Core
Imports
Main imports from other libraries.
DataBlock
DataBlock (blocks:list=None, dl_type:TfmdDL=None, getters:list=None, n_inp:int=None, item_tfms:list=None, batch_tfms:list=None, get_items=None, splitter=None, get_y=None, get_x=None)
Generic container to quickly build Datasets
and DataLoaders
.
Type | Default | Details | |
---|---|---|---|
blocks | list | None | One or more TransformBlock s |
dl_type | TfmdDL | None | Task specific TfmdDL , defaults to block ’s dl_type orTfmdDL |
getters | list | None | Getter functions applied to results of get_items |
n_inp | int | None | Number of inputs |
item_tfms | list | None | ItemTransform s, applied on an item |
batch_tfms | list | None | Transform s or RandTransform s, applied by batch |
get_items | NoneType | None | |
splitter | NoneType | None | |
get_y | NoneType | None | |
get_x | NoneType | None |
DataLoaders
DataLoaders (*loaders, path:str|pathlib.Path='.', device=None)
Basic wrapper around several DataLoader
s.
Learner
Group together a model
, some dls
and a loss_func
to handle training
ShowGraphCallback
ShowGraphCallback (after_create=None, before_fit=None, before_epoch=None, before_train=None, before_batch=None, after_pred=None, after_loss=None, before_backward=None, after_cancel_backward=None, after_backward=None, before_step=None, after_cancel_step=None, after_step=None, after_cancel_batch=None, after_batch=None, after_cancel_train=None, after_train=None, before_validate=None, after_cancel_validate=None, after_validate=None, after_cancel_epoch=None, after_epoch=None, after_cancel_fit=None, after_fit=None)
Update a graph of training and validation loss
CSVLogger
CSVLogger (fname='history.csv', append=False)
Log the results displayed in learn.path/fname
cells3d
cells3d ()
*3D fluorescence microscopy image of cells.
The returned data is a 3D multichannel array with dimensions provided in (z, c, y, x)
order. Each voxel has a size of (0.29 0.26 0.26)
micrometer. Channel 0 contains cell membranes, channel 1 contains nuclei.*
Engine
Core engine for model training. See tutorials for usage examples.
fastTrainer
fastTrainer (dataloaders:fastai.data.core.DataLoaders, model:<built- infunctioncallable>, loss_fn:typing.Any|None=None, optimizer :fastai.optimizer.Optimizer|fastai.optimizer.OptimWrapper=<f unction Adam>, lr:float|slice=0.001, splitter:<built- infunctioncallable>=<function trainable_params>, callbacks:U nion[fastai.callback.core.Callback,MutableSequence,NoneType] =None, metrics:Union[Any,MutableSequence,NoneType]=None, csv_log:bool=False, show_graph:bool=True, show_summary:bool=False, find_lr:bool=False, find_lr_fn=<function valley>, path:str|pathlib.Path|None=None, model_dir:str|pathlib.Path='models', wd:float|int|None=None, wd_bn_bias:bool=False, train_bn:bool=True, moms:tuple=Ellipsis, default_cbs:bool=True)
A custom implementation of the FastAI Learner class for training models in bioinformatics applications.
Type | Default | Details | |
---|---|---|---|
dataloaders | DataLoaders | The DataLoader objects containing training and validation datasets. | |
model | callable | A callable model that will be trained on the dataset. | |
loss_fn | typing.Any | None | None | The loss function to optimize during training. If None, defaults to a suitable default. |
optimizer | fastai.optimizer.Optimizer | fastai.optimizer.OptimWrapper | Adam | The optimizer function to use. Defaults to Adam if not specified. |
lr | float | slice | 0.001 | Learning rate for the optimizer. Can be a float or a slice object for learning rate scheduling. |
splitter | callable | trainable_params | |
callbacks | Union | None | A callable that determines which parameters of the model should be updated during training. |
metrics | Union | None | Optional list of callback functions to customize training behavior. |
csv_log | bool | False | Metrics to evaluate the performance of the model during training. |
show_graph | bool | True | Whether to log training history to a CSV file. If True, logs will be appended to ‘history.csv’. |
show_summary | bool | False | The base directory where models are saved or loaded from. Defaults to None. |
find_lr | bool | False | Subdirectory within the base path where trained models are stored. Default is ‘models’. |
find_lr_fn | function | valley | Weight decay factor for optimization. Defaults to None. |
path | str | pathlib.Path | None | None | Whether to apply weight decay to batch normalization and bias parameters. |
model_dir | str | pathlib.Path | models | Whether to update the batch normalization statistics during training. |
wd | float | int | None | None | |
wd_bn_bias | bool | False | |
train_bn | bool | True | |
moms | tuple | Ellipsis | Tuple of tuples representing the momentum values for different layers in the model. Defaults to FastAI’s default settings if not specified. |
default_cbs | bool | True | Automatically include default callbacks such as ShowGraphCallback and CSVLogger. |
visionTrainer
visionTrainer (dataloaders:fastai.data.core.DataLoaders, model:<built- infunctioncallable>, normalize=True, n_out=None, pretrained=True, weights=None, loss_fn:typing.Any|None=None, optimizer:fastai.optimizer.O ptimizer|fastai.optimizer.OptimWrapper=<function Adam>, lr:float|slice=0.001, splitter:<built- infunctioncallable>=<function trainable_params>, callbacks :Union[fastai.callback.core.Callback,MutableSequence,NoneT ype]=None, metrics:Union[Any,MutableSequence,NoneType]=None, csv_log:bool=False, show_graph:bool=True, show_summary:bool=False, find_lr:bool=False, find_lr_fn=<function valley>, path:str|pathlib.Path|None=None, model_dir:str|pathlib.Path='models', wd:float|int|None=None, wd_bn_bias:bool=False, train_bn:bool=True, moms:tuple=Ellipsis, default_cbs:bool=True, cut=None, init=<function kaiming_normal_>, custom_head=None, concat_pool=True, pool=True, lin_ftrs=None, ps=0.5, first_bn=True, bn_final=False, lin_first=False, y_range=None, n_in=3)
Build a vision trainer from dataloaders
and model
Type | Default | Details | |
---|---|---|---|
dataloaders | DataLoaders | The DataLoader objects containing training and validation datasets. | |
model | callable | A callable model that will be trained on the dataset. | |
normalize | bool | True | |
n_out | NoneType | None | |
pretrained | bool | True | |
weights | NoneType | None | |
loss_fn | typing.Any | None | None | The loss function to optimize during training. If None, defaults to a suitable default. |
optimizer | fastai.optimizer.Optimizer | fastai.optimizer.OptimWrapper | Adam | The optimizer function to use. Defaults to Adam if not specified. |
lr | float | slice | 0.001 | Learning rate for the optimizer. Can be a float or a slice object for learning rate scheduling. |
splitter | callable | trainable_params | |
callbacks | Union | None | A callable that determines which parameters of the model should be updated during training. |
metrics | Union | None | Optional list of callback functions to customize training behavior. |
csv_log | bool | False | Metrics to evaluate the performance of the model during training. |
show_graph | bool | True | Whether to log training history to a CSV file. If True, logs will be appended to ‘history.csv’. |
show_summary | bool | False | The base directory where models are saved or loaded from. Defaults to None. |
find_lr | bool | False | Subdirectory within the base path where trained models are stored. Default is ‘models’. |
find_lr_fn | function | valley | Weight decay factor for optimization. Defaults to None. |
path | str | pathlib.Path | None | None | Whether to apply weight decay to batch normalization and bias parameters. |
model_dir | str | pathlib.Path | models | Whether to update the batch normalization statistics during training. |
wd | float | int | None | None | |
wd_bn_bias | bool | False | |
train_bn | bool | True | |
moms | tuple | Ellipsis | Tuple of tuples representing the momentum values for different layers in the model. Defaults to FastAI’s default settings if not specified. |
default_cbs | bool | True | Automatically include default callbacks such as ShowGraphCallback and CSVLogger. |
cut | NoneType | None | model & head args |
init | function | kaiming_normal_ | |
custom_head | NoneType | None | |
concat_pool | bool | True | |
pool | bool | True | |
lin_ftrs | NoneType | None | |
ps | float | 0.5 | |
first_bn | bool | True | |
bn_final | bool | False | |
lin_first | bool | False | |
y_range | NoneType | None | |
n_in | int | 3 |
Utils
Utility functions.
attributesFromDict
attributesFromDict (d)
get_device
get_device ()
img2float
img2float (image, force_copy=False)
img2Tensor
img2Tensor (image)