Skip to content

Completed docs #400

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 5 commits into from
Apr 25, 2022
Merged

Completed docs #400

Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
5032c05
Completed docs
Fl4m3Ph03n1x Apr 21, 2022
abf73ca
Added numpy typing
Fl4m3Ph03n1x Apr 22, 2022
b30f7e0
updating deps
Fl4m3Ph03n1x Apr 22, 2022
baa9ea7
reverting back to np.ndarray
Fl4m3Ph03n1x Apr 22, 2022
e3b6a45
reverted changed to stup
Fl4m3Ph03n1x Apr 22, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
179 changes: 137 additions & 42 deletions darwin/torch/dataset.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -89,7 +89,7 @@ class ClassificationDataset(LocalDataset):

Parameters
----------
transform: Optional[Union[Callable, List]], default: None
transform: Optional[Union[Callable, List[Callable]]], default: None
torchvision function or list to set the ``transform`` attribute. If it is a list, it will
be composed via torchvision.
"""
Expand All @@ -105,7 +105,7 @@ def __init__(self, transform: Optional[Union[Callable, List]] = None, **kwargs):
self.is_multi_label = False
self.check_if_multi_label()

def __getitem__(self, index: int) -> Tuple[torch.Tensor, torch.Tensor]:
def __getitem__(self, index: int) -> Tuple[Tensor, Tensor]:
"""
See superclass for documentation.

Expand All @@ -116,7 +116,7 @@ def __getitem__(self, index: int) -> Tuple[torch.Tensor, torch.Tensor]:

Returns
-------
Tuple[torch.Tensor, torch.Tensor]
Tuple[Tensor, Tensor]
A tuple of tensors, where the first value is the image tensor and the second is the
target's tensor.
"""
Expand All @@ -130,7 +130,7 @@ def __getitem__(self, index: int) -> Tuple[torch.Tensor, torch.Tensor]:

return img_tensor, target

def get_target(self, index: int) -> torch.Tensor:
def get_target(self, index: int) -> Tensor:
"""
Returns the classification target.

Expand All @@ -141,7 +141,7 @@ def get_target(self, index: int) -> torch.Tensor:

Returns
-------
torch.Tensor
Tensor
The target's tensor.
"""

Expand All @@ -151,7 +151,7 @@ def get_target(self, index: int) -> torch.Tensor:

assert len(tags) >= 1, f"No tags were found for index={index}"

target: torch.Tensor = torch.tensor(self.classes.index(tags[0]))
target: Tensor = torch.tensor(self.classes.index(tags[0]))

if self.is_multi_label:
target = torch.zeros(len(self.classes))
Expand Down Expand Up @@ -190,26 +190,26 @@ def get_class_idx(self, index: int) -> int:
int
``category_id`` of the image.
"""
target: torch.Tensor = self.get_target(index)
target: Tensor = self.get_target(index)
return target["category_id"]

def measure_weights(self, **kwargs) -> np.ndarray:
Copy link
Member

@andreaazzini andreaazzini Apr 21, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can we run a test, just to make sure that calling measure_weights with no kwargs is valid?

Copy link
Contributor Author

@Fl4m3Ph03n1x Fl4m3Ph03n1x Apr 22, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

How would I run a test for that? (I don;t know enough to know how to use this :S )

def measure_weights(self) -> np.ndarray:
"""
Computes the class balancing weights (not the frequencies!!) given the train loader.
Gets the weights proportional to the inverse of their class frequencies.
The vector sums up to 1.

Returns
-------
ndarray[double]
np.ndarray[float]
Weight for each class in the train set (one for each class) as a 1D array normalized.
"""
# Collect all the labels by iterating over the whole dataset
labels = []
for i, _filename in enumerate(self.images_path):
target = self.get_target(i)
target: Tensor = self.get_target(i)
if self.is_multi_label:
# get the indixes of the class present
# get the indices of the class present
target = torch.where(target == 1)[0]
labels.extend(target.tolist())
else:
Expand All @@ -233,7 +233,7 @@ class InstanceSegmentationDataset(LocalDataset):

Parameters
----------
transform: Optional[Union[Callable, List]], default: None
transform: Optional[Union[Callable, List[Callable]]], default: None
torchvision function or list to set the ``transform`` attribute. If it is a list, it will
be composed via torchvision.
"""
Expand All @@ -248,7 +248,7 @@ def __init__(self, transform: Optional[Union[Callable, List]] = None, **kwargs):

self.convert_polygons = ConvertPolygonsToInstanceMasks()

def __getitem__(self, index: int) -> Tuple[torch.Tensor, Dict[str, Any]]:
def __getitem__(self, index: int) -> Tuple[Tensor, Dict[str, Any]]:
"""
Notes
-----
Expand Down Expand Up @@ -279,7 +279,30 @@ def __getitem__(self, index: int) -> Tuple[torch.Tensor, Dict[str, Any]]:

def get_target(self, index: int) -> Dict[str, Any]:
"""
Returns the instance segmentation target
Builds and returns the target dictionary for the item at the given index.
The target dictionary will have the following format:

.. code-block:: python
{
"annotations": [
{
"category_id": int,
"segmentation": List[List[int | float]],
"bbox": List[float],
"area": float
}
]
}

Parameters
----------
index : int
The actual index of the item in the ``Dataset``.

Returns
-------
Dict[str, Any]
The target.
"""
target = self.parse_json(index)

Expand All @@ -297,15 +320,15 @@ def get_target(self, index: int) -> Dict[str, Any]:
# Compute the bbox of the polygon
x_coords = [s[0::2] for s in sequences]
y_coords = [s[1::2] for s in sequences]
min_x = np.min([np.min(x_coord) for x_coord in x_coords])
min_y = np.min([np.min(y_coord) for y_coord in y_coords])
max_x = np.max([np.max(x_coord) for x_coord in x_coords])
max_y = np.max([np.max(y_coord) for y_coord in y_coords])
w = max_x - min_x + 1
h = max_y - min_y + 1
min_x: float = np.min([np.min(x_coord) for x_coord in x_coords])
min_y: float = np.min([np.min(y_coord) for y_coord in y_coords])
max_x: float = np.max([np.max(x_coord) for x_coord in x_coords])
max_y: float = np.max([np.max(y_coord) for y_coord in y_coords])
w: float = max_x - min_x + 1
h: float = max_y - min_y + 1
# Compute the area of the polygon
# TODO fix with addictive/subtractive paths in complex polygons
poly_area = np.sum([polygon_area(x_coord, y_coord) for x_coord, y_coord in zip(x_coords, y_coords)])
poly_area: float = np.sum([polygon_area(x_coord, y_coord) for x_coord, y_coord in zip(x_coords, y_coords)])

# Create and append the new entry for this annotation
annotations.append(
Expand All @@ -320,16 +343,16 @@ def get_target(self, index: int) -> Dict[str, Any]:

return target

def measure_weights(self, **kwargs) -> np.ndarray:
def measure_weights(self) -> np.ndarray:
"""
Computes the class balancing weights (not the frequencies!!) given the train loader
Get the weights proportional to the inverse of their class frequencies.
The vector sums up to 1
The vector sums up to 1.

Returns
-------
class_weights : ndarray[double]
Weight for each class in the train set (one for each class) as a 1D array normalized
class_weights : np.ndarray[float]
Weight for each class in the train set (one for each class) as a 1D array normalized.
"""
# Collect all the labels by iterating over the whole dataset
labels: List[int] = []
Expand All @@ -340,10 +363,25 @@ def measure_weights(self, **kwargs) -> np.ndarray:


class SemanticSegmentationDataset(LocalDataset):
def __init__(self, transform: Optional[Union[List, Callable]] = None, **kwargs):
"""
See `LocalDataset` class for documentation
"""
"""
Represents an instance of a LocalDataset used for training on semantic segmentation tasks.

Attributes
----------
transform : Optional[Callable], default: None
torchvision transform function(s) to run on the dataset.
convert_polygons : ConvertPolygonsToSemanticMask
Object used to convert polygons to semantic masks.

Parameters
----------
transform : Optional[Union[List[Callable], Callable]], default: None
torchvision function or list to set the ``transform`` attribute. If it is a list, it will
be composed via torchvision.
"""

def __init__(self, transform: Optional[Union[List[Callable], Callable]] = None, **kwargs):

super().__init__(annotation_type="polygon", **kwargs)

if transform is not None and isinstance(transform, list):
Expand All @@ -352,7 +390,7 @@ def __init__(self, transform: Optional[Union[List, Callable]] = None, **kwargs):
self.transform: Optional[Callable] = transform
self.convert_polygons = ConvertPolygonsToSemanticMask()

def __getitem__(self, index: int) -> Tuple[torch.Tensor, Dict[str, Any]]:
def __getitem__(self, index: int) -> Tuple[Tensor, Dict[str, Any]]:
"""
See superclass for documentation

Expand All @@ -379,11 +417,32 @@ def __getitem__(self, index: int) -> Tuple[torch.Tensor, Dict[str, Any]]:

def get_target(self, index: int) -> Dict[str, Any]:
"""
Returns the semantic segmentation target
Builds and returns the target dictionary for the item at the given index.
The returned dictionary has the following structure:

.. code-block:: python
{
"annotations": [
{
"category_id": int,
"segmentation": List[List[float | int]]
}
]
}

Parameters
----------
index : int
The actual index of the item in the ``Dataset``.

Returns
-------
Dict[str, Any]
The target.
"""
target = self.parse_json(index)

annotations = []
annotations: List[Dict[str, Union[int, List[List[Union[int, float]]]]]] = []
for obj in target["annotations"]:
sequences = convert_polygons_to_sequences(
obj["polygon"]["path"],
Expand All @@ -399,16 +458,16 @@ def get_target(self, index: int) -> Dict[str, Any]:

return target

def measure_weights(self, **kwargs) -> np.ndarray:
def measure_weights(self) -> np.ndarray:
"""
Computes the class balancing weights (not the frequencies!!) given the train loader
Get the weights proportional to the inverse of their class frequencies.
The vector sums up to 1
The vector sums up to 1.

Returns
-------
class_weights : ndarray[double]
Weight for each class in the train set (one for each class) as a 1D array normalized
class_weights : np.ndarray[float]
Weight for each class in the train set (one for each class) as a 1D array normalized.
"""
# Collect all the labels by iterating over the whole dataset
labels = []
Expand All @@ -419,8 +478,22 @@ def measure_weights(self, **kwargs) -> np.ndarray:


class ObjectDetectionDataset(LocalDataset):
"""
Represents an instance of a LocalDataset used for training on object detection tasks.

Attributes
----------
transform : Optional[Callable], default: None
torchvision transform function(s) to run on the dataset.

Parameters
----------
transform : Optional[Union[List[Callable], Callable]], default: None
torchvision function or list to set the ``transform`` attribute. If it is a list, it will
be composed via torchvision.
"""

def __init__(self, transform: Optional[List] = None, **kwargs):
"""See `LocalDataset` class for documentation"""
super().__init__(annotation_type="bounding_box", **kwargs)

if transform is not None and isinstance(transform, list):
Expand Down Expand Up @@ -455,7 +528,29 @@ def __getitem__(self, index: int):
return img_tensor, target

def get_target(self, index: int) -> Dict[str, Tensor]:
"""Returns the object detection target"""
"""
Builds and returns the target dictionary for the item at the given index.
The returned dictionary has the following structure:

.. code-block:: python
{
"boxes": Tensor,
"area": Tensor,
"labels": Tensor,
"image_id": Tensor,
"iscrowd": Tensor
}

Parameters
----------
index : int
The actual index of the item in the ``Dataset``.

Returns
-------
Dict[str, Any]
The target.
"""
target = self.parse_json(index)
annotations = target.pop("annotations")

Expand Down Expand Up @@ -487,16 +582,16 @@ def get_target(self, index: int) -> Dict[str, Tensor]:

return stacked_targets

def measure_weights(self, **kwargs) -> np.ndarray:
def measure_weights(self) -> np.ndarray:
"""
Computes the class balancing weights (not the frequencies!!) given the train loader
Get the weights proportional to the inverse of their class frequencies.
The vector sums up to 1
The vector sums up to 1.

Returns
-------
class_weights : ndarray[double]
Weight for each class in the train set (one for each class) as a 1D array normalized
class_weights : np.ndarray[float]
Weight for each class in the train set (one for each class) as a 1D array normalized.
"""
# Collect all the labels by iterating over the whole dataset
labels = []
Expand Down