size is None else tuple(max_img_size)
+        if data_df is not None:
+            check_valid_dataframe(df=data_df, source=f'{self.__class__.__name__} data_df', valid_dtypes=Formats.NUMERIC_DTYPES.value)
+        self.animal_bp_cols, self.animal_clrs = {}, {}
+        if config_path is not None:
+            check_file_exist_and_readable(file_path=config_path, raise_error=True)
+            ConfigReader.__init__(self, config_path=config_path, read_video_info=False, create_logger=False)
+            for animal_name, animal_data in self.animal_bp_dict.items():
+                if len(animal_data['X_bps']) != len(animal_data['Y_bps']):
+                    raise InvalidInputError(msg=f'The animal {animal_name} has {len(animal_data["X_bps"])} X and {len(animal_data["Y_bps"])} Y body-part columns in the SimBA project {config_path}.', source=self.__class__.__name__)
+                self.animal_bp_cols[animal_name] = list(zip(animal_data['X_bps'], animal_data['Y_bps']))
+                self.animal_clrs[animal_name] = [tuple(int(c) for c in clr) for clr in animal_data['colors']]
+        if (show_names or show_pose) and (not isinstance(data_df, pd.DataFrame) or config_path is None):
+            raise InvalidInputError(msg='If showing names and/or pose, please pass config_path and data_df')
+        self.video_meta_data = get_video_meta_data(video_path=video_path)
+        if config_path is not None:   # NOTE: the text and circle sizes are computed for any project - not just when names/pose are displayed at startup - as they can be switched on later through the preferences menu.
+            _, self.font_path = get_simba_font_name_and_path(font=TextOptions.DEFAULT_FONT.value)
+            self.font_size, _, _ = PlottingMixin().get_optimal_font_size_ttf(text=list(self.animal_bp_dict.keys()), font_path=self.font_path, accepted_px_width=int(self.video_meta_data['width']/10), accepted_px_height=int(self.video_meta_data['height']/10))
+            self.circle_size = PlottingMixin().get_optimal_circle_size(frame_size=(self.video_meta_data['width'], self.video_meta_data['height']), circle_frame_ratio=100)
+        else:
+            self.font_path, self.font_size, self.circle_size = None, None, None
+        self.show_names, self.show_pose, self.data_df, self.config_path = show_names, show_pose, data_df, config_path
+        missing_cols = self._get_missing_bp_cols(show_names=self.show_names, show_pose=self.show_pose)
+        if len(missing_cols) > 0:
+            raise ColumnNotFoundError(column_name=', '.join(missing_cols), file_name=self.data_path if self.data_path is not None else 'the passed data frame', source=self.__class__.__name__)
+        self.main_frm = Toplevel()
+        self.btn_icons = get_icons_paths()
+        for k in self.btn_icons.keys():
+            self.btn_icons[k]["img"] = ImageTk.PhotoImage(image=Img.open(os.path.join(os.path.dirname("__file__"), self.btn_icons[k]["icon_path"])))
+        self.main_frm.iconphoto(False, self.btn_icons['frames']['img'])
+        self.current_frm_number, self.jump_size = 0, 0
+        self.img_frm = Frame(self.main_frm)
+        self.clr_dict = get_color_dict()
+        self.text_bg_clr, self.name_bg_opacity = 'Black', 0.4
+        self.img_frm.grid(row=0, column=1, sticky=NW)
+        self.button_frame = Frame(self.main_frm, bd=2, width=700, height=300)
+        self.button_frame.grid(row=1, column=0)
+        self.main_frm.wm_title(self.video_meta_data['video_name'])
+        self.cap = cv2.VideoCapture(video_path)
+        self.max_frm = np.argmax(p_arr)
+        self.frame_id_lbl = SimBALabel(parent=self.button_frame, txt="FRAME NUMBER", justify='center')
+        self.frame_id_lbl.grid(row=0, column=0, pady=(0, PADDING))
+
+        self.nav_frm = Frame(self.button_frame)
+        self.nav_frm.grid(row=1, column=0)
+        self.back_first_frm = SimbaButton(parent=self.nav_frm, txt="<<", cmd=self.load_new_frame, cmd_kwargs={'frm_cnt': lambda: 0})
+        self.back_first_frm.grid(row=0, column=0)
+        self.back_one_frm_btn = SimbaButton(parent=self.nav_frm, txt="<", cmd=self.load_new_frame, cmd_kwargs={'frm_cnt': lambda: self.current_frm_number - 1})
+        self.back_one_frm_btn.grid(row=0, column=1, padx=(0, PADDING))
+        self.frame_entry_box = Entry_Box(parent=self.nav_frm, value=self.current_frm_number, allow_blank=False, justify='center', padx=PADDING, validation='numeric')
+        self.frame_entry_box.grid(row=0, column=2)
+        self.forward_next_frm_btn = SimbaButton(parent=self.nav_frm, txt=">", cmd=self.load_new_frame, cmd_kwargs={'frm_cnt': lambda: self.current_frm_number + 1})
+        self.forward_next_frm_btn.grid(row=0, column=3, padx=(PADDING, 0))
+        self.forward_last_frm_btn = SimbaButton(parent=self.nav_frm, txt=">>", cmd=self.load_new_frame, cmd_kwargs={'frm_cnt': lambda: self.video_meta_data["frame_count"] - 1})
+        self.forward_last_frm_btn.grid(row=0, column=4)
+        self.select_frm_btn = SimbaButton(parent=self.button_frame, txt="Jump to selected frame", cmd=self.load_new_frame, cmd_kwargs={'frm_cnt': lambda: int(self.frame_entry_box.entry_get)}, img='jump')
+        self.select_frm_btn.grid(row=2, column=0, pady=(PADDING, 0))
+
+        self.jump_frm = Frame(self.button_frame)
+        self.jump_frm.grid(row=3, column=0, pady=(PADDING, 0))
+
+        self.jump_size_scale = SimBAScaleBar(parent=self.jump_frm, label="Jump Size:", from_=0, to=100, length=200, value=0)
+        self.jump_size_scale.grid(row=0, column=0)
+        self.jump_back_btn = SimbaButton(parent=self.jump_frm, txt="<<", cmd=self.load_new_frame, cmd_kwargs={'frm_cnt': lambda: self.current_frm_number - self.jump_size_scale.get()})
+        self.jump_back_btn.grid(row=0, column=1, padx=(PADDING, 0))
+        self.jump_forward_btn = SimbaButton(parent=self.jump_frm, txt=">>", cmd=self.load_new_frame, cmd_kwargs={'frm_cnt': lambda: self.current_frm_number + self.jump_size_scale.get()})
+        self.jump_forward_btn.grid(row=0, column=2)
+
+        self.load_new_frame(frm_cnt=self.current_frm_number)
+
+        instructions_frm = Frame(self.main_frm, width=100, height=100)
+        instructions_frm.grid(row=0, column=2, sticky=N)
+
+        key_presses = SimBALabel(parent=instructions_frm, txt="\n\n Keyboard shortcuts for frame navigation: \n Right Arrow = +1 frame"
+                                                "\n Left Arrow = -1 frame"
+                                                "\n Ctrl + l = Last frame"
+                                                "\n Ctrl + o = First frame", font=Formats.FONT_REGULAR_BOLD.value)
+        key_presses.grid(row=0, column=0, sticky=S)
+        self.move_to_highest_p_btn = SimbaButton(parent=instructions_frm, txt="SHOW HIGHEST \n PROBABILITY FRAME", cmd=self.load_new_frame, cmd_kwargs={'frm_cnt': lambda: self.max_frm}, img='pct')
+        self.move_to_highest_p_btn.grid(row=1, column=0, sticky=N, pady=(PADDING, 0))
+
+        self.menu = Menu(self.main_frm, tearoff=0)
+        self.file_menu = Menu(self.menu, tearoff=0)
+        self.menu.add_cascade(label="File...", menu=self.file_menu)
+        self.file_menu.add_command(label="Preferences", compound="left", image=self.btn_icons["settings"]["img"], command=lambda: self.preferences_pop_up())
+        self.main_frm.config(menu=self.menu)
+
+        self._bind_keys()
+
+    def _get_missing_bp_cols(self, show_names: bool, show_pose: bool) -> List[str]:
+        """Return the body-part columns required for drawing animal names and/or pose that are missing from ``self.data_df``."""
+        if not (show_names or show_pose) or not isinstance(self.data_df, pd.DataFrame):
+            return []
+        required_cols = []
+        for bp_cols in self.animal_bp_cols.values():
+            required_cols.extend([c for bp in (bp_cols if show_pose else bp_cols[:1]) for c in bp])
+        df_cols = set(self.data_df.columns)
+        return [c for c in required_cols if c not in df_cols]
+
+    def _show_names(self, img: np.ndarray, frm_cnt: int) -> np.ndarray:
+        for animal_name, bp_cols in self.animal_bp_cols.items():
+            name_loc = self.data_df.loc[frm_cnt, list(bp_cols[0])].values.astype(np.float32)
+            if np.isnan(name_loc).any():
+                continue
+            img = PlottingMixin().put_text(img=img, text=animal_name, pos=(int(name_loc[0]), int(name_loc[1])), font_size=self.font_size, font_path=self.font_path, text_color=self.animal_clrs[animal_name][0], text_bg_alpha=self.name_bg_opacity, text_color_bg=self.clr_dict[self.text_bg_clr])
+        return img
+
+    def _show_pose(self, img: np.ndarray, frm_cnt: int) -> np.ndarray:
+        for animal_name, bp_cols in self.animal_bp_cols.items():
+            animal_pose_data = self.data_df.loc[frm_cnt, [c for bp in bp_cols for c in bp]].values.astype(np.float32).reshape(-1, 2)
+            for bp_idx in range(animal_pose_data.shape[0]):
+                if np.isnan(animal_pose_data[bp_idx]).any():
+                    continue
+                img = cv2.circle(img=img, center=(int(animal_pose_data[bp_idx][0]), int(animal_pose_data[bp_idx][1])), radius=self.circle_size, color=self.animal_clrs[animal_name][bp_idx], thickness=-1)
+        return img
+
+    def preferences_pop_up(self):
+        if hasattr(self, 'preferences_frm'):
+            self.preferences_frm.destroy()
+        self.preferences_frm = Toplevel()
+        self.preferences_frm.minsize(400, 300)
+        self.preferences_frm.wm_title("PREFERENCES")
+        self.preferences_frm.iconphoto(False, self.btn_icons['settings']["img"])
+
+        pref_frm_panel = CreateLabelFrameWithIcon(parent=self.preferences_frm, header="PREFERENCES", icon_name='settings', padx=5, pady=5)
+        self.show_names_dd = SimBADropDown(parent=pref_frm_panel, dropdown_options=['TRUE', 'FALSE'], label="SHOW ANIMAL NAMES: ", label_width=35, dropdown_width=35, value='TRUE' if self.show_names else 'FALSE', tooltip_key='ROI_POLYGON_TOLERANCE', img='label')
+        self.show_pose_dd = SimBADropDown(parent=pref_frm_panel, dropdown_options=['TRUE', 'FALSE'], label="SHOW ANIMAL POSE: ", label_width=35, dropdown_width=35, value='TRUE' if self.show_pose else 'FALSE', tooltip_key='ROI_KEYBOARD_SENSITIVITY', img='pose')
+        self.name_bg_clr_dd = SimBADropDown(parent=pref_frm_panel, dropdown_options=list(self.clr_dict.keys()), label="NAME BACKGROUND COLOR: ", label_width=35, dropdown_width=35, value=self.text_bg_clr, tooltip_key='ROI_KEYBOARD_SENSITIVITY', img='paint')
+        self.name_bg_opacity_dd = SimBADropDown(parent=pref_frm_panel, dropdown_options=[f"{x / 10:.1f}" for x in range(0, 11)], label="NAME BACKGROUND OPACITY: ", label_width=35, dropdown_width=35, value=self.name_bg_opacity, tooltip_key='ROI_KEYBOARD_SENSITIVITY', img='opacity')
+
+        pref_save_btn = SimbaButton(parent=pref_frm_panel, txt="SAVE", img='save_large', font=Formats.FONT_REGULAR.value, cmd=self._set_preferences)
+        pref_frm_panel.grid(row=0, column=0)
+        self.show_names_dd.grid(row=0, column=0, sticky=NW, pady=5)
+        self.show_pose_dd.grid(row=1, column=0, sticky=NW, pady=5)
+        self.name_bg_clr_dd.grid(row=2, column=0, sticky=NW, pady=5)
+        self.name_bg_opacity_dd.grid(row=3, column=0, sticky=NW, pady=5)
+        pref_save_btn.grid(row=4, column=0, sticky=NW, pady=5)
+        self.status_bar = SimBALabel(parent=self.preferences_frm, txt='', txt_clr='black', bg_clr=None, font=Formats.FONT_REGULAR.value, relief='sunken')
+        self.status_bar.grid(row=1, column=0, sticky='we')
+        self.preferences_frm.grid_rowconfigure(1, weight=0)
+
+    def _set_preferences(self):
+        show_pose_selection = str_2_bool(self.show_pose_dd.get_value())
+        show_name_selection = str_2_bool(self.show_names_dd.get_value())
+        missing_cols = self._get_missing_bp_cols(show_names=show_name_selection, show_pose=show_pose_selection)
+        if (show_pose_selection or show_name_selection) and (not isinstance(self.data_df, pd.DataFrame) or self.config_path is None):
+            self.show_names_dd.set_value(value='FALSE')
+            self.show_pose_dd.set_value(value='FALSE')
+            self.status_bar.configure(text='If showing names and/or pose, please pass config_path and data_df', fg='red')
+            self.status_bar.update_idletasks()
+            raise InvalidInputError(msg='If showing names and/or pose, please pass config_path and data_df')
+        elif len(missing_cols) > 0:
+            self.show_names_dd.set_value(value='TRUE' if self.show_names else 'FALSE')
+            self.show_pose_dd.set_value(value='TRUE' if self.show_pose else 'FALSE')
+            self.status_bar.configure(text=f'The column(s) {missing_cols} could not be found in the data: names and pose cannot be displayed', fg='red')
+            self.status_bar.update_idletasks()
+        else:
+            self.show_pose = deepcopy(show_pose_selection)
+            self.show_names = deepcopy(show_name_selection)
+            self.text_bg_clr = self.name_bg_clr_dd.get_value()
+            self.name_bg_opacity = float(self.name_bg_opacity_dd.get_value())
+            self.load_new_frame(frm_cnt=self.current_frm_number)
+            self.status_bar.configure(text='Updated interactive probability grapher visualization settings', fg='blue')
+            self.status_bar.update_idletasks()
+
+
+
+    def _bind_keys(self):
+        self.main_frm.bind("<Right>", lambda x: self.load_new_frame(frm_cnt=self.current_frm_number + 1))
+        self.main_frm.bind("<Left>", lambda x: self.load_new_frame(frm_cnt=self.current_frm_number - 1))
+        self.main_frm.bind("<Control-l>", lambda x: self.load_new_frame(frm_cnt=self.video_meta_data["frame_count"] - 1))
+        self.main_frm.bind("<Control-o>", lambda x: self.load_new_frame(frm_cnt=0))
+
+    def load_new_frame(self, frm_cnt: int):
+        if (frm_cnt > self.video_meta_data["frame_count"] - 1) or (frm_cnt < 0):
+            FrameRangeWarning(msg=f'Frame {str(frm_cnt)} is outside of the video frame range: (0-{self.video_meta_data["frame_count"]-1}).')
+        else:
+            self.new_frm = read_frm_of_video(video_path=self.cap, frame_index=frm_cnt)
+            if self.show_pose or self.show_names:
+                if frm_cnt in self.data_df.index:
+                    if self.show_pose:
+                        self.new_frm = self._show_pose(img=self.new_frm, frm_cnt=frm_cnt)
+                    if self.show_names:
+                        self.new_frm = self._show_names(img=self.new_frm, frm_cnt=frm_cnt)
+                else:
+                    FrameRangeWarning(msg=f'Frame {str(frm_cnt)} is outside of the data frame range: (0-{len(self.data_df)-1}). Pose and animal names cannot be displayed.', source=self.__class__.__name__)
+            self.new_frm = cv2.cvtColor(self.new_frm, cv2.COLOR_RGB2BGR)
+            self.new_frm = Img.fromarray(self.new_frm)
+            self.new_frm.thumbnail(self.max_img_size, Img.LANCZOS)
+            self.new_frm = ImageTk.PhotoImage(master=self.main_frm, image=self.new_frm)
+            self.img_frm = Label(self.main_frm, image=self.new_frm)
+            self.img_frm.image = self.new_frm
+            self.img_frm.grid(row=0, column=0)
+            self.current_frm_number = frm_cnt
+            self.frame_entry_box.entry_set(val=self.current_frm_number)

{simba_uw_tf_dev-5.5.2 → simba_uw_tf_dev-5.5.3}/simba_uw_tf_dev.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: simba-uw-tf-dev
-Version: 5.5.2
+Version: 5.5.3
 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