pipelinex.extras.ops package

Subpackages

• pipelinex.extras.ops.ignite package
  • Subpackages
    • pipelinex.extras.ops.ignite.declaratives package
      • Submodules
      • pipelinex.extras.ops.ignite.declaratives.declarative_trainer module
    • pipelinex.extras.ops.ignite.handlers package
      • Submodules
      • pipelinex.extras.ops.ignite.handlers.flexible_checkpoint module
      • pipelinex.extras.ops.ignite.handlers.time_limit module
    • pipelinex.extras.ops.ignite.metrics package
      • Submodules
      • pipelinex.extras.ops.ignite.metrics.cohen_kappa_score module
      • pipelinex.extras.ops.ignite.metrics.fbeta_score module
      • pipelinex.extras.ops.ignite.metrics.utils module

Submodules

pipelinex.extras.ops.allennlp_ops module

class pipelinex.extras.ops.allennlp_ops.AllennlpReaderToDict(**kwargs)

Bases: object

__init__(**kwargs)

Initialize self. See help(type(self)) for accurate signature.

pipelinex.extras.ops.argparse_ops module

class pipelinex.extras.ops.argparse_ops.FeedArgsDict(func, args={}, force_return=None)

Bases: object

__init__(func, args={}, force_return=None)

Initialize self. See help(type(self)) for accurate signature.

pipelinex.extras.ops.argparse_ops.namespace(d)

pipelinex.extras.ops.numpy_ops module

class pipelinex.extras.ops.numpy_ops.ReverseChannel(channel_first=False)

Bases: object

__init__(channel_first=False)

Initialize self. See help(type(self)) for accurate signature.

pipelinex.extras.ops.numpy_ops.reverse_channel(a, channel_first=False)

pipelinex.extras.ops.numpy_ops.to_channel_first_arr(a)

pipelinex.extras.ops.numpy_ops.to_channel_last_arr(a)

pipelinex.extras.ops.opencv_ops module

class pipelinex.extras.ops.opencv_ops.CvBGR2Gray(*args, **kwargs)

Bases: pipelinex.extras.ops.opencv_ops.CvDictToDict

__init__(*args, **kwargs)

Initialize self. See help(type(self)) for accurate signature.

fn = 'cvtColor'

class pipelinex.extras.ops.opencv_ops.CvBGR2HSV(*args, **kwargs)

Bases: pipelinex.extras.ops.opencv_ops.CvDictToDict

__init__(*args, **kwargs)

Initialize self. See help(type(self)) for accurate signature.

fn = 'cvtColor'

class pipelinex.extras.ops.opencv_ops.CvBilateralFilter(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.CvDictToDict

fn = 'bilateralFilter'

class pipelinex.extras.ops.opencv_ops.CvBlur(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.CvDictToDict

fn = 'blur'

class pipelinex.extras.ops.opencv_ops.CvBoxFilter(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.CvDictToDict

fn = 'boxFilter'

class pipelinex.extras.ops.opencv_ops.CvCanny(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.CvDictToDict

fn = 'Canny'

class pipelinex.extras.ops.opencv_ops.CvCvtColor(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.CvDictToDict

fn = 'cvtColor'

class pipelinex.extras.ops.opencv_ops.CvDiagonalEdgeFilter2d(kernel_type=2, **kwargs)

Bases: pipelinex.extras.ops.opencv_ops.CvModuleL2

__init__(kernel_type=2, **kwargs)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.opencv_ops.CvDictToDict(**kwargs)

Bases: pipelinex.utils.DictToDict

module = <module 'cv2' from '/home/docs/checkouts/readthedocs.org/user_builds/pipelinex/envs/latest/lib/python3.8/site-packages/cv2/__init__.py'>

class pipelinex.extras.ops.opencv_ops.CvDilate(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.CvDictToDict

fn = 'dilate'

class pipelinex.extras.ops.opencv_ops.CvEqualizeHist(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.CvDictToDict

fn = 'equalizeHist'

class pipelinex.extras.ops.opencv_ops.CvErode(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.CvDictToDict

fn = 'erode'

class pipelinex.extras.ops.opencv_ops.CvFilter2d(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.CvDictToDict

fn = 'filter2D'

class pipelinex.extras.ops.opencv_ops.CvGaussianBlur(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.CvDictToDict

fn = 'GaussianBlur'

class pipelinex.extras.ops.opencv_ops.CvHoughLinesP(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.CvDictToDict

fn = 'HoughLinesP'

class pipelinex.extras.ops.opencv_ops.CvLine(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.CvDictToDict

fn = 'line'

class pipelinex.extras.ops.opencv_ops.CvMedianBlur(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.CvDictToDict

fn = 'medianBlur'

class pipelinex.extras.ops.opencv_ops.CvModuleConcat(*modules)

Bases: pipelinex.extras.ops.opencv_ops.CvModuleListMerge

class pipelinex.extras.ops.opencv_ops.CvModuleL1(*modules)

Bases: pipelinex.extras.ops.opencv_ops.CvModuleStack

class pipelinex.extras.ops.opencv_ops.CvModuleL2(*modules)

Bases: pipelinex.extras.ops.opencv_ops.CvModuleStack

class pipelinex.extras.ops.opencv_ops.CvModuleListMerge(*modules)

Bases: object

__init__(*modules)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.opencv_ops.CvModuleMean(*modules)

Bases: pipelinex.extras.ops.opencv_ops.CvModuleStack

class pipelinex.extras.ops.opencv_ops.CvModuleStack(*modules)

Bases: pipelinex.extras.ops.opencv_ops.CvModuleListMerge

class pipelinex.extras.ops.opencv_ops.CvModuleSum(*modules)

Bases: pipelinex.extras.ops.opencv_ops.CvModuleStack

class pipelinex.extras.ops.opencv_ops.CvResize(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.CvDictToDict

fn = 'resize'

class pipelinex.extras.ops.opencv_ops.CvScale(width, height)

Bases: object

__init__(width, height)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.opencv_ops.CvSobel(ddepth='CV_64F', **kwargs)

Bases: pipelinex.extras.ops.opencv_ops.CvDictToDict

__init__(ddepth='CV_64F', **kwargs)

Initialize self. See help(type(self)) for accurate signature.

fn = 'Sobel'

class pipelinex.extras.ops.opencv_ops.CvThreshold(type='THRESH_BINARY', **kwargs)

Bases: pipelinex.extras.ops.opencv_ops.CvDictToDict

__init__(type='THRESH_BINARY', **kwargs)

Initialize self. See help(type(self)) for accurate signature.

fn = 'threshold'

class pipelinex.extras.ops.opencv_ops.NpAbs(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.NpDictToDict

fn = 'abs'

class pipelinex.extras.ops.opencv_ops.NpConcat(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.NpDictToDict

fn = 'concatenate'

class pipelinex.extras.ops.opencv_ops.NpDictToDict(**kwargs)

Bases: pipelinex.utils.DictToDict

module = <module 'numpy' from '/home/docs/checkouts/readthedocs.org/user_builds/pipelinex/envs/latest/lib/python3.8/site-packages/numpy/__init__.py'>

class pipelinex.extras.ops.opencv_ops.NpFullLike(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.NpDictToDict

fn = 'full_like'

class pipelinex.extras.ops.opencv_ops.NpMean(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.NpDictToDict

fn = 'mean'

class pipelinex.extras.ops.opencv_ops.NpOnesLike(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.NpDictToDict

fn = 'ones_like'

class pipelinex.extras.ops.opencv_ops.NpSqrt(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.NpDictToDict

fn = 'sqrt'

class pipelinex.extras.ops.opencv_ops.NpSquare(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.NpDictToDict

fn = 'square'

class pipelinex.extras.ops.opencv_ops.NpStack(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.NpDictToDict

fn = 'stack'

class pipelinex.extras.ops.opencv_ops.NpSum(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.NpDictToDict

fn = 'sum'

class pipelinex.extras.ops.opencv_ops.NpZerosLike(**kwargs)

Bases: pipelinex.extras.ops.opencv_ops.NpDictToDict

fn = 'zeros_like'

pipelinex.extras.ops.opencv_ops.expand_repeat(a, repeats=1, axis=None)

pipelinex.extras.ops.opencv_ops.fit_to_1(a)

pipelinex.extras.ops.opencv_ops.fit_to_uint8(a)

pipelinex.extras.ops.opencv_ops.mix_up(*imgs)

pipelinex.extras.ops.opencv_ops.overlay(*imgs)

pipelinex.extras.ops.opencv_ops.sum_up(*imgs)

pipelinex.extras.ops.pandas_ops module

class pipelinex.extras.ops.pandas_ops.DfAddRowStat(**kwargs)

Bases: object

__init__(**kwargs)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfAgg(groupby=None, columns=None, keep_others=False, method=None, **kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseTask

method = 'agg'

class pipelinex.extras.ops.pandas_ops.DfAggregate(groupby=None, columns=None, keep_others=False, method=None, **kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseTask

method = 'aggregate'

class pipelinex.extras.ops.pandas_ops.DfApply(groupby=None, columns=None, keep_others=False, method=None, **kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseTask

method = 'apply'

class pipelinex.extras.ops.pandas_ops.DfApplymap(groupby=None, columns=None, keep_others=False, method=None, **kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseTask

method = 'applymap'

class pipelinex.extras.ops.pandas_ops.DfAssignColumns(names=None, name_fmt='{:03d}')

Bases: object

__init__(names=None, name_fmt='{:03d}')

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfBaseMethod(**kwargs)

Bases: object

__init__(**kwargs)

Initialize self. See help(type(self)) for accurate signature.

method = None

class pipelinex.extras.ops.pandas_ops.DfBaseTask(groupby=None, columns=None, keep_others=False, method=None, **kwargs)

Bases: object

__init__(groupby=None, columns=None, keep_others=False, method=None, **kwargs)

Initialize self. See help(type(self)) for accurate signature.

method = None

class pipelinex.extras.ops.pandas_ops.DfColApply(func, **kwargs)

Bases: object

__init__(func, **kwargs)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfConcat(new_col_name=None, new_col_values=None, col_id=None, sort=False)

Bases: object

__init__(new_col_name=None, new_col_values=None, col_id=None, sort=False)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfCondReplace(flag, columns, value=nan, replace_if_flag=True, **kwargs)

Bases: object

__init__(flag, columns, value=nan, replace_if_flag=True, **kwargs)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfDrop(**kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseMethod

method = 'drop'

class pipelinex.extras.ops.pandas_ops.DfDropDuplicates(**kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseMethod

method = 'drop_duplicates'

class pipelinex.extras.ops.pandas_ops.DfDropFilter(**kwargs)

Bases: object

__init__(**kwargs)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfDtypesApply(func, **kwargs)

Bases: object

__init__(func, **kwargs)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfDuplicate(columns)

Bases: object

__init__(columns)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfEval(expr, parser='pandas', engine=None, truediv=True)

Bases: object

__init__(expr, parser='pandas', engine=None, truediv=True)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfEwm(groupby=None, columns=None, keep_others=False, method=None, **kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseTask

method = 'ewm'

class pipelinex.extras.ops.pandas_ops.DfExpanding(groupby=None, columns=None, keep_others=False, method=None, **kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseTask

method = 'expanding'

class pipelinex.extras.ops.pandas_ops.DfFillna(groupby=None, columns=None, keep_others=False, method=None, **kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseTask

method = 'fillna'

class pipelinex.extras.ops.pandas_ops.DfFilter(groupby=None, columns=None, keep_others=False, method=None, **kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseTask

class pipelinex.extras.ops.pandas_ops.DfFilterCols(groupby=None, columns=None, keep_others=False, method=None, **kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfFilter

class pipelinex.extras.ops.pandas_ops.DfFocusTransform(focus, columns, groupby=None, keep_others=False, func='max', **kwargs)

Bases: object

__init__(focus, columns, groupby=None, keep_others=False, func='max', **kwargs)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfGetColIndexes(**kwargs)

Bases: object

__init__(**kwargs)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfGetCols

Bases: object

class pipelinex.extras.ops.pandas_ops.DfGetDummies(**kwargs)

Bases: object

__init__(**kwargs)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfGroupby(**kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseMethod

method = 'groupby'

class pipelinex.extras.ops.pandas_ops.DfHead(groupby=None, columns=None, keep_others=False, method=None, **kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseTask

method = 'head'

class pipelinex.extras.ops.pandas_ops.DfMap(arg, prefix='', suffix='', **kwargs)

Bases: object

__init__(arg, prefix='', suffix='', **kwargs)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfMerge(**kwargs)

Bases: object

__init__(**kwargs)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfNgroup(groupby=None, columns=None, keep_others=False, method=None, **kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseTask

method = 'ngroup'

class pipelinex.extras.ops.pandas_ops.DfPipe(groupby=None, columns=None, keep_others=False, method=None, **kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseTask

method = 'pipe'

class pipelinex.extras.ops.pandas_ops.DfQuery(**kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseMethod

method = 'query'

class pipelinex.extras.ops.pandas_ops.DfRelative(focus, columns, groupby=None)

Bases: object

__init__(focus, columns, groupby=None)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfRename(index=None, columns=None, copy=True, level=None)

Bases: object

__init__(index=None, columns=None, copy=True, level=None)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfResample(groupby=None, columns=None, keep_others=False, method=None, **kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseTask

method = 'resample'

class pipelinex.extras.ops.pandas_ops.DfResetIndex(**kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseMethod

method = 'reset_index'

class pipelinex.extras.ops.pandas_ops.DfRolling(groupby=None, columns=None, keep_others=False, method=None, **kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseTask

method = 'rolling'

class pipelinex.extras.ops.pandas_ops.DfRowApply(func, **kwargs)

Bases: object

__init__(func, **kwargs)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfSample(**kwargs)

Bases: object

__init__(**kwargs)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfSelectDtypes(**kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseMethod

method = 'select_dtypes'

class pipelinex.extras.ops.pandas_ops.DfSelectDtypesCols(**kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfSelectDtypes

class pipelinex.extras.ops.pandas_ops.DfSetIndex(**kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseMethod

method = 'set_index'

class pipelinex.extras.ops.pandas_ops.DfShift(groupby=None, columns=None, keep_others=False, method=None, **kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseTask

method = 'shift'

class pipelinex.extras.ops.pandas_ops.DfSlice(**kwargs)

Bases: object

__init__(**kwargs)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfSortValues(**kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseMethod

method = 'sort_values'

class pipelinex.extras.ops.pandas_ops.DfSpatialFeatures(output='distance', coo_cols=['X', 'Y'], groupby=None, ord=None, unit_distance=1.0, affinity_scale=1.0, binary_affinity=False, min_affinity=1e-06, col_name_fmt='feat_{:03d}', keep_others=True, sort=True)

Bases: object

__init__(output='distance', coo_cols=['X', 'Y'], groupby=None, ord=None, unit_distance=1.0, affinity_scale=1.0, binary_affinity=False, min_affinity=1e-06, col_name_fmt='feat_{:03d}', keep_others=True, sort=True)

Available values for output:

distance affinity laplacian eigenvalues eigenvectors n_connected

class pipelinex.extras.ops.pandas_ops.DfStrftime(cols, **kwargs)

Bases: object

__init__(cols, **kwargs)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfTail(groupby=None, columns=None, keep_others=False, method=None, **kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseTask

method = 'tail'

class pipelinex.extras.ops.pandas_ops.DfToDatetime(cols=None, **kwargs)

Bases: object

__init__(cols=None, **kwargs)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfToTimedelta(cols, **kwargs)

Bases: object

__init__(cols, **kwargs)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfTotalSeconds(cols, **kwargs)

Bases: object

__init__(cols, **kwargs)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.DfTransform(groupby=None, columns=None, keep_others=False, method=None, **kwargs)

Bases: pipelinex.extras.ops.pandas_ops.DfBaseTask

method = 'transform'

class pipelinex.extras.ops.pandas_ops.NestedDictToDf(row_oriented=True, index_name='index', reset_index=True)

Bases: object

__init__(row_oriented=True, index_name='index', reset_index=True)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pandas_ops.SrMap(**kwargs)

Bases: object

__init__(**kwargs)

Initialize self. See help(type(self)) for accurate signature.

pipelinex.extras.ops.pandas_ops.affinity_matrix(coo_2darr, ord=None, unit_distance=1.0, affinity_scale=1.0, binary_affinity=False, min_affinity=1e-06, zero_diag=True)

pipelinex.extras.ops.pandas_ops.degree_matrix(affinity_2darr)

pipelinex.extras.ops.pandas_ops.distance_matrix(coo_2darr, ord=None)

pipelinex.extras.ops.pandas_ops.distance_to_affinity(dist_2darr, unit_distance=1.0, affinity_scale=1.0, binary_affinity=False, min_affinity=1e-06)

pipelinex.extras.ops.pandas_ops.eigen(a, return_values=True, values_as_square_matrix=False, return_vectors=False, sort=False)

pipelinex.extras.ops.pandas_ops.laplacian_eigen(coo_2darr, return_values=True, return_vectors=False, ord=None, unit_distance=1.0, affinity_scale=1.0, binary_affinity=False, min_affinity=1e-06, sort=False)

pipelinex.extras.ops.pandas_ops.laplacian_matrix(coo_2darr, ord=None, unit_distance=1.0, affinity_scale=1.0, binary_affinity=False, min_affinity=1e-06)

pipelinex.extras.ops.pandas_ops.nested_dict_to_df(d, row_oriented=True, index_name='index', reset_index=True)

pipelinex.extras.ops.pandas_ops.row_vector_to_square_matrix(a)

pipelinex.extras.ops.pytorch_ops module

class pipelinex.extras.ops.pytorch_ops.CrossEntropyLoss2d(weight=None, size_average=None, ignore_index=- 100, reduce=None, reduction='mean', label_smoothing=0.0)

Bases: torch.nn.modules.loss.CrossEntropyLoss

forward(input, target)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

ignore_index: int

label_smoothing: float

class pipelinex.extras.ops.pytorch_ops.ModuleAvg(*args: torch.nn.modules.module.Module)

class pipelinex.extras.ops.pytorch_ops.ModuleAvg(arg: OrderedDict[str, Module])

Bases: pipelinex.extras.ops.pytorch_ops.ModuleListMerge

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class pipelinex.extras.ops.pytorch_ops.ModuleBottleneck2d(in_channels, out_channels, kernel_size=(1, 1), stride=(1, 1), mid_channels=None, batch_norm=None, activation=None, **kwargs)

Bases: torch.nn.modules.container.Sequential

__init__(in_channels, out_channels, kernel_size=(1, 1), stride=(1, 1), mid_channels=None, batch_norm=None, activation=None, **kwargs)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

class pipelinex.extras.ops.pytorch_ops.ModuleConcat(*args: torch.nn.modules.module.Module)

class pipelinex.extras.ops.pytorch_ops.ModuleConcat(arg: OrderedDict[str, Module])

Bases: pipelinex.extras.ops.pytorch_ops.ModuleListMerge

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class pipelinex.extras.ops.pytorch_ops.ModuleConcatSkip(*modules)

Bases: pipelinex.extras.ops.pytorch_ops.ModuleConcat

__init__(*modules)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

class pipelinex.extras.ops.pytorch_ops.ModuleConvWrap(batchnorm=None, activation=None, *args, **kwargs)

Bases: torch.nn.modules.container.Sequential

__init__(batchnorm=None, activation=None, *args, **kwargs)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

core = None

class pipelinex.extras.ops.pytorch_ops.ModuleListMerge(*args: torch.nn.modules.module.Module)

class pipelinex.extras.ops.pytorch_ops.ModuleListMerge(arg: OrderedDict[str, Module])

Bases: torch.nn.modules.container.Sequential

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class pipelinex.extras.ops.pytorch_ops.ModuleProd(*args: torch.nn.modules.module.Module)

class pipelinex.extras.ops.pytorch_ops.ModuleProd(arg: OrderedDict[str, Module])

Bases: pipelinex.extras.ops.pytorch_ops.ModuleListMerge

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class pipelinex.extras.ops.pytorch_ops.ModuleSum(*args: torch.nn.modules.module.Module)

class pipelinex.extras.ops.pytorch_ops.ModuleSum(arg: OrderedDict[str, Module])

Bases: pipelinex.extras.ops.pytorch_ops.ModuleListMerge

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class pipelinex.extras.ops.pytorch_ops.ModuleSumSkip(*modules)

Bases: pipelinex.extras.ops.pytorch_ops.ModuleSum

__init__(*modules)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

class pipelinex.extras.ops.pytorch_ops.NLLoss(weight=None, size_average=None, ignore_index=- 100, reduce=None, reduction='mean')

Bases: torch.nn.modules.loss.NLLLoss

The negative likelihood loss. To compute Cross Entropy Loss, there are 3 options. NLLoss with torch.nn.Softmax torch.nn.NLLLoss with torch.nn.LogSoftmax torch.nn.CrossEntropyLoss

forward(input, target)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

ignore_index: int

class pipelinex.extras.ops.pytorch_ops.Pool1dMixIn(keepdim=False)

Bases: object

__init__(keepdim=False)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pytorch_ops.Pool2dMixIn(keepdim=False)

Bases: object

__init__(keepdim=False)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pytorch_ops.Pool3dMixIn(keepdim=False)

Bases: object

__init__(keepdim=False)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.pytorch_ops.StatModule(dim, keepdim=False)

Bases: torch.nn.modules.module.Module

__init__(dim, keepdim=False)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

training: bool

class pipelinex.extras.ops.pytorch_ops.StepBinary(size, desc=False, compare=None, dtype=None)

Bases: torch.nn.modules.module.Module

__init__(size, desc=False, compare=None, dtype=None)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorAvgPool1d(batchnorm=None, activation=None, *args, **kwargs)

Bases: pipelinex.extras.ops.pytorch_ops.ModuleConvWrap

core

alias of torch.nn.modules.pooling.AvgPool1d

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorAvgPool2d(batchnorm=None, activation=None, *args, **kwargs)

Bases: pipelinex.extras.ops.pytorch_ops.ModuleConvWrap

core

alias of torch.nn.modules.pooling.AvgPool2d

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorAvgPool3d(batchnorm=None, activation=None, *args, **kwargs)

Bases: pipelinex.extras.ops.pytorch_ops.ModuleConvWrap

core

alias of torch.nn.modules.pooling.AvgPool3d

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorClamp(min=None, max=None)

Bases: torch.nn.modules.module.Module

__init__(min=None, max=None)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorClampMax(max=None)

Bases: torch.nn.modules.module.Module

__init__(max=None)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorClampMin(min=None)

Bases: torch.nn.modules.module.Module

__init__(min=None)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorConstantLinear(weight=1, bias=0)

Bases: torch.nn.modules.module.Module

__init__(weight=1, bias=0)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorConv1d(batchnorm=None, activation=None, *args, **kwargs)

Bases: pipelinex.extras.ops.pytorch_ops.ModuleConvWrap

core

alias of torch.nn.modules.conv.Conv1d

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorConv2d(batchnorm=None, activation=None, *args, **kwargs)

Bases: pipelinex.extras.ops.pytorch_ops.ModuleConvWrap

core

alias of torch.nn.modules.conv.Conv2d

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorConv3d(batchnorm=None, activation=None, *args, **kwargs)

Bases: pipelinex.extras.ops.pytorch_ops.ModuleConvWrap

core

alias of torch.nn.modules.conv.Conv3d

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorCumsum(dim=1)

Bases: torch.nn.modules.module.Module

__init__(dim=1)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorExp(*args, **kwargs)

Bases: torch.nn.modules.module.Module

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorFlatten(*args, **kwargs)

Bases: torch.nn.modules.module.Module

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorForward(func=None)

Bases: torch.nn.modules.module.Module

__init__(func=None)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorGlobalAvgPool1d(keepdim=False)

Bases: pipelinex.extras.ops.pytorch_ops.Pool1dMixIn, pipelinex.extras.ops.pytorch_ops.TensorMean

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorGlobalAvgPool2d(keepdim=False)

Bases: pipelinex.extras.ops.pytorch_ops.Pool2dMixIn, pipelinex.extras.ops.pytorch_ops.TensorMean

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorGlobalAvgPool3d(keepdim=False)

Bases: pipelinex.extras.ops.pytorch_ops.Pool3dMixIn, pipelinex.extras.ops.pytorch_ops.TensorMean

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorGlobalMaxPool1d(keepdim=False)

Bases: pipelinex.extras.ops.pytorch_ops.Pool1dMixIn, pipelinex.extras.ops.pytorch_ops.TensorMax

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorGlobalMaxPool2d(keepdim=False)

Bases: pipelinex.extras.ops.pytorch_ops.Pool2dMixIn, pipelinex.extras.ops.pytorch_ops.TensorMax

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorGlobalMaxPool3d(keepdim=False)

Bases: pipelinex.extras.ops.pytorch_ops.Pool3dMixIn, pipelinex.extras.ops.pytorch_ops.TensorMax

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorGlobalMinPool1d(keepdim=False)

Bases: pipelinex.extras.ops.pytorch_ops.Pool1dMixIn, pipelinex.extras.ops.pytorch_ops.TensorMin

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorGlobalMinPool2d(keepdim=False)

Bases: pipelinex.extras.ops.pytorch_ops.Pool2dMixIn, pipelinex.extras.ops.pytorch_ops.TensorMin

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorGlobalMinPool3d(keepdim=False)

Bases: pipelinex.extras.ops.pytorch_ops.Pool3dMixIn, pipelinex.extras.ops.pytorch_ops.TensorMin

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorGlobalRangePool1d(keepdim=False)

Bases: pipelinex.extras.ops.pytorch_ops.Pool1dMixIn, pipelinex.extras.ops.pytorch_ops.TensorRange

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorGlobalRangePool2d(keepdim=False)

Bases: pipelinex.extras.ops.pytorch_ops.Pool2dMixIn, pipelinex.extras.ops.pytorch_ops.TensorRange

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorGlobalRangePool3d(keepdim=False)

Bases: pipelinex.extras.ops.pytorch_ops.Pool3dMixIn, pipelinex.extras.ops.pytorch_ops.TensorRange

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorGlobalSumPool1d(keepdim=False)

Bases: pipelinex.extras.ops.pytorch_ops.Pool1dMixIn, pipelinex.extras.ops.pytorch_ops.TensorSum

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorGlobalSumPool2d(keepdim=False)

Bases: pipelinex.extras.ops.pytorch_ops.Pool2dMixIn, pipelinex.extras.ops.pytorch_ops.TensorSum

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorGlobalSumPool3d(keepdim=False)

Bases: pipelinex.extras.ops.pytorch_ops.Pool3dMixIn, pipelinex.extras.ops.pytorch_ops.TensorSum

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorIdentity(*args, **kwargs)

Bases: torch.nn.modules.module.Module

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorLog(*args, **kwargs)

Bases: torch.nn.modules.module.Module

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorMax(dim, keepdim=False)

Bases: pipelinex.extras.ops.pytorch_ops.StatModule, torch.nn.modules.module.Module

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorMaxPool1d(batchnorm=None, activation=None, *args, **kwargs)

Bases: pipelinex.extras.ops.pytorch_ops.ModuleConvWrap

core

alias of torch.nn.modules.pooling.MaxPool1d

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorMaxPool2d(batchnorm=None, activation=None, *args, **kwargs)

Bases: pipelinex.extras.ops.pytorch_ops.ModuleConvWrap

core

alias of torch.nn.modules.pooling.MaxPool2d

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorMaxPool3d(batchnorm=None, activation=None, *args, **kwargs)

Bases: pipelinex.extras.ops.pytorch_ops.ModuleConvWrap

core

alias of torch.nn.modules.pooling.MaxPool3d

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorMean(dim, keepdim=False)

Bases: pipelinex.extras.ops.pytorch_ops.StatModule

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorMin(dim, keepdim=False)

Bases: pipelinex.extras.ops.pytorch_ops.StatModule, torch.nn.modules.module.Module

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorNearestPad(lower=1, upper=1)

Bases: torch.nn.modules.module.Module

__init__(lower=1, upper=1)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorProba(dim=1)

Bases: torch.nn.modules.module.Module

__init__(dim=1)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorRange(dim, keepdim=False)

Bases: pipelinex.extras.ops.pytorch_ops.StatModule, torch.nn.modules.module.Module

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorSkip(*args, **kwargs)

Bases: torch.nn.modules.module.Module

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorSlice(start=0, end=None, step=1)

Bases: torch.nn.modules.module.Module

__init__(start=0, end=None, step=1)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorSqueeze(dim=None)

Bases: torch.nn.modules.module.Module

__init__(dim=None)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorSum(dim, keepdim=False)

Bases: pipelinex.extras.ops.pytorch_ops.StatModule

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class pipelinex.extras.ops.pytorch_ops.TensorUnsqueeze(dim)

Bases: torch.nn.modules.module.Module

__init__(dim)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(input)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

pipelinex.extras.ops.pytorch_ops.as_tuple(x)

pipelinex.extras.ops.pytorch_ops.element_wise_average(tt_list)

pipelinex.extras.ops.pytorch_ops.element_wise_prod(tt_list)

pipelinex.extras.ops.pytorch_ops.element_wise_sum(tt_list)

pipelinex.extras.ops.pytorch_ops.nl_loss(input, *args, **kwargs)

pipelinex.extras.ops.pytorch_ops.setup_conv_params(kernel_size=1, dilation=None, padding=None, stride=None, raise_error=False, *args, **kwargs)

pipelinex.extras.ops.pytorch_ops.step_binary(input, output_size, compare=<built-in method ge of type object>)

pipelinex.extras.ops.pytorch_ops.tensor_max(input, dim, keepdim=False)

pipelinex.extras.ops.pytorch_ops.tensor_min(input, dim, keepdim=False)

pipelinex.extras.ops.pytorch_ops.to_array(input)

pipelinex.extras.ops.pytorch_ops.to_channel_first_tensor(a)

pipelinex.extras.ops.pytorch_ops.to_channel_last_tensor(a)

pipelinex.extras.ops.shap_ops module

class pipelinex.extras.ops.shap_ops.ExplainModel(**kwargs)

Bases: object

__init__(**kwargs)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.shap_ops.Scale(**kwargs)

Bases: object

__init__(**kwargs)

Initialize self. See help(type(self)) for accurate signature.

pipelinex.extras.ops.skimage_ops module

class pipelinex.extras.ops.skimage_ops.SkimageMarkBoundaries(**kwargs)

Bases: pipelinex.extras.ops.skimage_ops.SkimageSegmentationDictToDict

fn = 'mark_boundaries'

class pipelinex.extras.ops.skimage_ops.SkimageSegmentationDictToDict(**kwargs)

Bases: pipelinex.utils.DictToDict

module = <module 'skimage.segmentation' from '/home/docs/checkouts/readthedocs.org/user_builds/pipelinex/envs/latest/lib/python3.8/site-packages/skimage/segmentation/__init__.py'>

class pipelinex.extras.ops.skimage_ops.SkimageSegmentationFelzenszwalb(**kwargs)

Bases: pipelinex.extras.ops.skimage_ops.SkimageSegmentationDictToDict

fn = 'felzenszwalb'

class pipelinex.extras.ops.skimage_ops.SkimageSegmentationQuickshift(**kwargs)

Bases: pipelinex.extras.ops.skimage_ops.SkimageSegmentationDictToDict

fn = 'quickshift'

class pipelinex.extras.ops.skimage_ops.SkimageSegmentationSlic(**kwargs)

Bases: pipelinex.extras.ops.skimage_ops.SkimageSegmentationDictToDict

fn = 'slic'

class pipelinex.extras.ops.skimage_ops.SkimageSegmentationWatershed(**kwargs)

Bases: pipelinex.extras.ops.skimage_ops.SkimageSegmentationDictToDict

fn = 'watershed'

pipelinex.extras.ops.sklearn_ops module

class pipelinex.extras.ops.sklearn_ops.DfBaseTransformer(cols=None, target_col=None, **kwargs)

Bases: pipelinex.extras.ops.sklearn_ops.ZeroToZeroTransformer

__init__(cols=None, target_col=None, **kwargs)

Initialize self. See help(type(self)) for accurate signature.

fit(df)

fit_transform(df)

Fit to data, then transform it.

Fits transformer to X and y with optional parameters fit_params and returns a transformed version of X.

Parameters:

• X (array-like of shape (n_samples, n_features)) – Input samples.

• y (array-like of shape (n_samples,) or (n_samples, n_outputs), default=None) – Target values (None for unsupervised transformations).

• **fit_params (dict) – Additional fit parameters.

Returns:

X_new – Transformed array.

Return type:

ndarray array of shape (n_samples, n_features_new)

set_fit_request(*, df: Union[bool, None, str] = '$UNCHANGED$') → pipelinex.extras.ops.sklearn_ops.DfBaseTransformer

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

New in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

df (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for df parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_transform_request(*, df: Union[bool, None, str] = '$UNCHANGED$') → pipelinex.extras.ops.sklearn_ops.DfBaseTransformer

Request metadata passed to the transform method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to transform if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to transform.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

New in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

df (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for df parameter in transform.

Returns:

self – The updated object.

Return type:

object

transform(df)

class pipelinex.extras.ops.sklearn_ops.DfMinMaxScaler(cols=None, target_col=None, **kwargs)

Bases: pipelinex.extras.ops.sklearn_ops.DfBaseTransformer, sklearn.preprocessing._data.MinMaxScaler

set_fit_request(*, df: Union[bool, None, str] = '$UNCHANGED$') → pipelinex.extras.ops.sklearn_ops.DfMinMaxScaler

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

New in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

df (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for df parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_transform_request(*, df: Union[bool, None, str] = '$UNCHANGED$') → pipelinex.extras.ops.sklearn_ops.DfMinMaxScaler

Request metadata passed to the transform method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to transform if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to transform.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

New in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

df (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for df parameter in transform.

Returns:

self – The updated object.

Return type:

object

class pipelinex.extras.ops.sklearn_ops.DfQuantileTransformer(cols=None, target_col=None, **kwargs)

Bases: pipelinex.extras.ops.sklearn_ops.DfBaseTransformer, sklearn.preprocessing._data.QuantileTransformer

set_fit_request(*, df: Union[bool, None, str] = '$UNCHANGED$') → pipelinex.extras.ops.sklearn_ops.DfQuantileTransformer

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

New in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

df (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for df parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_transform_request(*, df: Union[bool, None, str] = '$UNCHANGED$') → pipelinex.extras.ops.sklearn_ops.DfQuantileTransformer

Request metadata passed to the transform method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to transform if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to transform.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

New in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

df (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for df parameter in transform.

Returns:

self – The updated object.

Return type:

object

class pipelinex.extras.ops.sklearn_ops.DfStandardScaler(cols=None, target_col=None, **kwargs)

Bases: pipelinex.extras.ops.sklearn_ops.DfBaseTransformer, sklearn.preprocessing._data.StandardScaler

set_fit_request(*, df: Union[bool, None, str] = '$UNCHANGED$') → pipelinex.extras.ops.sklearn_ops.DfStandardScaler

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

New in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

df (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for df parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_inverse_transform_request(*, copy: Union[bool, None, str] = '$UNCHANGED$') → pipelinex.extras.ops.sklearn_ops.DfStandardScaler

Request metadata passed to the inverse_transform method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to inverse_transform if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to inverse_transform.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

New in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

copy (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for copy parameter in inverse_transform.

Returns:

self – The updated object.

Return type:

object

set_partial_fit_request(*, sample_weight: Union[bool, None, str] = '$UNCHANGED$') → pipelinex.extras.ops.sklearn_ops.DfStandardScaler

Request metadata passed to the partial_fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to partial_fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to partial_fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

New in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

sample_weight (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_weight parameter in partial_fit.

Returns:

self – The updated object.

Return type:

object

set_transform_request(*, df: Union[bool, None, str] = '$UNCHANGED$') → pipelinex.extras.ops.sklearn_ops.DfStandardScaler

Request metadata passed to the transform method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to transform if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to transform.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

New in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

df (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for df parameter in transform.

Returns:

self – The updated object.

Return type:

object

class pipelinex.extras.ops.sklearn_ops.DfTrainTestSplit(**kwargs)

Bases: object

__init__(**kwargs)

Initialize self. See help(type(self)) for accurate signature.

class pipelinex.extras.ops.sklearn_ops.EstimatorTransformer

Bases: sklearn.base.TransformerMixin, sklearn.base.BaseEstimator

class pipelinex.extras.ops.sklearn_ops.ZeroToZeroTransformer(zero_to_zero=False, **kwargs)

Bases: pipelinex.extras.ops.sklearn_ops.EstimatorTransformer

__init__(zero_to_zero=False, **kwargs)

Initialize self. See help(type(self)) for accurate signature.

fit_transform(X)

Fit to data, then transform it.

Fits transformer to X and y with optional parameters fit_params and returns a transformed version of X.

Parameters:

• X (array-like of shape (n_samples, n_features)) – Input samples.

• y (array-like of shape (n_samples,) or (n_samples, n_outputs), default=None) – Target values (None for unsupervised transformations).

• **fit_params (dict) – Additional fit parameters.

Returns:

X_new – Transformed array.

Return type:

ndarray array of shape (n_samples, n_features_new)

transform(X)

pipelinex.extras.ops.sklearn_ops.extract_from_df(df, cols, target_col)