project_name
stringlengths
6
104
file_name
stringlengths
4
89
full_name
stringlengths
1
102
func_name
stringlengths
1
85
docstring
stringlengths
13
836
docstring_tokens
listlengths
4
122
code
stringlengths
23
39.7k
code_tokens
stringlengths
29
44.6k
url
int64
3
986k
ZumoLabs/zpy
color.py
irgb_to_frgb
irgb_to_frgb
Convert integer rgb (0 to 255) to float rgb (0 to 1).
[ "Convert", "integer", "rgb", "(0", "to", "255)", "to", "float", "rgb", "(0", "to", "1)." ]
def irgb_to_frgb(irgb: Tuple[int]) -> Tuple[float]: max_rgb_value = 255 return tuple((x / max_rgb_value for x in irgb))
['def', 'irgb_to_frgb(irgb:', 'Tuple[int])', '->', 'Tuple[float]:', 'max_rgb_value', '=', '255', 'return', 'tuple((x', '/', 'max_rgb_value', 'for', 'x', 'in', 'irgb))']
971,999
ZumoLabs/zpy
color.py
irgb_to_hex
irgb_to_hex
Convert integer rgb (0 to 255) to hex.
[ "Convert", "integer", "rgb", "(0", "to", "255)", "to", "hex." ]
def irgb_to_hex(irgb: Tuple[int]) -> str: (r, g, b) = irgb return '#%02x%02x%02x' % (r, g, b)
['def', 'irgb_to_hex(irgb:', 'Tuple[int])', '->', 'str:', '(r,', 'g,', 'b)', '=', 'irgb', 'return', "'#%02x%02x%02x'", '%', '(r,', 'g,', 'b)']
972,000
ZumoLabs/zpy
color.py
frgb_to_irgb
frgb_to_irgb
Convert float rgb (0 to 1) to integer rgb (0 to 255).
[ "Convert", "float", "rgb", "(0", "to", "1)", "to", "integer", "rgb", "(0", "to", "255)." ]
def frgb_to_irgb(frgb: Tuple[float]) -> Tuple[int]: max_rgb_value = 255 return tuple((int(x * max_rgb_value) for x in frgb))
['def', 'frgb_to_irgb(frgb:', 'Tuple[float])', '->', 'Tuple[int]:', 'max_rgb_value', '=', '255', 'return', 'tuple((int(x', '*', 'max_rgb_value)', 'for', 'x', 'in', 'frgb))']
972,001
ZumoLabs/zpy
color.py
frgb_to_srgba
frgb_to_srgba
Convert float rgb (0 to 1) to the gamma-corrected sRGBA float (0 to 1).
[ "Convert", "float", "rgb", "(0", "to", "1)", "to", "the", "gamma-corrected", "sRGBA", "float", "(0", "to", "1)." ]
def frgb_to_srgba(frgb: Tuple[float], a=1.0) -> Tuple[float]: srgb = frgb_to_srgb(frgb) srgba = frgb_to_frgba(srgb, a=a) return srgba
['def', 'frgb_to_srgba(frgb:', 'Tuple[float],', 'a=1.0)', '->', 'Tuple[float]:', 'srgb', '=', 'frgb_to_srgb(frgb)', 'srgba', '=', 'frgb_to_frgba(srgb,', 'a=a)', 'return', 'srgba']
972,004
ZumoLabs/zpy
color.py
closest_color
closest_color
Get the index of the closest color in a list to the input color.
[ "Get", "the", "index", "of", "the", "closest", "color", "in", "a", "list", "to", "the", "input", "color." ]
def closest_color(color: Tuple[float], colors: List[Tuple[float]], max_dist: float=0.01) -> Union[None, Tuple[float]]: min_dist = 3.0 nearest_idx = 0 for (i, _color) in enumerate(colors): dist = (color[0] - _color[0]) ** 2 + (color[1] - _color[1]) ** 2 + (color[2] - _color[2]) ** 2 if dist <...
['def', 'closest_color(color:', 'Tuple[float],', 'colors:', 'List[Tuple[float]],', 'max_dist:', 'float=0.01)', '->', 'Union[None,', 'Tuple[float]]:', 'min_dist', '=', '3.0', 'nearest_idx', '=', '0', 'for', '(i,', '_color)', 'in', 'enumerate(colors):', 'dist', '=', '(color[0]', '-', '_color[0])', '**', '2', '+', '(color...
972,006
ZumoLabs/zpy
files.py
dataset_contents
dataset_contents
Use regex to search inside a data directory.
[ "Use", "regex", "to", "search", "inside", "a", "data", "directory." ]
def dataset_contents(path: Union[Path, str], filetype_regex: Dict=FILE_REGEX) -> Dict: path = verify_path(path, check_dir=True, make=False) contents = {'dirs': []} for (dirpath, _, files) in os.walk(path): contents['dirs'].append(dirpath) for filename in files: for (name, re_patt...
['def', 'dataset_contents(path:', 'Union[Path,', 'str],', 'filetype_regex:', 'Dict=FILE_REGEX)', '->', 'Dict:', 'path', '=', 'verify_path(path,', 'check_dir=True,', 'make=False)', 'contents', '=', "{'dirs':", '[]}', 'for', '(dirpath,', '_,', 'files)', 'in', 'os.walk(path):', "contents['dirs'].append(dirpath)", 'for', '...
972,007
ZumoLabs/zpy
files.py
make_rgb_image_name
make_rgb_image_name
Creates a RGB image name given an integer id.
[ "Creates", "a", "RGB", "image", "name", "given", "an", "integer", "id." ]
def make_rgb_image_name(id: int, extension: str='.png') -> str: return 'image.%06d.rgb' % id + extension
['def', 'make_rgb_image_name(id:', 'int,', 'extension:', "str='.png')", '->', 'str:', 'return', "'image.%06d.rgb'", '%', 'id', '+', 'extension']
972,009
ZumoLabs/zpy
files.py
make_cseg_image_name
make_cseg_image_name
Return category (class) segmentation image name from integer id.
[ "Return", "category", "(class)", "segmentation", "image", "name", "from", "integer", "id." ]
def make_cseg_image_name(id: int, extension: str='.png') -> str: return 'image.%06d.cseg' % id + extension
['def', 'make_cseg_image_name(id:', 'int,', 'extension:', "str='.png')", '->', 'str:', 'return', "'image.%06d.cseg'", '%', 'id', '+', 'extension']
972,010
ZumoLabs/zpy
files.py
make_iseg_image_name
make_iseg_image_name
Return instance segmentation image name from integer id.
[ "Return", "instance", "segmentation", "image", "name", "from", "integer", "id." ]
def make_iseg_image_name(id: int, extension: str='.png') -> str: return 'image.%06d.iseg' % id + extension
['def', 'make_iseg_image_name(id:', 'int,', 'extension:', "str='.png')", '->', 'str:', 'return', "'image.%06d.iseg'", '%', 'id', '+', 'extension']
972,011
ZumoLabs/zpy
files.py
id_from_image_name
id_from_image_name
Extract integer id from image name.
[ "Extract", "integer", "id", "from", "image", "name." ]
def id_from_image_name(image_name: str) -> int: return int(''.join([s for s in image_name if s.isdigit()]))
['def', 'id_from_image_name(image_name:', 'str)', '->', 'int:', 'return', "int(''.join([s", 'for', 's', 'in', 'image_name', 'if', 's.isdigit()]))']
972,014
ZumoLabs/zpy
files.py
replace_id_in_image_name
replace_id_in_image_name
Replace the integer id in an image name.
[ "Replace", "the", "integer", "id", "in", "an", "image", "name." ]
def replace_id_in_image_name(image_name: str, new_id: int) -> str: return 'image.%06d' % new_id + image_name[12:]
['def', 'replace_id_in_image_name(image_name:', 'str,', 'new_id:', 'int)', '->', 'str:', 'return', "'image.%06d'", '%', 'new_id', '+', 'image_name[12:]']
972,015
ZumoLabs/zpy
files.py
clean_dir
clean_dir
Delete everything at the provided directory.
[ "Delete", "everything", "at", "the", "provided", "directory." ]
def clean_dir(path: Union[Path, str], keep_dir: bool=True) -> None: path = verify_path(path, make=False, check_dir=True) if keep_dir: for _path in path.iterdir(): try: if _path.is_file() or _path.is_symlink(): _path.unlink() elif _path.is_d...
['def', 'clean_dir(path:', 'Union[Path,', 'str],', 'keep_dir:', 'bool=True)', '->', 'None:', 'path', '=', 'verify_path(path,', 'make=False,', 'check_dir=True)', 'if', 'keep_dir:', 'for', '_path', 'in', 'path.iterdir():', 'try:', 'if', '_path.is_file()', 'or', '_path.is_symlink():', '_path.unlink()', 'elif', '_path.is_d...
972,019
ZumoLabs/zpy
files.py
verify_path
verify_path
Checks to make sure Path exists and optionally creates it.
[ "Checks", "to", "make", "sure", "Path", "exists", "and", "optionally", "creates", "it." ]
def verify_path(path: Union[Path, str], make: bool=False, check_dir: bool=False) -> Path: path = to_pathlib_path(path) if not path.exists(): log.warning(f'Could not find path at {path}') if make: log.info(f'Making {path.name} dir at {path}') path.mkdir(exist_ok=True, pare...
['def', 'verify_path(path:', 'Union[Path,', 'str],', 'make:', 'bool=False,', 'check_dir:', 'bool=False)', '->', 'Path:', 'path', '=', 'to_pathlib_path(path)', 'if', 'not', 'path.exists():', "log.warning(f'Could", 'not', 'find', 'path', 'at', "{path}')", 'if', 'make:', "log.info(f'Making", '{path.name}', 'dir', 'at', "{...
972,020
ZumoLabs/zpy
files.py
read_json
read_json
Read a json from a path.
[ "Read", "a", "json", "from", "a", "path." ]
def read_json(path: Union[Path, str]) -> Union[Dict, List]: path = to_pathlib_path(path) if not path.suffix == '.json': raise ValueError(f'{path} is not a JSON file.') log.info(f'Reading JSON file at {path}') with path.open() as f: data = json.load(f) return data
['def', 'read_json(path:', 'Union[Path,', 'str])', '->', 'Union[Dict,', 'List]:', 'path', '=', 'to_pathlib_path(path)', 'if', 'not', 'path.suffix', '==', "'.json':", 'raise', "ValueError(f'{path}", 'is', 'not', 'a', 'JSON', "file.')", "log.info(f'Reading", 'JSON', 'file', 'at', "{path}')", 'with', 'path.open()', 'as', ...
972,022
ZumoLabs/zpy
gin.py
parse_gin_bindings
parse_gin_bindings
Parse any extra gin bindings to the config.
[ "Parse", "any", "extra", "gin", "bindings", "to", "the", "config." ]
def parse_gin_bindings(gin_bindings: Dict=None) -> None: if gin_bindings is None: log.info('No additional gin bindings to parse') else: log.info(f'Parsing additional bindings: {pformat(gin_bindings)}') with gin.unlock_config(): for (key, value) in replace_human_redable_kwargs...
['def', 'parse_gin_bindings(gin_bindings:', 'Dict=None)', '->', 'None:', 'if', 'gin_bindings', 'is', 'None:', "log.info('No", 'additional', 'gin', 'bindings', 'to', "parse')", 'else:', "log.info(f'Parsing", 'additional', 'bindings:', "{pformat(gin_bindings)}')", 'with', 'gin.unlock_config():', 'for', '(key,', 'value)',...
972,032
ZumoLabs/zpy
hdris.py
load_hdri
load_hdri
Load an HDRI from path.
[ "Load", "an", "HDRI", "from", "path." ]
def load_hdri(path: Union[Path, str], scale: Tuple[float]=(1.0, 1.0, 1.0), random_z_rot: bool=True) -> None: scene = zpy.blender.verify_blender_scene() scene.world.use_nodes = True tree = scene.world.node_tree out_node = zpy.nodes.get_or_make('World Output', 'ShaderNodeOutputWorld', tree, pos=(0, 0)) ...
['def', 'load_hdri(path:', 'Union[Path,', 'str],', 'scale:', 'Tuple[float]=(1.0,', '1.0,', '1.0),', 'random_z_rot:', 'bool=True)', '->', 'None:', 'scene', '=', 'zpy.blender.verify_blender_scene()', 'scene.world.use_nodes', '=', 'True', 'tree', '=', 'scene.world.node_tree', 'out_node', '=', "zpy.nodes.get_or_make('World...
972,035
ZumoLabs/zpy
image.py
open_image
open_image
Open image from path to ndarray.
[ "Open", "image", "from", "path", "to", "ndarray." ]
def open_image(image_path: Union[Path, str]) -> np.ndarray: image_path = zpy.files.verify_path(image_path, make=False) img = None try: img = io.imread(image_path) if img.shape[2] > 3: log.debug('RGBA image detected!') img = img[:, :, :3] if img.max() > 2.0: ...
['def', 'open_image(image_path:', 'Union[Path,', 'str])', '->', 'np.ndarray:', 'image_path', '=', 'zpy.files.verify_path(image_path,', 'make=False)', 'img', '=', 'None', 'try:', 'img', '=', 'io.imread(image_path)', 'if', 'img.shape[2]', '>', '3:', "log.debug('RGBA", 'image', "detected!')", 'img', '=', 'img[:,', ':,', '...
972,037
ZumoLabs/zpy
image.py
remove_alpha_channel
remove_alpha_channel
Remove the alpha channel in an image (overwrites image).
[ "Remove", "the", "alpha", "channel", "in", "an", "image", "(overwrites", "image)." ]
def remove_alpha_channel(image_path: Union[Path, str]) -> None: img = open_image(image_path) io.imsave(image_path, img) log.info(f'Saving image with no alpha channel at {image_path}')
['def', 'remove_alpha_channel(image_path:', 'Union[Path,', 'str])', '->', 'None:', 'img', '=', 'open_image(image_path)', 'io.imsave(image_path,', 'img)', "log.info(f'Saving", 'image', 'with', 'no', 'alpha', 'channel', 'at', "{image_path}')"]
972,038
ZumoLabs/zpy
image.py
jpeg_compression
jpeg_compression
Add jpeg compression to an image (overwrites image).
[ "Add", "jpeg", "compression", "to", "an", "image", "(overwrites", "image)." ]
def jpeg_compression(image_path: Union[Path, str], quality: int=40) -> Path: image_path = zpy.files.verify_path(image_path, make=False) img = io.imread(image_path) if not image_path.suffix == '.jpeg': image_path = image_path.with_suffix('.jpeg') io.imsave(image_path, arr=img, quality=quality) ...
['def', 'jpeg_compression(image_path:', 'Union[Path,', 'str],', 'quality:', 'int=40)', '->', 'Path:', 'image_path', '=', 'zpy.files.verify_path(image_path,', 'make=False)', 'img', '=', 'io.imread(image_path)', 'if', 'not', 'image_path.suffix', '==', "'.jpeg':", 'image_path', '=', "image_path.with_suffix('.jpeg')", 'io....
972,039
ZumoLabs/zpy
image.py
resize_image
resize_image
Resize an image (overwrites image).
[ "Resize", "an", "image", "(overwrites", "image)." ]
def resize_image(image_path: Union[Path, str], width: int=640, height: int=480) -> Path: img = open_image(image_path) resized_img = resize(img, (height, width), anti_aliasing=True) io.imsave(image_path, resized_img)
['def', 'resize_image(image_path:', 'Union[Path,', 'str],', 'width:', 'int=640,', 'height:', 'int=480)', '->', 'Path:', 'img', '=', 'open_image(image_path)', 'resized_img', '=', 'resize(img,', '(height,', 'width),', 'anti_aliasing=True)', 'io.imsave(image_path,', 'resized_img)']
972,040
ZumoLabs/zpy
kdtree.py
volume_occupancy
volume_occupancy
Get occupancy percentage for volume.
[ "Get", "occupancy", "percentage", "for", "volume." ]
def volume_occupancy(kdtree: mathutils.kdtree.KDTree, x_bounds: Tuple[float], y_bounds: Tuple[float], z_bounds: Tuple[float], num_voxels: int=100) -> float: log.info('Calculating volume occupancy ....') x_side_length = abs(x_bounds[1] - x_bounds[0]) y_side_length = abs(y_bounds[1] - y_bounds[0]) z_side_...
['def', 'volume_occupancy(kdtree:', 'mathutils.kdtree.KDTree,', 'x_bounds:', 'Tuple[float],', 'y_bounds:', 'Tuple[float],', 'z_bounds:', 'Tuple[float],', 'num_voxels:', 'int=100)', '->', 'float:', "log.info('Calculating", 'volume', 'occupancy', "....')", 'x_side_length', '=', 'abs(x_bounds[1]', '-', 'x_bounds[0])', 'y_...
972,048
ZumoLabs/zpy
keypoints.py
Keypoints.update
update
Add a keypoint skeleton.
[ "Add", "a", "keypoint", "skeleton." ]
def update(self, world_transform=None) -> None: self.num_keypoints = 0 self.keypoints_xyv = [] self.keypoints_xyz = [] for (name, bone_name) in self.bone_lookup.items(): bone = self.bones.get(bone_name, None) if bone is None: log.warning(f'Could not find keypoint bone {name} ...
['def', 'update(self,', 'world_transform=None)', '->', 'None:', 'self.num_keypoints', '=', '0', 'self.keypoints_xyv', '=', '[]', 'self.keypoints_xyz', '=', '[]', 'for', '(name,', 'bone_name)', 'in', 'self.bone_lookup.items():', 'bone', '=', 'self.bones.get(bone_name,', 'None)', 'if', 'bone', 'is', 'None:', "log.warning...
972,049
ZumoLabs/zpy
logging.py
linebreaker_log
linebreaker_log
Good looking line-breaker log message.
[ "Good", "looking", "line-breaker", "log", "message." ]
def linebreaker_log(message: str, line_length: int=80): message = message[:line_length] whitespace = ' ' * int((line_length - len(message)) / 2) log.info('-' * line_length) log.info(f'{whitespace}{message.upper()}{whitespace}') log.info('-' * line_length)
['def', 'linebreaker_log(message:', 'str,', 'line_length:', 'int=80):', 'message', '=', 'message[:line_length]', 'whitespace', '=', "'", "'", '*', 'int((line_length', '-', 'len(message))', '/', '2)', "log.info('-'", '*', 'line_length)', "log.info(f'{whitespace}{message.upper()}{whitespace}')", "log.info('-'", '*', 'lin...
972,051
ZumoLabs/zpy
material.py
verify
verify
Get a material given either its name or the object itself.
[ "Get", "a", "material", "given", "either", "its", "name", "or", "the", "object", "itself." ]
def verify(mat: Union[bpy.types.Material, str], check_none: bool=True) -> bpy.types.Material: if isinstance(mat, str): mat = bpy.data.materials.get(mat) if check_none and mat is None: raise ValueError(f'Could not find material {mat}.') return mat
['def', 'verify(mat:', 'Union[bpy.types.Material,', 'str],', 'check_none:', 'bool=True)', '->', 'bpy.types.Material:', 'if', 'isinstance(mat,', 'str):', 'mat', '=', 'bpy.data.materials.get(mat)', 'if', 'check_none', 'and', 'mat', 'is', 'None:', 'raise', "ValueError(f'Could", 'not', 'find', 'material', "{mat}.')", 'retu...
972,054
ZumoLabs/zpy
material.py
restore_mat_props
restore_mat_props
Restore an object to a position.
[ "Restore", "an", "object", "to", "a", "position." ]
def restore_mat_props(mat: Union[bpy.types.Material, str]) -> None: log.info(f'Restoring material properties for {mat.name}') set_mat_props(mat, _SAVED_MATERIALS[mat.name])
['def', 'restore_mat_props(mat:', 'Union[bpy.types.Material,', 'str])', '->', 'None:', "log.info(f'Restoring", 'material', 'properties', 'for', "{mat.name}')", 'set_mat_props(mat,', '_SAVED_MATERIALS[mat.name])']
972,057
ZumoLabs/zpy
material.py
restore_all_mat_props
restore_all_mat_props
Restore all jittered materials to original look.
[ "Restore", "all", "jittered", "materials", "to", "original", "look." ]
def restore_all_mat_props() -> None: for (mat_name, mat_props) in _SAVED_MATERIALS.items(): set_mat_props(mat_name, mat_props)
['def', 'restore_all_mat_props()', '->', 'None:', 'for', '(mat_name,', 'mat_props)', 'in', '_SAVED_MATERIALS.items():', 'set_mat_props(mat_name,', 'mat_props)']
972,058
ZumoLabs/zpy
material.py
get_mat_props
get_mat_props
Get (some of the) material properties.
[ "Get", "(some", "of", "the)", "material", "properties." ]
def get_mat_props(mat: Union[bpy.types.Material, str]) -> Tuple[float]: mat = verify(mat) bsdf_node = mat.node_tree.nodes.get('Principled BSDF') if bsdf_node is None: log.warning(f'No BSDF node in {mat.name}') return (0.0, 0.0, 0.0) return (bsdf_node.inputs['Roughness'].default_value, bs...
['def', 'get_mat_props(mat:', 'Union[bpy.types.Material,', 'str])', '->', 'Tuple[float]:', 'mat', '=', 'verify(mat)', 'bsdf_node', '=', "mat.node_tree.nodes.get('Principled", "BSDF')", 'if', 'bsdf_node', 'is', 'None:', "log.warning(f'No", 'BSDF', 'node', 'in', "{mat.name}')", 'return', '(0.0,', '0.0,', '0.0)', 'return'...
972,059
ZumoLabs/zpy
material.py
set_mat_props
set_mat_props
Set (some of the) material properties.
[ "Set", "(some", "of", "the)", "material", "properties." ]
def set_mat_props(mat: Union[bpy.types.Material, str], prop_tuple: Tuple[float]) -> None: mat = verify(mat) bsdf_node = mat.node_tree.nodes.get('Principled BSDF', None) if bsdf_node is None: log.warning(f'No BSDF node in {mat.name}') return bsdf_node.inputs['Roughness'].default_value = c...
['def', 'set_mat_props(mat:', 'Union[bpy.types.Material,', 'str],', 'prop_tuple:', 'Tuple[float])', '->', 'None:', 'mat', '=', 'verify(mat)', 'bsdf_node', '=', "mat.node_tree.nodes.get('Principled", "BSDF',", 'None)', 'if', 'bsdf_node', 'is', 'None:', "log.warning(f'No", 'BSDF', 'node', 'in', "{mat.name}')", 'return', ...
972,060
ZumoLabs/zpy
material.py
jitter
jitter
Randomize an existing material a little.
[ "Randomize", "an", "existing", "material", "a", "little." ]
def jitter(mat: Union[bpy.types.Material, str], std: float=0.2, save_first_time: bool=True) -> None: mat = verify(mat) if save_first_time: if _SAVED_MATERIALS.get(mat.name, None) is None: save_mat_props(mat) else: restore_mat_props(mat) log.info(f'Jittering material {...
['def', 'jitter(mat:', 'Union[bpy.types.Material,', 'str],', 'std:', 'float=0.2,', 'save_first_time:', 'bool=True)', '->', 'None:', 'mat', '=', 'verify(mat)', 'if', 'save_first_time:', 'if', '_SAVED_MATERIALS.get(mat.name,', 'None)', 'is', 'None:', 'save_mat_props(mat)', 'else:', 'restore_mat_props(mat)', "log.info(f'J...
972,061
ZumoLabs/zpy
material.py
make_mat_from_color
make_mat_from_color
Makes a material given a color.
[ "Makes", "a", "material", "given", "a", "color." ]
def make_mat_from_color(color: Tuple[float], name: str=None) -> bpy.types.Material: if name is None: name = str(color) mat = bpy.data.materials.get(name, None) if mat is None: log.debug(f'Material {name} does not exist, creating it.') mat = bpy.data.materials.new(name=name) mat.u...
['def', 'make_mat_from_color(color:', 'Tuple[float],', 'name:', 'str=None)', '->', 'bpy.types.Material:', 'if', 'name', 'is', 'None:', 'name', '=', 'str(color)', 'mat', '=', 'bpy.data.materials.get(name,', 'None)', 'if', 'mat', 'is', 'None:', "log.debug(f'Material", '{name}', 'does', 'not', 'exist,', 'creating', "it.')...
972,065
ZumoLabs/zpy
material.py
set_mat
set_mat
Set the material for an object.
[ "Set", "the", "material", "for", "an", "object." ]
def set_mat(obj: Union[bpy.types.Object, str], mat: Union[bpy.types.Material, str], recursive: bool=True) -> None: obj = zpy.objects.verify(obj) mat = zpy.material.verify(mat) if hasattr(obj, 'active_material'): log.debug(f'Setting object {obj.name} material {mat.name}') obj.active_material ...
['def', 'set_mat(obj:', 'Union[bpy.types.Object,', 'str],', 'mat:', 'Union[bpy.types.Material,', 'str],', 'recursive:', 'bool=True)', '->', 'None:', 'obj', '=', 'zpy.objects.verify(obj)', 'mat', '=', 'zpy.material.verify(mat)', 'if', 'hasattr(obj,', "'active_material'):", "log.debug(f'Setting", 'object', '{obj.name}', ...
972,066
ZumoLabs/zpy
ml.py
log
log
Log an update to experiment.
[ "Log", "an", "update", "to", "experiment." ]
def log(metrics: str=None, file_path: str=None) -> None: global experiment exp = experiment if file_path: file_path = Path(file_path).resolve() exp._update(file_path=file_path, metrics=metrics)
['def', 'log(metrics:', 'str=None,', 'file_path:', 'str=None)', '->', 'None:', 'global', 'experiment', 'exp', '=', 'experiment', 'if', 'file_path:', 'file_path', '=', 'Path(file_path).resolve()', 'exp._update(file_path=file_path,', 'metrics=metrics)']
972,069
ZumoLabs/zpy
nodes.py
get_or_make
get_or_make
Verify existence or create a node.
[ "Verify", "existence", "or", "create", "a", "node." ]
def get_or_make(name: str, node_type: str, tree: bpy.types.NodeTree, label_tag: str='(zpy) ', pos: Tuple[float]=None) -> bpy.types.Node: node = tree.nodes.get(name, None) if node is None: node = tree.nodes.new(node_type) node.name = name node.label = f'{label_tag}{name}' node.bl_descript...
['def', 'get_or_make(name:', 'str,', 'node_type:', 'str,', 'tree:', 'bpy.types.NodeTree,', 'label_tag:', "str='(zpy)", "',", 'pos:', 'Tuple[float]=None)', '->', 'bpy.types.Node:', 'node', '=', 'tree.nodes.get(name,', 'None)', 'if', 'node', 'is', 'None:', 'node', '=', 'tree.nodes.new(node_type)', 'node.name', '=', 'name...
972,070
ZumoLabs/zpy
objects.py
verify
verify
Return object given name or Object type object.
[ "Return", "object", "given", "name", "or", "Object", "type", "object." ]
def verify(obj: Union[bpy.types.Object, str], check_none=True) -> bpy.types.Object: if isinstance(obj, str): obj = bpy.data.objects.get(obj) if check_none and obj is None: raise ValueError(f'Could not find object {obj}.') return obj
['def', 'verify(obj:', 'Union[bpy.types.Object,', 'str],', 'check_none=True)', '->', 'bpy.types.Object:', 'if', 'isinstance(obj,', 'str):', 'obj', '=', 'bpy.data.objects.get(obj)', 'if', 'check_none', 'and', 'obj', 'is', 'None:', 'raise', "ValueError(f'Could", 'not', 'find', 'object', "{obj}.')", 'return', 'obj']
972,072
ZumoLabs/zpy
objects.py
delete_obj_context
delete_obj_context
Alternative way to delete an object.
[ "Alternative", "way", "to", "delete", "an", "object." ]
def delete_obj_context(obj: Union[bpy.types.Object, str]) -> None: obj = verify(obj) log.debug(f'Removing obj: {obj.name}') context_remove = bpy.context.copy() context_remove['selected_objects'] = [obj] bpy.ops.object.delete(context_remove)
['def', 'delete_obj_context(obj:', 'Union[bpy.types.Object,', 'str])', '->', 'None:', 'obj', '=', 'verify(obj)', "log.debug(f'Removing", 'obj:', "{obj.name}')", 'context_remove', '=', 'bpy.context.copy()', "context_remove['selected_objects']", '=', '[obj]', 'bpy.ops.object.delete(context_remove)']
972,074
ZumoLabs/zpy
objects.py
randomly_hide_within_collection
randomly_hide_within_collection
Randomly hide objects in a list of collections.
[ "Randomly", "hide", "objects", "in", "a", "list", "of", "collections." ]
def randomly_hide_within_collection(collections: List[bpy.types.Collection], chance_to_hide: float=0.9) -> None: to_hide = [] for obj in for_obj_in_collections(collections): if random.random() < chance_to_hide: to_hide.append(obj.name) for name in to_hide: bpy.data.objects[name]....
['def', 'randomly_hide_within_collection(collections:', 'List[bpy.types.Collection],', 'chance_to_hide:', 'float=0.9)', '->', 'None:', 'to_hide', '=', '[]', 'for', 'obj', 'in', 'for_obj_in_collections(collections):', 'if', 'random.random()', '<', 'chance_to_hide:', 'to_hide.append(obj.name)', 'for', 'name', 'in', 'to_h...
972,080
ZumoLabs/zpy
objects.py
populate_vertex_colors
populate_vertex_colors
Fill the given Vertex Color Layer with the color parameter values.
[ "Fill", "the", "given", "Vertex", "Color", "Layer", "with", "the", "color", "parameter", "values." ]
def populate_vertex_colors(obj: Union[bpy.types.Object, str], color_rgba: Tuple[float], seg_type: str='instance') -> None: obj = verify(obj) if not obj.type == 'MESH': log.warning(f'Object {obj.name} is not a mesh, has no vertices.') return if len(obj.data.sculpt_vertex_colors): for ...
['def', 'populate_vertex_colors(obj:', 'Union[bpy.types.Object,', 'str],', 'color_rgba:', 'Tuple[float],', 'seg_type:', "str='instance')", '->', 'None:', 'obj', '=', 'verify(obj)', 'if', 'not', 'obj.type', '==', "'MESH':", "log.warning(f'Object", '{obj.name}', 'is', 'not', 'a', 'mesh,', 'has', 'no', "vertices.')", 'ret...
972,081
ZumoLabs/zpy
objects.py
random_position_within_constraints
random_position_within_constraints
Randomize position of object within constraints.
[ "Randomize", "position", "of", "object", "within", "constraints." ]
def random_position_within_constraints(obj: Union[bpy.types.Object, str]) -> None: obj = verify(obj) _constraints = obj.constraints.get('Limit Location', None) if _constraints is not None: obj.location.x = random.uniform(obj.constraints['Limit Location'].min_x, obj.constraints['Limit Location'].max_...
['def', 'random_position_within_constraints(obj:', 'Union[bpy.types.Object,', 'str])', '->', 'None:', 'obj', '=', 'verify(obj)', '_constraints', '=', "obj.constraints.get('Limit", "Location',", 'None)', 'if', '_constraints', 'is', 'not', 'None:', 'obj.location.x', '=', "random.uniform(obj.constraints['Limit", "Location...
972,082
ZumoLabs/zpy
objects.py
jitter
jitter
Apply random scale (blender units) and rotation (radians) to object.
[ "Apply", "random", "scale", "(blender", "units)", "and", "rotation", "(radians)", "to", "object." ]
def jitter(obj: Union[bpy.types.Object, str], translate_range: Tuple[Tuple[float]]=((0, 0), (0, 0), (0, 0)), rotate_range: Tuple[Tuple[float]]=((0, 0), (0, 0), (0, 0)), scale_range: Tuple[Tuple[float]]=((1.0, 1.0), (1.0, 1.0), (1.0, 1.0))) -> None: obj = verify(obj) translate(obj, translation=(random.uniform(tr...
['def', 'jitter(obj:', 'Union[bpy.types.Object,', 'str],', 'translate_range:', 'Tuple[Tuple[float]]=((0,', '0),', '(0,', '0),', '(0,', '0)),', 'rotate_range:', 'Tuple[Tuple[float]]=((0,', '0),', '(0,', '0),', '(0,', '0)),', 'scale_range:', 'Tuple[Tuple[float]]=((1.0,', '1.0),', '(1.0,', '1.0),', '(1.0,', '1.0)))', '->'...
972,087
ZumoLabs/zpy
objects.py
save_pose
save_pose
Save a pose (rot and pos) to dict.
[ "Save", "a", "pose", "(rot", "and", "pos)", "to", "dict." ]
def save_pose(obj: Union[bpy.types.Object, str], pose_name: str=None) -> None: obj = verify(obj) log.info(f'Saving pose {pose_name} based on object {obj.name}') if pose_name is None: pose_name = obj.name _SAVED_POSES[pose_name] = obj.matrix_world.copy()
['def', 'save_pose(obj:', 'Union[bpy.types.Object,', 'str],', 'pose_name:', 'str=None)', '->', 'None:', 'obj', '=', 'verify(obj)', "log.info(f'Saving", 'pose', '{pose_name}', 'based', 'on', 'object', "{obj.name}')", 'if', 'pose_name', 'is', 'None:', 'pose_name', '=', 'obj.name', '_SAVED_POSES[pose_name]', '=', 'obj.mat...
972,088
ZumoLabs/zpy
output_coco.py
parse_coco_annotations
parse_coco_annotations
Parse COCO annotations, optionally output a ImageSaver object.
[ "Parse", "COCO", "annotations,", "optionally", "output", "a", "ImageSaver", "object." ]
def parse_coco_annotations(annotation_file: Union[Path, str], data_dir: Union[Path, str]=None, output_saver: bool=False, image_keys_to_add: List[str]=None) -> zpy.saver_image.ImageSaver: log.info(f'Parsing COCO annotations at {annotation_file}...') annotation_file = zpy.files.verify_path(annotation_file) if...
['def', 'parse_coco_annotations(annotation_file:', 'Union[Path,', 'str],', 'data_dir:', 'Union[Path,', 'str]=None,', 'output_saver:', 'bool=False,', 'image_keys_to_add:', 'List[str]=None)', '->', 'zpy.saver_image.ImageSaver:', "log.info(f'Parsing", 'COCO', 'annotations', 'at', "{annotation_file}...')", 'annotation_file...
972,091
ZumoLabs/zpy
output_coco.py
OutputCOCO.output_annotations
output_annotations
Output COCO annotations to file.
[ "Output", "COCO", "annotations", "to", "file." ]
def output_annotations(self, annotation_path: Union[Path, str]=None, splitseg: bool=False) -> Path: annotation_path = super().output_annotations(annotation_path=annotation_path) coco_dict = {'info': self.coco_info(), 'licenses': self.coco_license(), 'categories': self.coco_categories(), 'images': self.coco_imag...
['def', 'output_annotations(self,', 'annotation_path:', 'Union[Path,', 'str]=None,', 'splitseg:', 'bool=False)', '->', 'Path:', 'annotation_path', '=', 'super().output_annotations(annotation_path=annotation_path)', 'coco_dict', '=', "{'info':", 'self.coco_info(),', "'licenses':", 'self.coco_license(),', "'categories':"...
972,092
ZumoLabs/zpy
output_csv.py
OutputCSV.output_annotations
output_annotations
Output CSV annotations to file.
[ "Output", "CSV", "annotations", "to", "file." ]
def output_annotations(self, annotation_path: Union[Path, str]=None, annotation_dict_to_csv_row_func: Callable=None, header: List[str]=None) -> Path: annotation_path = super().output_annotations(annotation_path=annotation_path) if annotation_dict_to_csv_row_func is None: raise CSVParseError('Output CSV ...
['def', 'output_annotations(self,', 'annotation_path:', 'Union[Path,', 'str]=None,', 'annotation_dict_to_csv_row_func:', 'Callable=None,', 'header:', 'List[str]=None)', '->', 'Path:', 'annotation_path', '=', 'super().output_annotations(annotation_path=annotation_path)', 'if', 'annotation_dict_to_csv_row_func', 'is', 'N...
972,094
ZumoLabs/zpy
output_zumo.py
OutputZUMO.output_annotations
output_annotations
Output annotations to file.
[ "Output", "annotations", "to", "file." ]
def output_annotations(self, annotation_path: Union[Path, str]=None) -> Path: annotation_path = super().output_annotations(annotation_path=annotation_path) zumo_dict = {'metadata': self.saver.metadata, 'categories': self.saver.categories, 'images': self.saver.images, 'annotations': self.saver.annotations} z...
['def', 'output_annotations(self,', 'annotation_path:', 'Union[Path,', 'str]=None)', '->', 'Path:', 'annotation_path', '=', 'super().output_annotations(annotation_path=annotation_path)', 'zumo_dict', '=', "{'metadata':", 'self.saver.metadata,', "'categories':", 'self.saver.categories,', "'images':", 'self.saver.images,...
972,097
ZumoLabs/zpy
render.py
make_aov_pass
make_aov_pass
Make AOV pass in Cycles.
[ "Make", "AOV", "pass", "in", "Cycles." ]
def make_aov_pass(style: str='instance') -> None: scene = zpy.blender.verify_blender_scene() if not scene.render.engine == 'CYCLES': log.warning(' Setting render engine to CYCLES to use AOV') scene.render.engine = 'CYCLES' scene.render.use_compositing = True valid_styles = ['instance...
['def', 'make_aov_pass(style:', "str='instance')", '->', 'None:', 'scene', '=', 'zpy.blender.verify_blender_scene()', 'if', 'not', 'scene.render.engine', '==', "'CYCLES':", "log.warning('", 'Setting', 'render', 'engine', 'to', 'CYCLES', 'to', 'use', "AOV')", 'scene.render.engine', '=', "'CYCLES'", 'scene.render.use_com...
972,098
ZumoLabs/zpy
render.py
lens_dirt_node
lens_dirt_node
TODO: Add lens dirt effect to a compositor node.
[ "TODO:", "Add", "lens", "dirt", "effect", "to", "a", "compositor", "node." ]
def lens_dirt_node(node_tree: bpy.types.NodeTree, input_node: bpy.types.Node) -> bpy.types.Node: log.warn('NotImplemented: lens dirt ') return input_node
['def', 'lens_dirt_node(node_tree:', 'bpy.types.NodeTree,', 'input_node:', 'bpy.types.Node)', '->', 'bpy.types.Node:', "log.warn('NotImplemented:", 'lens', 'dirt', "')", 'return', 'input_node']
972,101
ZumoLabs/zpy
render.py
render
render
Render images using AOV nodes.
[ "Render", "images", "using", "AOV", "nodes." ]
def render(rgb_path: Union[Path, str]=None, depth_path: Union[Path, str]=None, iseg_path: Union[Path, str]=None, cseg_path: Union[Path, str]=None, width: int=640, height: int=480, hsv: Tuple[float]=None): scene = zpy.blender.verify_blender_scene() scene.render.resolution_x = width scene.render.resolution_y ...
['def', 'render(rgb_path:', 'Union[Path,', 'str]=None,', 'depth_path:', 'Union[Path,', 'str]=None,', 'iseg_path:', 'Union[Path,', 'str]=None,', 'cseg_path:', 'Union[Path,', 'str]=None,', 'width:', 'int=640,', 'height:', 'int=480,', 'hsv:', 'Tuple[float]=None):', 'scene', '=', 'zpy.blender.verify_blender_scene()', 'scen...
972,102
ZumoLabs/zpy
render.py
default_render_settings
default_render_settings
Render settings for normal color images.
[ "Render", "settings", "for", "normal", "color", "images." ]
def default_render_settings(samples: int=96, tile_size: int=48, spatial_splits: bool=False, is_aggressive: bool=False) -> None: scene = zpy.blender.verify_blender_scene() if not scene.render.engine == 'CYCLES': log.warning(' Setting render engine to CYCLES') scene.render.engine = 'CYCLES' sc...
['def', 'default_render_settings(samples:', 'int=96,', 'tile_size:', 'int=48,', 'spatial_splits:', 'bool=False,', 'is_aggressive:', 'bool=False)', '->', 'None:', 'scene', '=', 'zpy.blender.verify_blender_scene()', 'if', 'not', 'scene.render.engine', '==', "'CYCLES':", "log.warning('", 'Setting', 'render', 'engine', 'to...
972,103
ZumoLabs/zpy
render.py
segmentation_render_settings
segmentation_render_settings
Render settings for segmentation images.
[ "Render", "settings", "for", "segmentation", "images." ]
def segmentation_render_settings(): scene = zpy.blender.verify_blender_scene() if not scene.render.engine == 'CYCLES': log.warning(' Setting render engine to CYCLES') scene.render.engine = 'CYCLES' scene.render.film_transparent = True scene.render.dither_intensity = 0.0 scene.render....
['def', 'segmentation_render_settings():', 'scene', '=', 'zpy.blender.verify_blender_scene()', 'if', 'not', 'scene.render.engine', '==', "'CYCLES':", "log.warning('", 'Setting', 'render', 'engine', 'to', "CYCLES')", 'scene.render.engine', '=', "'CYCLES'", 'scene.render.film_transparent', '=', 'True', 'scene.render.dith...
972,104
ZumoLabs/zpy
requests.py
verify_key
verify_key
Check a request dict for key, raise error if not present or wrong type.
[ "Check", "a", "request", "dict", "for", "key,", "raise", "error", "if", "not", "present", "or", "wrong", "type." ]
def verify_key(request: Dict, key: str, key_type: type=None) -> Any: value = request.get(key, None) if value is None: raise InvalidRequest(f'Required key {key} not found.') if key_type is not None: if not isinstance(value, key_type): raise InvalidRequest(f'Key {key} must be of ty...
['def', 'verify_key(request:', 'Dict,', 'key:', 'str,', 'key_type:', 'type=None)', '->', 'Any:', 'value', '=', 'request.get(key,', 'None)', 'if', 'value', 'is', 'None:', 'raise', "InvalidRequest(f'Required", 'key', '{key}', 'not', "found.')", 'if', 'key_type', 'is', 'not', 'None:', 'if', 'not', 'isinstance(value,', 'ke...
972,105
ZumoLabs/zpy
requests.py
request_as_process
request_as_process
Decorator for running a request as seperate processes.
[ "Decorator", "for", "running", "a", "request", "as", "seperate", "processes." ]
def request_as_process(request_func): @wraps(request_func) def wrapped_request_func(request: Dict) -> None: _reply = multiprocessing.Manager().dict() p = Process(target=request_func, args=(request, _reply)) p.start() p.join() global reply reply.update(_reply) ...
['def', 'request_as_process(request_func):', '@wraps(request_func)', 'def', 'wrapped_request_func(request:', 'Dict)', '->', 'None:', '_reply', '=', 'multiprocessing.Manager().dict()', 'p', '=', 'Process(target=request_func,', 'args=(request,', '_reply))', 'p.start()', 'p.join()', 'global', 'reply', 'reply.update(_reply...
972,106
ZumoLabs/zpy
requests.py
send_request
send_request
Send a request over a uri.
[ "Send", "a", "request", "over", "a", "uri." ]
def send_request(request: Dict, ip: str='127.0.0.1', port: str='5555') -> Dict: log.info(f'Connecting to {ip}:{port} ...') context = zmq.Context() socket = context.socket(zmq.REQ) socket.connect(f'tcp://{ip}:{port}') log.info('... Done!') log.info(f'Sending request: {request}') socket.send_j...
['def', 'send_request(request:', 'Dict,', 'ip:', "str='127.0.0.1',", 'port:', "str='5555')", '->', 'Dict:', "log.info(f'Connecting", 'to', '{ip}:{port}', "...')", 'context', '=', 'zmq.Context()', 'socket', '=', 'context.socket(zmq.REQ)', "socket.connect(f'tcp://{ip}:{port}')", "log.info('...", "Done!')", "log.info(f'Se...
972,108
ZumoLabs/zpy
saver.py
Saver.clip_bbox
clip_bbox
Clip a bounding box in [x, y, width, height] format.
[ "Clip", "a", "bounding", "box", "in", "[x,", "y,", "width,", "height]", "format." ]
def clip_bbox(bbox: List[Union[int, float]]=None, height: Union[int, float]=None, width: Union[int, float]=None, normalized: bool=False) -> List[Union[int, float]]: if normalized: (max_x, max_y) = (1.0, 1.0) else: (max_x, max_y) = (width, height) new_bbox = [0] * 4 new_bbox[0] = max(0, m...
['def', 'clip_bbox(bbox:', 'List[Union[int,', 'float]]=None,', 'height:', 'Union[int,', 'float]=None,', 'width:', 'Union[int,', 'float]=None,', 'normalized:', 'bool=False)', '->', 'List[Union[int,', 'float]]:', 'if', 'normalized:', '(max_x,', 'max_y)', '=', '(1.0,', '1.0)', 'else:', '(max_x,', 'max_y)', '=', '(width,',...
972,115
ZumoLabs/zpy
saver_image.py
ImageSaver.parse_annotations_from_seg_image
parse_annotations_from_seg_image
Populate annotation field based on segmentation image.
[ "Populate", "annotation", "field", "based", "on", "segmentation", "image." ]
def parse_annotations_from_seg_image(self, image_name: str) -> None: is_iseg = zpy.files.file_is_of_type(image_name, 'instance segmentation image') is_cseg = zpy.files.file_is_of_type(image_name, 'class segmentation image') if not (is_iseg or is_cseg): raise ValueError('Image is not segmentation ima...
['def', 'parse_annotations_from_seg_image(self,', 'image_name:', 'str)', '->', 'None:', 'is_iseg', '=', 'zpy.files.file_is_of_type(image_name,', "'instance", 'segmentation', "image')", 'is_cseg', '=', 'zpy.files.file_is_of_type(image_name,', "'class", 'segmentation', "image')", 'if', 'not', '(is_iseg', 'or', 'is_cseg):...
972,118
ZumoLabs/zpy
saver_image.py
ImageSaver.output_annotated_images
output_annotated_images
Dump annotated sampled images to the meta folder.
[ "Dump", "annotated", "sampled", "images", "to", "the", "meta", "folder." ]
def output_annotated_images(self, num_annotated_images: int=10) -> None: log.info('Output annotated images...') import zpy.viz output_path = self.output_dir / self.HIDDEN_METAFOLDER_FILENAME output_path = zpy.files.verify_path(output_path, make=True, check_dir=True) for (i, image) in enumerate(self....
['def', 'output_annotated_images(self,', 'num_annotated_images:', 'int=10)', '->', 'None:', "log.info('Output", 'annotated', "images...')", 'import', 'zpy.viz', 'output_path', '=', 'self.output_dir', '/', 'self.HIDDEN_METAFOLDER_FILENAME', 'output_path', '=', 'zpy.files.verify_path(output_path,', 'make=True,', 'check_d...
972,119
ZumoLabs/zpy
saver_image.py
ImageSaver.output_meta_analysis
output_meta_analysis
Perform a full meta analysis, outputting some meta files.
[ "Perform", "a", "full", "meta", "analysis,", "outputting", "some", "meta", "files." ]
def output_meta_analysis(self, image_sample_size: int=50) -> None: log.info(f'perform meta analysis image_sample_size:{image_sample_size}...') import zpy.files image_paths = [i['output_path'] for i in self.images.values() if i['style'] == 'default'] image_paths = zpy.files.sample(image_paths, sample_siz...
['def', 'output_meta_analysis(self,', 'image_sample_size:', 'int=50)', '->', 'None:', "log.info(f'perform", 'meta', 'analysis', "image_sample_size:{image_sample_size}...')", 'import', 'zpy.files', 'image_paths', '=', "[i['output_path']", 'for', 'i', 'in', 'self.images.values()', 'if', "i['style']", '==', "'default']", ...
972,120
ZumoLabs/zpy
viz.py
pretty_axes
pretty_axes
Better looking matplotlib axes object.
[ "Better", "looking", "matplotlib", "axes", "object." ]
def pretty_axes(ax: matplotlib.axes.Axes) -> matplotlib.axes.Axes: ax.spines['top'].set_visible(False) ax.spines['right'].set_visible(False) ax.spines['left'].set_visible(False) ax.get_xaxis().tick_bottom() ax.get_yaxis().set_visible(False) ax.grid(axis='y', alpha=0.75) return ax
['def', 'pretty_axes(ax:', 'matplotlib.axes.Axes)', '->', 'matplotlib.axes.Axes:', "ax.spines['top'].set_visible(False)", "ax.spines['right'].set_visible(False)", "ax.spines['left'].set_visible(False)", 'ax.get_xaxis().tick_bottom()', 'ax.get_yaxis().set_visible(False)', "ax.grid(axis='y',", 'alpha=0.75)', 'return', 'a...
972,124
ZumoLabs/zpy
viz.py
image_grid_plot
image_grid_plot
Plots images in a grid.
[ "Plots", "images", "in", "a", "grid." ]
def image_grid_plot(images: List[np.ndarray]=None, rows: int=4, cols: int=4) -> Tuple[str, matplotlib.figure.Figure]: assert images is not None, 'Images required.' sample_size = min(rows * cols, len(images)) images = random.sample(images, sample_size) fig = plt.figure(figsize=(16, 16)) plt.suptitle(...
['def', 'image_grid_plot(images:', 'List[np.ndarray]=None,', 'rows:', 'int=4,', 'cols:', 'int=4)', '->', 'Tuple[str,', 'matplotlib.figure.Figure]:', 'assert', 'images', 'is', 'not', 'None,', "'Images", "required.'", 'sample_size', '=', 'min(rows', '*', 'cols,', 'len(images))', 'images', '=', 'random.sample(images,', 's...
972,126
ZumoLabs/zpy
viz.py
color_correlations_plot
color_correlations_plot
Plots 2D histograms of color correlations: RG, RB, and BG.
[ "Plots", "2D", "histograms", "of", "color", "correlations:", "RG,", "RB,", "and", "BG." ]
def color_correlations_plot(flat_images: List[np.ndarray]=None) -> Tuple[str, matplotlib.figure.Figure]: assert flat_images is not None, 'Images required.' flat_images = flat_images[0] fig = plt.figure(figsize=(16, 5)) plt.rcParams['axes.grid'] = False plt.suptitle('Pixel Color Correlations \n\n\n',...
['def', 'color_correlations_plot(flat_images:', 'List[np.ndarray]=None)', '->', 'Tuple[str,', 'matplotlib.figure.Figure]:', 'assert', 'flat_images', 'is', 'not', 'None,', "'Images", "required.'", 'flat_images', '=', 'flat_images[0]', 'fig', '=', 'plt.figure(figsize=(16,', '5))', "plt.rcParams['axes.grid']", '=', 'False...
972,128
ZumoLabs/zpy
viz.py
draw_annotations
draw_annotations
Given an path to an image draw annotations.
[ "Given", "an", "path", "to", "an", "image", "draw", "annotations." ]
def draw_annotations(image_path: Union[Path, str]=None, annotations: List=None, categories: Dict[str, Dict]=None) -> None: log.info(f'draw annotations on {image_path}...') image = zpy.image.open_image(image_path) (_, ax) = plt.subplots() ax.imshow(image) for (i, annotation) in enumerate(annotations)...
['def', 'draw_annotations(image_path:', 'Union[Path,', 'str]=None,', 'annotations:', 'List=None,', 'categories:', 'Dict[str,', 'Dict]=None)', '->', 'None:', "log.info(f'draw", 'annotations', 'on', "{image_path}...')", 'image', '=', 'zpy.image.open_image(image_path)', '(_,', 'ax)', '=', 'plt.subplots()', 'ax.imshow(imag...
972,131
ZumoLabs/zpy
viz.py
draw_bbox
draw_bbox
Draw a bounding box on the matplotlib axes object.
[ "Draw", "a", "bounding", "box", "on", "the", "matplotlib", "axes", "object." ]
def draw_bbox(ax: matplotlib.axes.Axes, bbox: List, color: Tuple[int], text: str=None, alpha: float=0.2) -> None: log.debug(f'Drawing bbox {bbox} {color}') r = Rectangle((bbox[0], bbox[1]), bbox[2], bbox[3], linewidth=3, facecolor=color, edgecolor=color, alpha=alpha) if text is not None: ax.text(x=b...
['def', 'draw_bbox(ax:', 'matplotlib.axes.Axes,', 'bbox:', 'List,', 'color:', 'Tuple[int],', 'text:', 'str=None,', 'alpha:', 'float=0.2)', '->', 'None:', "log.debug(f'Drawing", 'bbox', '{bbox}', "{color}')", 'r', '=', 'Rectangle((bbox[0],', 'bbox[1]),', 'bbox[2],', 'bbox[3],', 'linewidth=3,', 'facecolor=color,', 'edgec...
972,132
ZumoLabs/zpy
output_panel.py
registerSceneProperties
registerSceneProperties
Properties applied to scenes.
[ "Properties", "applied", "to", "scenes." ]
def registerSceneProperties(): bpy.types.Scene.zpy_output_path = bpy.props.StringProperty(name='Output Path', description='Output path for rendered images, annotations, etc.', default=str(zpy.files.default_temp_path()), subtype='DIR_PATH')
['def', 'registerSceneProperties():', 'bpy.types.Scene.zpy_output_path', '=', "bpy.props.StringProperty(name='Output", "Path',", "description='Output", 'path', 'for', 'rendered', 'images,', 'annotations,', "etc.',", 'default=str(zpy.files.default_temp_path()),', "subtype='DIR_PATH')"]
972,153
ZumoLabs/zpy
segment_panel.py
registerObjectProperties
registerObjectProperties
Properties applied to object.
[ "Properties", "applied", "to", "object." ]
def registerObjectProperties(): bpy.types.Object.seg = bpy.props.PointerProperty(type=SegmentableProperties)
['def', 'registerObjectProperties():', 'bpy.types.Object.seg', '=', 'bpy.props.PointerProperty(type=SegmentableProperties)']
972,155
ZumoLabs/zpy
__init__.py
install_pip_depenencies
install_pip_depenencies
Install pip dependencies required by zpy addon.
[ "Install", "pip", "dependencies", "required", "by", "zpy", "addon." ]
def install_pip_depenencies(): try: log.info('Installing zpy and dependencies...') pip_install = [sys.executable, '-m', 'pip', 'install'] subprocess.run(pip_install + ['--upgrade', 'pip'], check=True) pkg_path = Path(sys.executable).parent.parent / 'lib' / 'site-packages' / 'zpy' ...
['def', 'install_pip_depenencies():', 'try:', "log.info('Installing", 'zpy', 'and', "dependencies...')", 'pip_install', '=', '[sys.executable,', "'-m',", "'pip',", "'install']", 'subprocess.run(pip_install', '+', "['--upgrade',", "'pip'],", 'check=True)', 'pkg_path', '=', 'Path(sys.executable).parent.parent', '/', "'li...
972,157
MendelXu/zsseg.baseline
classification_evaluation.py
accuracy
accuracy
Computes the accuracy over the k top predictions for the specified values of k In top-5 accuracy you give yourself credit for having the right answer if the right answer appears in your top five guesses.
[ "Computes", "the", "accuracy", "over", "the", "k", "top", "predictions", "for", "the", "specified", "values", "of", "k", "In", "top-5", "accuracy", "you", "give", "yourself", "credit", "for", "having", "the", "right", "answer", "if", "the", "right", "answer"...
def accuracy(output: torch.Tensor, target: torch.Tensor, topk=(1,)): maxk = max(topk) (_, pred) = output.topk(maxk, 1, True, True) pred = pred.t() correct = (pred == target.unsqueeze(dim=0)).expand_as(pred) res = [] for k in topk: correct_k = correct[:k].float().sum(0) res.append...
['def', 'accuracy(output:', 'torch.Tensor,', 'target:', 'torch.Tensor,', 'topk=(1,)):', 'maxk', '=', 'max(topk)', '(_,', 'pred)', '=', 'output.topk(maxk,', '1,', 'True,', 'True)', 'pred', '=', 'pred.t()', 'correct', '=', '(pred', '==', 'target.unsqueeze(dim=0)).expand_as(pred)', 'res', '=', '[]', 'for', 'k', 'in', 'top...
972,208
albanie/zsvision
zs_data_structures.py
ExpertStore.todict
todict
Convert the current datastructure into a vanilla python dictionary Returns: a dictionary with the same keys and values as the current object.
[ "Convert", "the", "current", "datastructure", "into", "a", "vanilla", "python", "dictionary", "Returns:", "a", "dictionary", "with", "the", "same", "keys", "and", "values", "as", "the", "current", "object." ]
def todict(self): return {key: self[key] for key in self.keymap}
['def', 'todict(self):', 'return', '{key:', 'self[key]', 'for', 'key', 'in', 'self.keymap}']
972,232
albanie/zsvision
zs_multiproc.py
apply_kwargs
apply_kwargs
Wrapper for unpacking keyword function calls.
[ "Wrapper", "for", "unpacking", "keyword", "function", "calls." ]
def apply_kwargs(func, kwargs): return func(**kwargs)
['def', 'apply_kwargs(func,', 'kwargs):', 'return', 'func(**kwargs)']
972,239
albanie/zsvision
zs_utils.py
pickle_loader
pickle_loader
Deserialise object from pickle.
[ "Deserialise", "object", "from", "pickle." ]
def pickle_loader(pkl_path: Path, verbose: bool, backwards_compatible: bool=True) -> object: tic = time.time() with open(pkl_path, 'rb') as f: buffer = f.read() if verbose: print(f'[I/O: {time.time() - tic:.1f}s]', end=' ') tic = time.time() if backwards_compatible: ...
['def', 'pickle_loader(pkl_path:', 'Path,', 'verbose:', 'bool,', 'backwards_compatible:', 'bool=True)', '->', 'object:', 'tic', '=', 'time.time()', 'with', 'open(pkl_path,', "'rb')", 'as', 'f:', 'buffer', '=', 'f.read()', 'if', 'verbose:', "print(f'[I/O:", '{time.time()', '-', "tic:.1f}s]',", "end='", "')", 'tic', '=',...
972,241
albanie/zsvision
zs_utils.py
msgpack_loader
msgpack_loader
Msgpack provides a faster serialisation routine than pickle, so is preferable for loading and deserialising large feature sets from disk.
[ "Msgpack", "provides", "a", "faster", "serialisation", "routine", "than", "pickle,", "so", "is", "preferable", "for", "loading", "and", "deserialising", "large", "feature", "sets", "from", "disk." ]
def msgpack_loader(mp_path: Path, verbose: bool): tic = time.time() with open(mp_path, 'rb') as f: buffer = f.read() if verbose: print(f'[I/O: {time.time() - tic:.1f}s]', end=' ') tic = time.time() data = msgpack_np.unpackb(buffer, raw=False) if verbose: ...
['def', 'msgpack_loader(mp_path:', 'Path,', 'verbose:', 'bool):', 'tic', '=', 'time.time()', 'with', 'open(mp_path,', "'rb')", 'as', 'f:', 'buffer', '=', 'f.read()', 'if', 'verbose:', "print(f'[I/O:", '{time.time()', '-', "tic:.1f}s]',", "end='", "')", 'tic', '=', 'time.time()', 'data', '=', 'msgpack_np.unpackb(buffer,...
972,242
albanie/zsvision
zs_utils.py
load_json_or_yaml_config
load_json_or_yaml_config
Load a configuration file into memory.
[ "Load", "a", "configuration", "file", "into", "memory." ]
def load_json_or_yaml_config(cfg_fname: (Path, str)) -> dict: ancestors = find_ancestors(cfg_fname) config = ancestors.pop() ancestors = reversed(ancestors) for ancestor in ancestors: merge(ancestor, config, strategy=Strategy.REPLACE) config = ancestor return config
['def', 'load_json_or_yaml_config(cfg_fname:', '(Path,', 'str))', '->', 'dict:', 'ancestors', '=', 'find_ancestors(cfg_fname)', 'config', '=', 'ancestors.pop()', 'ancestors', '=', 'reversed(ancestors)', 'for', 'ancestor', 'in', 'ancestors:', 'merge(ancestor,', 'config,', 'strategy=Strategy.REPLACE)', 'config', '=', 'an...
972,246
albanie/zsvision
zs_utils.py
load_json_config
load_json_config
Load a json configuration file into memory.
[ "Load", "a", "json", "configuration", "file", "into", "memory." ]
def load_json_config(cfg_fname: (Path, str)) -> dict: return load_json_or_yaml_config(cfg_fname)
['def', 'load_json_config(cfg_fname:', '(Path,', 'str))', '->', 'dict:', 'return', 'load_json_or_yaml_config(cfg_fname)']
972,247
albanie/zsvision
zs_utils.py
load_yaml_config
load_yaml_config
Load a yaml configuration file into memory.
[ "Load", "a", "yaml", "configuration", "file", "into", "memory." ]
def load_yaml_config(cfg_fname: (Path, str)) -> dict: return load_json_or_yaml_config(cfg_fname)
['def', 'load_yaml_config(cfg_fname:', '(Path,', 'str))', '->', 'dict:', 'return', 'load_json_or_yaml_config(cfg_fname)']
972,248
albanie/zsvision
zs_utils.py
quote_and_escape_ffmpeg_path
quote_and_escape_ffmpeg_path
Quote and escape paths for use with ffmpeg/ffprobe.
[ "Quote", "and", "escape", "paths", "for", "use", "with", "ffmpeg/ffprobe." ]
def quote_and_escape_ffmpeg_path(path: (str, Path)) -> str: escaped = str(path).replace('$', '\\$').replace('%', '\\%') if "'" in escaped: quoted = f'"{escaped}"' else: quoted = f"'{escaped}'" return quoted
['def', 'quote_and_escape_ffmpeg_path(path:', '(str,', 'Path))', '->', 'str:', 'escaped', '=', "str(path).replace('$',", "'\\\\$').replace('%',", "'\\\\%')", 'if', '"\'"', 'in', 'escaped:', 'quoted', '=', 'f\'"{escaped}"\'', 'else:', 'quoted', '=', 'f"\'{escaped}\'"', 'return', 'quoted']
972,251
mazzzystar/WaveGAN-pytorch
utils.py
numpy_to_var
numpy_to_var
Convert numpy array to Variable.
[ "Convert", "numpy", "array", "to", "Variable." ]
def numpy_to_var(numpy_data, cuda): data = numpy_data[:, np.newaxis, :] data = torch.Tensor(data) if cuda: data = data.cuda() return Variable(data, requires_grad=False)
['def', 'numpy_to_var(numpy_data,', 'cuda):', 'data', '=', 'numpy_data[:,', 'np.newaxis,', ':]', 'data', '=', 'torch.Tensor(data)', 'if', 'cuda:', 'data', '=', 'data.cuda()', 'return', 'Variable(data,', 'requires_grad=False)']
972,260
NoaCahan/WavenetAutoEncoder
generate.py
decode1
decode1
Synthesize audio from an array of embeddings.
[ "Synthesize", "audio", "from", "an", "array", "of", "embeddings." ]
def decode1(model_path, model_name, encoding, decoder_path, decoder_name, sr=16000, duration=10): if os.path.exists(decoder_path) is False: os.makedirs(decoder_path) with open('./params/model_params.json') as f: model_params = json.load(f) f.close() net = WavenetAutoencoder(**model_param...
['def', 'decode1(model_path,', 'model_name,', 'encoding,', 'decoder_path,', 'decoder_name,', 'sr=16000,', 'duration=10):', 'if', 'os.path.exists(decoder_path)', 'is', 'False:', 'os.makedirs(decoder_path)', 'with', "open('./params/model_params.json')", 'as', 'f:', 'model_params', '=', 'json.load(f)', 'f.close()', 'net',...
972,267
sek788432/Waymo-2D-Object-Detection
base_trainer.py
Recovery.maybe_recover
maybe_recover
Conditionally recovers the training by triggering checkpoint restoration.
[ "Conditionally", "recovers", "the", "training", "by", "triggering", "checkpoint", "restoration." ]
def maybe_recover(self, loss_value, global_step): if not self.should_recover(loss_value, global_step): return self.recover_counter += 1 if self.recover_counter > self.recovery_max_trials: raise RuntimeError('The loss value is NaN after training loop and it happens %d times.' % self.recover_c...
['def', 'maybe_recover(self,', 'loss_value,', 'global_step):', 'if', 'not', 'self.should_recover(loss_value,', 'global_step):', 'return', 'self.recover_counter', '+=', '1', 'if', 'self.recover_counter', '>', 'self.recovery_max_trials:', 'raise', "RuntimeError('The", 'loss', 'value', 'is', 'NaN', 'after', 'training', 'l...
972,302
sek788432/Waymo-2D-Object-Detection
base_trainer.py
_AsyncTrainer.init_async
init_async
Initializes the Async Trainer base class.
[ "Initializes", "the", "Async", "Trainer", "base", "class." ]
def init_async(self): assert isinstance(self._strategy, tf.distribute.Strategy) self._is_async = isinstance(self._strategy, tf.distribute.experimental.ParameterServerStrategy) self._coordinator = None if self._is_async: self._coordinator = tf.distribute.experimental.coordinator.ClusterCoordinato...
['def', 'init_async(self):', 'assert', 'isinstance(self._strategy,', 'tf.distribute.Strategy)', 'self._is_async', '=', 'isinstance(self._strategy,', 'tf.distribute.experimental.ParameterServerStrategy)', 'self._coordinator', '=', 'None', 'if', 'self._is_async:', 'self._coordinator', '=', 'tf.distribute.experimental.coo...
972,303
sek788432/Waymo-2D-Object-Detection
base_trainer.py
_AsyncTrainer.create_train_loop_fn
create_train_loop_fn
Creates a eval loop from the given step function and options.
[ "Creates", "a", "eval", "loop", "from", "the", "given", "step", "function", "and", "options." ]
def create_train_loop_fn(self): train_loop_fn = super().create_train_loop_fn() if getattr(self, '_is_async', False): def _async_loop_fn(iterator, num_steps): self._coordinator.schedule(train_loop_fn, args=(iterator, num_steps)) return _async_loop_fn else: return train_lo...
['def', 'create_train_loop_fn(self):', 'train_loop_fn', '=', 'super().create_train_loop_fn()', 'if', 'getattr(self,', "'_is_async',", 'False):', 'def', '_async_loop_fn(iterator,', 'num_steps):', 'self._coordinator.schedule(train_loop_fn,', 'args=(iterator,', 'num_steps))', 'return', '_async_loop_fn', 'else:', 'return',...
972,305
sek788432/Waymo-2D-Object-Detection
base_trainer_test.py
create_in_process_cluster
create_in_process_cluster
Creates and starts local servers and returns the cluster_resolver.
[ "Creates", "and", "starts", "local", "servers", "and", "returns", "the", "cluster_resolver." ]
def create_in_process_cluster(num_workers, num_ps): worker_ports = [portpicker.pick_unused_port() for _ in range(num_workers)] ps_ports = [portpicker.pick_unused_port() for _ in range(num_ps)] cluster_dict = {} cluster_dict['worker'] = ['localhost:%s' % port for port in worker_ports] if num_ps > 0: ...
['def', 'create_in_process_cluster(num_workers,', 'num_ps):', 'worker_ports', '=', '[portpicker.pick_unused_port()', 'for', '_', 'in', 'range(num_workers)]', 'ps_ports', '=', '[portpicker.pick_unused_port()', 'for', '_', 'in', 'range(num_ps)]', 'cluster_dict', '=', '{}', "cluster_dict['worker']", '=', "['localhost:%s'"...
972,314
sek788432/Waymo-2D-Object-Detection
export_base.py
export
export
Exports to SavedModel format.
[ "Exports", "to", "SavedModel", "format." ]
def export(export_module: ExportModule, function_keys: Union[List[Text], Dict[Text, Text]], export_savedmodel_dir: Text, checkpoint_path: Optional[Text]=None, timestamped: bool=True, save_options: Optional[tf.saved_model.SaveOptions]=None) -> Text: ckpt_dir_or_file = checkpoint_path if tf.io.gfile.isdir(ckpt_di...
['def', 'export(export_module:', 'ExportModule,', 'function_keys:', 'Union[List[Text],', 'Dict[Text,', 'Text]],', 'export_savedmodel_dir:', 'Text,', 'checkpoint_path:', 'Optional[Text]=None,', 'timestamped:', 'bool=True,', 'save_options:', 'Optional[tf.saved_model.SaveOptions]=None)', '->', 'Text:', 'ckpt_dir_or_file',...
972,315
sek788432/Waymo-2D-Object-Detection
train_utils.py
cast_leaf_nested_dict
cast_leaf_nested_dict
Cast the leaves of a dictionary with arbitrary depth in place.
[ "Cast", "the", "leaves", "of", "a", "dictionary", "with", "arbitrary", "depth", "in", "place." ]
def cast_leaf_nested_dict(d: Dict[str, Any], cast_fn: Callable[[Any], Any]) -> Dict[str, Any]: for (key, value) in d.items(): if isinstance(value, dict): d[key] = cast_leaf_nested_dict(value, cast_fn) else: d[key] = cast_fn(value) return d
['def', 'cast_leaf_nested_dict(d:', 'Dict[str,', 'Any],', 'cast_fn:', 'Callable[[Any],', 'Any])', '->', 'Dict[str,', 'Any]:', 'for', '(key,', 'value)', 'in', 'd.items():', 'if', 'isinstance(value,', 'dict):', 'd[key]', '=', 'cast_leaf_nested_dict(value,', 'cast_fn)', 'else:', 'd[key]', '=', 'cast_fn(value)', 'return', ...
972,328
sek788432/Waymo-2D-Object-Detection
train_utils.py
try_count_params
try_count_params
Count the number of parameters if model is possible.
[ "Count", "the", "number", "of", "parameters", "if", "model", "is", "possible." ]
def try_count_params(model: tf.keras.Model): if hasattr(model, 'count_params'): try: return model.count_params() except ValueError: logging.info('Number of trainable params unknown, because the build() methods in keras layers were not called. This is probably because the mode...
['def', 'try_count_params(model:', 'tf.keras.Model):', 'if', 'hasattr(model,', "'count_params'):", 'try:', 'return', 'model.count_params()', 'except', 'ValueError:', "logging.info('Number", 'of', 'trainable', 'params', 'unknown,', 'because', 'the', 'build()', 'methods', 'in', 'keras', 'layers', 'were', 'not', 'called.'...
972,337
sek788432/Waymo-2D-Object-Detection
train_utils.py
BestCheckpointExporter.best_ckpt_path
best_ckpt_path
Returns the best ckpt path or None if there is no ckpt yet.
[ "Returns", "the", "best", "ckpt", "path", "or", "None", "if", "there", "is", "no", "ckpt", "yet." ]
def best_ckpt_path(self): return tf.train.latest_checkpoint(self._export_dir)
['def', 'best_ckpt_path(self):', 'return', 'tf.train.latest_checkpoint(self._export_dir)']
972,338
sek788432/Waymo-2D-Object-Detection
base_model.py
MultiTaskBaseModel.initialize
initialize
Optional function that loads a pre-train checkpoint.
[ "Optional", "function", "that", "loads", "a", "pre-train", "checkpoint." ]
def initialize(self): return
['def', 'initialize(self):', 'return']
972,368
sek788432/Waymo-2D-Object-Detection
base_trainer.py
MultiTaskBaseTrainer.train_loop_begin
train_loop_begin
Clean up states that hold losses and metrics.
[ "Clean", "up", "states", "that", "hold", "losses", "and", "metrics." ]
def train_loop_begin(self): for (_, train_loss_metric) in self.training_losses.items(): train_loss_metric.reset_states() for (_, metrics) in self.training_metrics.items(): for metric in metrics: metric.reset_states()
['def', 'train_loop_begin(self):', 'for', '(_,', 'train_loss_metric)', 'in', 'self.training_losses.items():', 'train_loss_metric.reset_states()', 'for', '(_,', 'metrics)', 'in', 'self.training_metrics.items():', 'for', 'metric', 'in', 'metrics:', 'metric.reset_states()']
972,369
sek788432/Waymo-2D-Object-Detection
base_trainer.py
MultiTaskBaseTrainer.train_loop_end
train_loop_end
Record loss and metric values per task.
[ "Record", "loss", "and", "metric", "values", "per", "task." ]
def train_loop_end(self): result = {} for (task_name, loss) in self.training_losses.items(): result[task_name] = {loss.name: loss.result()} for (task_name, task_metrics) in self.training_metrics.items(): result[task_name].update({metric.name: metric.result() for metric in task_metrics}) ...
['def', 'train_loop_end(self):', 'result', '=', '{}', 'for', '(task_name,', 'loss)', 'in', 'self.training_losses.items():', 'result[task_name]', '=', '{loss.name:', 'loss.result()}', 'for', '(task_name,', 'task_metrics)', 'in', 'self.training_metrics.items():', 'result[task_name].update({metric.name:', 'metric.result()...
972,370
sek788432/Waymo-2D-Object-Detection
base_trainer.py
MultiTaskBaseTrainer.training_losses
training_losses
Access training loss metric objects for all tasks.
[ "Access", "training", "loss", "metric", "objects", "for", "all", "tasks." ]
def training_losses(self): if self._training_losses is None: self._training_losses = dict(total_loss=tf.keras.metrics.Mean('training_loss', dtype=tf.float32)) for name in self.multi_task.tasks: self._training_losses[name] = tf.keras.metrics.Mean('training_loss', dtype=tf.float32) ret...
['def', 'training_losses(self):', 'if', 'self._training_losses', 'is', 'None:', 'self._training_losses', '=', "dict(total_loss=tf.keras.metrics.Mean('training_loss',", 'dtype=tf.float32))', 'for', 'name', 'in', 'self.multi_task.tasks:', 'self._training_losses[name]', '=', "tf.keras.metrics.Mean('training_loss',", 'dtyp...
972,372
sek788432/Waymo-2D-Object-Detection
base_trainer.py
MultiTaskBaseTrainer.train_step
train_step
The default train step calling the multi-task train step.
[ "The", "default", "train", "step", "calling", "the", "multi-task", "train", "step." ]
def train_step(self, iterator_map): def step_fn(inputs): losses = self.multi_task.joint_train_step(inputs, multi_task_model=self.multi_task_model, optimizer=self.optimizer, task_metrics=self.training_metrics) for (key, loss) in losses.items(): self.training_losses[key].update_state(loss...
['def', 'train_step(self,', 'iterator_map):', 'def', 'step_fn(inputs):', 'losses', '=', 'self.multi_task.joint_train_step(inputs,', 'multi_task_model=self.multi_task_model,', 'optimizer=self.optimizer,', 'task_metrics=self.training_metrics)', 'for', '(key,', 'loss)', 'in', 'losses.items():', 'self.training_losses[key]....
972,374
sek788432/Waymo-2D-Object-Detection
task_sampler.py
get_task_sampler
get_task_sampler
Utils to create task sampler with configuration and task weights.
[ "Utils", "to", "create", "task", "sampler", "with", "configuration", "and", "task", "weights." ]
def get_task_sampler(config: configs.TaskSamplingConfig, task_weights: Dict[Text, float]) -> TaskSampler: oneof_config = config.get() if config.type == 'uniform': return UniformTaskSampler(task_weights=task_weights) elif config.type == 'proportional': return ProportionalTaskSampler(task_weig...
['def', 'get_task_sampler(config:', 'configs.TaskSamplingConfig,', 'task_weights:', 'Dict[Text,', 'float])', '->', 'TaskSampler:', 'oneof_config', '=', 'config.get()', 'if', 'config.type', '==', "'uniform':", 'return', 'UniformTaskSampler(task_weights=task_weights)', 'elif', 'config.type', '==', "'proportional':", 'ret...
972,380
sek788432/Waymo-2D-Object-Detection
train_lib_test.py
ProgMockTask.get_optimizer
get_optimizer
Build optimizer for each stage.
[ "Build", "optimizer", "for", "each", "stage." ]
def get_optimizer(self, stage_id): params = optimization.OptimizationConfig({'optimizer': {'type': 'adamw'}, 'learning_rate': {'type': 'polynomial', 'polynomial': {'initial_learning_rate': 0.01, 'end_learning_rate': 0.0, 'power': 1.0, 'decay_steps': 10}}, 'warmup': {'polynomial': {'power': 1, 'warmup_steps': 2}, 't...
['def', 'get_optimizer(self,', 'stage_id):', 'params', '=', "optimization.OptimizationConfig({'optimizer':", "{'type':", "'adamw'},", "'learning_rate':", "{'type':", "'polynomial',", "'polynomial':", "{'initial_learning_rate':", '0.01,', "'end_learning_rate':", '0.0,', "'power':", '1.0,', "'decay_steps':", '10}},', "'w...
972,401
sek788432/Waymo-2D-Object-Detection
model_saving_utils.py
export_bert_model
export_bert_model
Export BERT model for serving which does not include the optimizer.
[ "Export", "BERT", "model", "for", "serving", "which", "does", "not", "include", "the", "optimizer." ]
def export_bert_model(model_export_path: typing.Text, model: tf.keras.Model, checkpoint_dir: typing.Optional[typing.Text]=None, restore_model_using_load_weights: bool=False) -> None: if not model_export_path: raise ValueError('model_export_path must be specified.') if not isinstance(model, tf.keras.Mode...
['def', 'export_bert_model(model_export_path:', 'typing.Text,', 'model:', 'tf.keras.Model,', 'checkpoint_dir:', 'typing.Optional[typing.Text]=None,', 'restore_model_using_load_weights:', 'bool=False)', '->', 'None:', 'if', 'not', 'model_export_path:', 'raise', "ValueError('model_export_path", 'must', 'be', "specified.'...
972,431
sek788432/Waymo-2D-Object-Detection
model_training_utils.py
steps_to_run
steps_to_run
Calculates steps to run on device.
[ "Calculates", "steps", "to", "run", "on", "device." ]
def steps_to_run(current_step, steps_per_epoch, steps_per_loop): if steps_per_loop <= 0: raise ValueError('steps_per_loop should be positive integer.') if steps_per_loop == 1: return steps_per_loop remainder_in_epoch = current_step % steps_per_epoch if remainder_in_epoch != 0: re...
['def', 'steps_to_run(current_step,', 'steps_per_epoch,', 'steps_per_loop):', 'if', 'steps_per_loop', '<=', '0:', 'raise', "ValueError('steps_per_loop", 'should', 'be', 'positive', "integer.')", 'if', 'steps_per_loop', '==', '1:', 'return', 'steps_per_loop', 'remainder_in_epoch', '=', 'current_step', '%', 'steps_per_ep...
972,432
sek788432/Waymo-2D-Object-Detection
run_classifier.py
run_keras_compile_fit
run_keras_compile_fit
Runs BERT classifier model using Keras compile/fit API.
[ "Runs", "BERT", "classifier", "model", "using", "Keras", "compile/fit", "API." ]
def run_keras_compile_fit(model_dir, strategy, model_fn, train_input_fn, eval_input_fn, loss_fn, metric_fn, init_checkpoint, epochs, steps_per_epoch, steps_per_loop, eval_steps, training_callbacks=True, custom_callbacks=None): with strategy.scope(): training_dataset = train_input_fn() evaluation_dat...
['def', 'run_keras_compile_fit(model_dir,', 'strategy,', 'model_fn,', 'train_input_fn,', 'eval_input_fn,', 'loss_fn,', 'metric_fn,', 'init_checkpoint,', 'epochs,', 'steps_per_epoch,', 'steps_per_loop,', 'eval_steps,', 'training_callbacks=True,', 'custom_callbacks=None):', 'with', 'strategy.scope():', 'training_dataset'...
972,442
sek788432/Waymo-2D-Object-Detection
run_squad_helper.py
get_squad_model_to_predict
get_squad_model_to_predict
Gets a squad model to make predictions.
[ "Gets", "a", "squad", "model", "to", "make", "predictions." ]
def get_squad_model_to_predict(strategy, bert_config, checkpoint_path, input_meta_data): with strategy.scope(): tf.keras.mixed_precision.set_global_policy('float32') (squad_model, _) = bert_models.squad_model(bert_config, input_meta_data['max_seq_length'], hub_module_url=FLAGS.hub_module_url) if...
['def', 'get_squad_model_to_predict(strategy,', 'bert_config,', 'checkpoint_path,', 'input_meta_data):', 'with', 'strategy.scope():', "tf.keras.mixed_precision.set_global_policy('float32')", '(squad_model,', '_)', '=', 'bert_models.squad_model(bert_config,', "input_meta_data['max_seq_length'],", 'hub_module_url=FLAGS.h...
972,458
sek788432/Waymo-2D-Object-Detection
create_pretraining_data.py
write_instance_to_example_files
write_instance_to_example_files
Creates TF example files from `TrainingInstance`s.
[ "Creates", "TF", "example", "files", "from", "`TrainingInstance`s." ]
def write_instance_to_example_files(instances, tokenizer, max_seq_length, max_predictions_per_seq, output_files, gzip_compress, use_v2_feature_names): writers = [] for output_file in output_files: writers.append(tf.io.TFRecordWriter(output_file, options='GZIP' if gzip_compress else '')) writer_index...
['def', 'write_instance_to_example_files(instances,', 'tokenizer,', 'max_seq_length,', 'max_predictions_per_seq,', 'output_files,', 'gzip_compress,', 'use_v2_feature_names):', 'writers', '=', '[]', 'for', 'output_file', 'in', 'output_files:', 'writers.append(tf.io.TFRecordWriter(output_file,', "options='GZIP'", 'if', '...
972,499
sek788432/Waymo-2D-Object-Detection
create_xlnet_pretraining_data.py
create_tfrecords
create_tfrecords
Runs the end-to-end preprocessing pipeline.
[ "Runs", "the", "end-to-end", "preprocessing", "pipeline." ]
def create_tfrecords(tokenizer: tokenization.FullSentencePieceTokenizer, input_file_or_files: str, use_eod_token: bool, do_lower_case: bool, per_host_batch_size: int, seq_length: int, reuse_length: int, bi_data: bool, num_cores_per_host: int, save_dir: str, prefix: str='', suffix: str='', num_tasks: Optional[int]=None,...
['def', 'create_tfrecords(tokenizer:', 'tokenization.FullSentencePieceTokenizer,', 'input_file_or_files:', 'str,', 'use_eod_token:', 'bool,', 'do_lower_case:', 'bool,', 'per_host_batch_size:', 'int,', 'seq_length:', 'int,', 'reuse_length:', 'int,', 'bi_data:', 'bool,', 'num_cores_per_host:', 'int,', 'save_dir:', 'str,'...
972,508
sek788432/Waymo-2D-Object-Detection
train_sentencepiece.py
dump_chars_to_textfile
dump_chars_to_textfile
Write part of a TFDS sentence dataset to lines in a text file.
[ "Write", "part", "of", "a", "TFDS", "sentence", "dataset", "to", "lines", "in", "a", "text", "file." ]
def dump_chars_to_textfile(dataset: tf.data.Dataset, data_keys: Tuple[str], max_char: int=-1): ds_iter = dataset.as_numpy_iterator() with tempfile.NamedTemporaryFile(delete=False) as outfp: char_count = 0 while True: example = next(ds_iter, None) if example is None or (ma...
['def', 'dump_chars_to_textfile(dataset:', 'tf.data.Dataset,', 'data_keys:', 'Tuple[str],', 'max_char:', 'int=-1):', 'ds_iter', '=', 'dataset.as_numpy_iterator()', 'with', 'tempfile.NamedTemporaryFile(delete=False)', 'as', 'outfp:', 'char_count', '=', '0', 'while', 'True:', 'example', '=', 'next(ds_iter,', 'None)', 'if...
972,529
sek788432/Waymo-2D-Object-Detection
train_sentencepiece.py
train_sentencepiece
train_sentencepiece
Train SentencePiece tokenizer from subset of tf dataset.
[ "Train", "SentencePiece", "tokenizer", "from", "subset", "of", "tf", "dataset." ]
def train_sentencepiece(file_path: str, model_path: str, vocab_size: int, character_coverage: float, model_type: str): argstr = ' '.join([f'--input={file_path}', f'--vocab_size={vocab_size}', f'--character_coverage={character_coverage}', f'--model_prefix={model_path}', f'--model_type={model_type}', '--bos_id=-1', '...
['def', 'train_sentencepiece(file_path:', 'str,', 'model_path:', 'str,', 'vocab_size:', 'int,', 'character_coverage:', 'float,', 'model_type:', 'str):', 'argstr', '=', "'", "'.join([f'--input={file_path}',", "f'--vocab_size={vocab_size}',", "f'--character_coverage={character_coverage}',", "f'--model_prefix={model_path}...
972,530
sek788432/Waymo-2D-Object-Detection
binary_helper.py
override_qa_task_config
override_qa_task_config
Overrides a `QuestionAnsweringConfig` object.
[ "Overrides", "a", "`QuestionAnsweringConfig`", "object." ]
def override_qa_task_config(task_cfg: question_answering.QuestionAnsweringConfig, model_config_file: str, init_checkpoint: str, hub_module_url: str, global_batch_size: int, train_input_path: str, validation_input_path: str, seq_length: int, tokenization: str, vocab_file: str, do_lower_case: bool, version_2_with_negativ...
['def', 'override_qa_task_config(task_cfg:', 'question_answering.QuestionAnsweringConfig,', 'model_config_file:', 'str,', 'init_checkpoint:', 'str,', 'hub_module_url:', 'str,', 'global_batch_size:', 'int,', 'train_input_path:', 'str,', 'validation_input_path:', 'str,', 'seq_length:', 'int,', 'tokenization:', 'str,', 'v...
972,534
sek788432/Waymo-2D-Object-Detection
binary_helper.py
override_tagging_task_config
override_tagging_task_config
Overrides a `TaggingConfig` object.
[ "Overrides", "a", "`TaggingConfig`", "object." ]
def override_tagging_task_config(task_cfg: tagging.TaggingConfig, model_config_file: str, init_checkpoint: str, hub_module_url: str, global_batch_size: int, train_input_path: str, validation_input_path: str, seq_length: int, class_names: List[str]): task_cfg.override({'init_checkpoint': init_checkpoint, 'model': {'...
['def', 'override_tagging_task_config(task_cfg:', 'tagging.TaggingConfig,', 'model_config_file:', 'str,', 'init_checkpoint:', 'str,', 'hub_module_url:', 'str,', 'global_batch_size:', 'int,', 'train_input_path:', 'str,', 'validation_input_path:', 'str,', 'seq_length:', 'int,', 'class_names:', 'List[str]):', "task_cfg.ov...
972,535