{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Type abbreviations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The code and docs sometimes use *type abbreviations* to avoid type signatures getting unwieldy. Here's a list of all abbreviations for composite types for convenient access."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## From [`core`](/core.html#core)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- `AnnealFunc` = `Callable`\\[\\[`Number`,`Number`,`float`], `Number`]\n",
    "- `ArgStar` = `Collection`\\[`Any`]\n",
    "- `BatchSamples` = `Collection`\\[`Tuple`\\[`Collection`\\[`int`], `int`]]\n",
    "- `Classes` = `Collection`\\[`Any`]\n",
    "- `DataFrameOrChunks` = `Union[DataFrame, pd.io.parsers.TextFileReader]`\n",
    "- `FilePathList` = `Collection`\\[`Path`]\n",
    "- `Floats` = `Union`\\[`float`, `Collection`\\[`float`]]\n",
    "- `ImgLabels` = `Collection`\\[`ImgLabel`]\n",
    "- `KeyFunc` = `Callable`\\[\\[`int`], `int`]\n",
    "- `KWArgs` = `Dict`\\[`str`,`Any`]\n",
    "- `ListOrItem` = `Union`\\[`Collection`\\[`Any`],`int`,`float`,`str`]\n",
    "- `ListRules` = `Collection`\\[`Callable`\\[\\[`str`],`str`]]\n",
    "- `ListSizes` = `Collection`\\[`Tuple`\\[`int`,`int`]]\n",
    "- `NPArrayableList` = `Collection`\\[`Union`\\[`np`.`ndarray`, `list`]]\n",
    "- `NPArrayList` = `Collection`\\[`np`.`ndarray`]\n",
    "- `OptDataFrame` = `Optional`\\[`DataFrame`]\n",
    "- `OptListOrItem` = `Optional`\\[`ListOrItem`]\n",
    "- `OptRange` = `Optional`\\[`Tuple`\\[`float`,`float`]]\n",
    "- `OptStrTuple` = `Optional`\\[`Tuple`\\[`str`,`str`]]\n",
    "- `OptStats` = `Optional`\\[`Tuple`\\[`np`.`ndarray`, `np`.`ndarray`]]\n",
    "- `PathOrStr` = `Union`\\[`Path`,`str`]\n",
    "- `PBar` = `Union`\\[`MasterBar`, `ProgressBar`]\n",
    "- `Point`=`Tuple`\\[`float`,`float`]\n",
    "- `Points`=`Collection`\\[`Point`]\n",
    "- `Sizes` = `List`\\[`List`\\[`int`]]\n",
    "- `SplitArrayList` = `List`\\[`Tuple`\\[`np`.`ndarray`,`np`.`ndarray`]]\n",
    "- `StartOptEnd`=`Union`\\[`float`,`Tuple`\\[`float`,`float`]]\n",
    "- `StrList` = `Collection`\\[`str`]\n",
    "- `Tokens` = `Collection`\\[`Collection`\\[`str`]]\n",
    "- `OptStrList` = `Optional`\\[`StrList`]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## From [`torch_core`](/torch_core.html#torch_core)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- `BoolOrTensor` = `Union`\\[`bool`,`Tensor`]\n",
    "- `FloatOrTensor` = `Union`\\[`float`,`Tensor`]\n",
    "- `IntOrTensor` = `Union`\\[`int`,`Tensor`]\n",
    "- `ItemsList` = `Collection`\\[`Union`\\[`Tensor`,[`ItemBase`](/core.html#ItemBase),'`ItemsList`',`float`,`int`]]\n",
    "- `LambdaFunc` = `Callable`\\[\\[`Tensor`],`Tensor`]\n",
    "- `LayerFunc` = `Callable`\\[ [[`nn.Module`](https://pytorch.org/docs/stable/nn.html#torch.nn.Module)],`None`]\n",
    "- [`Model`](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) = [`nn`](https://pytorch.org/docs/stable/nn.html#torch-nn).[`Module`](/torch_core.html#Module)\n",
    "- `ModuleList` = `Collection`\\[[`nn.Module`](https://pytorch.org/docs/stable/nn.html#torch.nn.Module)]\n",
    "- `OptOptimizer` = `Optional`\\[[`optim.Optimizer`](https://pytorch.org/docs/stable/optim.html#torch.optim.Optimizer)]\n",
    "- `ParamList` = `Collection`\\[[`nn.Parameter`](https://pytorch.org/docs/stable/nn.html#torch.nn.Parameter)]\n",
    "- `Rank0Tensor` = `NewType`('`OneEltTensor`', `Tensor`)\n",
    "- `SplitFunc` = `Callable`\\[[`Model`](https://pytorch.org/docs/stable/nn.html#torch.nn.Module)], `List`[`Model`](https://pytorch.org/docs/stable/nn.html#torch.nn.Module)]]\n",
    "- `SplitFuncOrIdxList` = `Union`\\[`Callable`, `Collection`\\[`ModuleList`]]\n",
    "- `TensorOrNumber` = `Union`\\[`Tensor`,`Number`]\n",
    "- `TensorOrNumList` = `Collection`\\[`TensorOrNumber`]\n",
    "- `TensorImageSize` = `Tuple`\\[`int`,`int`,`int`]\n",
    "- `Tensors` = `Union`\\[`Tensor`, `Collection`\\['`Tensors`']]\n",
    "- `Weights` = `Dict`\\[`str`,`Tensor`]\n",
    "- `AffineFunc` = `Callable`\\[\\[`KWArgs`], [`AffineMatrix`](https://pytorch.org/docs/stable/tensors.html#torch-tensor)]\n",
    "- `HookFunc` = `Callable`\\[[`Model`](https://pytorch.org/docs/stable/nn.html#torch.nn.Module), `Tensors`, `Tensors`], `Any`]\n",
    "- `LogitTensorImage` = `TensorImage`\n",
    "- `LossFunction` = `Callable`\\[\\[`Tensor`, `Tensor`], `Rank0Tensor`]\n",
    "- `MetricFunc` = `Callable`\\[\\[`Tensor`,`Tensor`],`TensorOrNumber`]\n",
    "- `MetricFuncList` = `Collection`\\[`MetricFunc`]\n",
    "- `MetricsList` = `Collection`\\[`TensorOrNumber`]\n",
    "- `OptLossFunc` = `Optional`\\[`LossFunction`]\n",
    "- `OptMetrics` = `Optional`\\[`MetricsList`]\n",
    "- `OptSplitFunc` = `Optional`\\[`SplitFunc`]\n",
    "- `PixelFunc` = `Callable`\\[\\[`TensorImage`, `ArgStar`, `KWArgs`], `TensorImage`]\n",
    "- `CoordFunc` = `Callable`\\[[`FlowField`](/vision.image.html#FlowField), `TensorImageSize`, `ArgStar`, `KWArgs`], `LogitTensorImage`]\n",
    "- `LightingFunc` = `Callable`\\[\\[`LogitTensorImage`, `ArgStar`, `KWArgs`], `LogitTensorImage`]"
   ]
  }
 ],
 "metadata": {
  "jekyll": {
   "keywords": "fastai",
   "summary": "Type annotations names",
   "title": "fastai_typing"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}