Compocyte.core.models.fit_methods

Functions

dataloaders_from_dask(x, y, batch_size, ...)
dataloaders_from_dense(x, y, batch_size, ...)
fit(model, x, y[, standardize_idx])
fit_logreg(model, x, y, **fit_kwargs)
fit_torch(model, x, y[, epochs, batch_size, ...])
fit_trees(model, x, y, **fit_kwargs)
predict(model, x[, threshold, monte_carlo])
predict_logits(model, x)
samples_per_class(y)
set_threads(num_threads, parallelize)
to_categorical(y, num_classes[, dtype])	Simplified from keras to avoid dependency and premature conversion to a Tensor.

Classes

DaskBatchDataset(X, y)

Compocyte.core.models.fit_methods.to_categorical(y, num_classes, dtype='float32')

Simplified from keras to avoid dependency and premature conversion to a Tensor.

class Compocyte.core.models.fit_methods.DaskBatchDataset(X, y)

Bases: IterableDataset

__init__(X, y)

set_epoch(epoch)

Compocyte.core.models.fit_methods.predict_logits(model, x)

Compocyte.core.models.fit_methods.predict(model, x, threshold=-1, monte_carlo=None)

Parameters:

monte_carlo (int)

Compocyte.core.models.fit_methods.samples_per_class(y)

Compocyte.core.models.fit_methods.set_threads(num_threads, parallelize)

Compocyte.core.models.fit_methods.dataloaders_from_dask(x, y, batch_size, num_workers)

Compocyte.core.models.fit_methods.dataloaders_from_dense(x, y, batch_size, num_workers)

Compocyte.core.models.fit_methods.fit_torch(model, x, y, epochs=40, batch_size=64, starting_lr=0.01, max_lr=0.1, momentum=0.5, parallelize=True, num_threads=1, beta=0.8, gamma=2.0, class_balance=True, max_cells=1000000)

Parameters:

• model (DenseTorch)

• x (array)

• y (array)

• epochs (int)

• batch_size (int)

• starting_lr (float)

• max_lr (float)

• momentum (float)

• parallelize (bool)

• num_threads (int)

• beta (float)

• gamma (float)

• class_balance (bool)

• max_cells (int)

Compocyte.core.models.fit_methods.fit_logreg(model, x, y, **fit_kwargs)

Parameters:

model (LogisticRegression)

Compocyte.core.models.fit_methods.fit_trees(model, x, y, **fit_kwargs)

Parameters:

model (BoostedTrees)

Compocyte.core.models.fit_methods.fit(model, x, y, standardize_idx=None, **fit_kwargs)

Parameters:

• model (Union[DenseTorch, LogisticRegression, DummyClassifier]) – Model to be fitted.

• x (np.array) – Input data.

• y (np.array) – Target data in the shape of a 1-dimensional array of label strings.

• standardize_idx (list)

Returns:

_description_

Return type:

_type_