How to use set_doc method in autotest

Best Python code snippet using autotest_python

array.py

Source:array.py

1import chainerx2from chainerx import _docs3def set_docs():4    ndarray = chainerx.ndarray5    _docs.set_doc(6        ndarray,7        """ndarray(shape, dtype, device=None)8Multi-dimensional array, the central data structure of ChainerX.9This class, along with other APIs in the :mod:`chainerx` module, provides a10subset of NumPy APIs. This class works similar to :class:`numpy.ndarray`,11except for some differences including the following noticeable points:12- :class:`chainerx.ndarray` has a :attr:`device` attribute. It indicates on13  which device the array is allocated.14- :class:`chainerx.ndarray` supports :ref:`Define-by-Run <define_by_run>`15  backpropagation. Once you call :meth:`require_grad`, the array starts16  recording the operations applied to it recursively. Gradient of the result17  with respect to the original array can be computed then with the18  :meth:`backward` method or the :func:`chainerx.backward` function.19Args:20    shape (tuple of ints): Shape of the new array.21    dtype: Data type.22    device (~chainerx.Device): Device on which the array is allocated.23        If omitted, :ref:`the default device <chainerx_device>` is chosen.24.. seealso:: :class:`numpy.ndarray`25""")26    _docs.set_doc(27        ndarray.data_ptr,28        """int: Address of the underlying memory allocation.29The meaning of the address is device-dependent.30""")31    _docs.set_doc(32        ndarray.data_size,33        'int: Total size of the underlying memory allocation.')34    _docs.set_doc(35        ndarray.device, '~chainerx.Device: Device on which the data exists.')36    _docs.set_doc(ndarray.dtype, 'Data type of the array.')37    # TODO(beam2d): Write about backprop id.38    _docs.set_doc(39        ndarray.grad,40        """~chainerx.ndarray: Gradient held by the array.41It is ``None`` if the gradient is not available.42Setter of this property overwrites the gradient.43""")44    _docs.set_doc(45        ndarray.is_contiguous,46        'bool: ``True`` iff the array is stored in the C-contiguous order.')47    _docs.set_doc(ndarray.itemsize, 'int: Size of each element in bytes.')48    _docs.set_doc(49        ndarray.nbytes,50        """int: Total size of all elements in bytes.51It does not count skips between elements.""")52    _docs.set_doc(ndarray.ndim, 'int: Number of dimensions.')53    _docs.set_doc(54        ndarray.offset,55        'int: Offset of the first element from the memory allocation in bytes.'56    )57    _docs.set_doc(58        ndarray.shape,59        """tuple of int: Lengths of axes.60.. note::61    Currently, this property does not support setter.""")62    _docs.set_doc(ndarray.size, 'int: Number of elements in the array.')63    _docs.set_doc(ndarray.strides, 'tuple of int: Strides of axes in bytes.')64    _docs.set_doc(65        ndarray.T,66        """~chainerx.ndarray: Shape-reversed view of the array.67New array is created at every access to this property.68``x.T`` is just a shorthand of ``x.transpose()``.69""")70    _docs.set_doc(71        ndarray.__getitem__,72        """___getitem__(self, key)73Returns self[key].74.. note::75    Currently, only basic indexing is supported not advanced indexing.76""")77    def unary_op(name, s):78        _docs.set_doc(getattr(ndarray, name), '{}()\n{}'.format(name, s))79    unary_op('__bool__', 'Casts a size-one array into a :class:`bool` value.')80    unary_op('__float__',81             'Casts a size-one array into a :class:`float` value.')82    unary_op('__int__', 'Casts a size-one array into :class:`int` value.')83    unary_op('__len__', 'Returns the length of the first axis.')84    unary_op('__neg__', 'Computes ``-x`` elementwise.')85    def binary_op(name, s):86        _docs.set_doc(getattr(ndarray, name), '{}(other)\n{}'.format(name, s))87    binary_op('__eq__', 'Computes ``x == y`` elementwise.')88    binary_op('__ne__', 'Computes ``x != y`` elementwise.')89    binary_op('__lt__', 'Computes ``x < y`` elementwise.')90    binary_op('__le__', 'Computes ``x <= y`` elementwise.')91    binary_op('__ge__', 'Computes ``x >= y`` elementwise.')92    binary_op('__gt__', 'Computes ``x > y`` elementwise.')93    binary_op('__iadd__', 'Computes ``x += y`` elementwise.')94    binary_op('__isub__', 'Computes ``x -= y`` elementwise.')95    binary_op('__imul__', 'Computes ``x *= y`` elementwise.')96    binary_op('__itruediv__', 'Computes ``x /= y`` elementwise.')97    binary_op('__iand__', 'Computes ``x &= y`` elementwise.')98    binary_op('__ior__', 'Computes ``x |= y`` elementwise.')99    binary_op('__ixor__', 'Computes ``x ^= y`` elementwise.')100    binary_op('__add__', 'Computes ``x + y`` elementwise.')101    binary_op('__sub__', 'Computes ``x - y`` elementwise.')102    binary_op('__mul__', 'Computes ``x * y`` elementwise.')103    binary_op('__truediv__', 'Computes ``x / y`` elementwise.')104    binary_op('__and__', 'Computes ``x & y`` elementwise.')105    binary_op('__or__', 'Computes ``x | y`` elementwise.')106    binary_op('__xor__', 'Computes ``x ^ y`` elementwise.')107    binary_op('__radd__', 'Computes ``y + x`` elementwise.')108    binary_op('__rsub__', 'Computes ``y - x`` elementwise.')109    binary_op('__rmul__', 'Computes ``y * x`` elementwise.')110    binary_op('__rand__', 'Computes ``y & x`` elementwise.')111    binary_op('__ror__', 'Computes ``y | x`` elementwise.')112    binary_op('__rxor__', 'Computes ``y ^ x`` elementwise.')113    # TODO(beam2d): Write about as_grad_stopped(backprop_ids, copy) overload.114    _docs.set_doc(115        ndarray.as_grad_stopped,116        """as_grad_stopped(copy=False)117Creates a view or a copy of the array that stops gradient propagation.118This method behaves similar to :meth:`view` and :meth:`copy`, except that119the gradient is not propagated through this operation (internally, this120method creates a copy or view of the array without connecting the computational121graph for backprop).122Args:123    copy (bool): If ``True``, it copies the array. Otherwise, it returns a view124        of the original array.125Returns:126    ~chainerx.ndarray:127        A view or a copy of the array without propagating the  gradient on128        backprop.129""")130    _docs.set_doc(131        ndarray.argmax,132        """argmax(axis=None)133Returns the indices of the maximum elements along a given axis.134See :func:`chainerx.argmax` for the full documentation.135""")136    _docs.set_doc(137        ndarray.argmin,138        """argmin(axis=None)139Returns the indices of the minimum elements along a given axis.140See :func:`chainerx.argmin` for the full documentation.141""")142    _docs.set_doc(143        ndarray.astype,144        """astype(dtype, copy=True)145Casts each element to the specified data type.146Args:147    dtype: Data type of the new array.148    copy (bool): If ``True``, this method always copies the data. Otherwise,149        it creates a view of the array if possible.150Returns:151    ~chainerx.ndarray: An array with the specified dtype.152""")153    _docs.set_doc(154        ndarray.backward,155        """backward(backprop_id=None, enable_double_backprop=False)156Performs backpropagation starting from this array.157This method is equivalent to ``chainerx.backward([self], *args)``.158See :func:`chainerx.backward` for the full documentation.159""")160    # TODO(beam2d): Write about backprop id.161    _docs.set_doc(162        ndarray.cleargrad,163        """cleargrad()164Clears the gradient held by this array.165""")166    _docs.set_doc(167        ndarray.copy,168        """copy()169Creates an array and copies all the elements to it.170The copied array is allocated on the same device as ``self``.171.. seealso:: :func:`chainerx.copy`172""")173    _docs.set_doc(174        ndarray.dot,175        """dot(b)176Returns the dot product with a given array.177See :func:`chainerx.dot` for the full documentation.178""")179    _docs.set_doc(180        ndarray.fill,181        """fill(value)182Fills the array with a scalar value in place.183Args:184    value: Scalar value with which the array will be filled.185""")186    # TODO(beam2d): Write about backprop_id argument.187    _docs.set_doc(188        ndarray.get_grad,189        """get_grad()190Returns the gradient held by the array.191If the gradient is not available, it returns ``None``.192""")193    # TODO(beam2d): Write about backprop_id argument.194    _docs.set_doc(195        ndarray.is_backprop_required,196        """is_backprop_required()197Returns ``True`` if gradient propagates through this array on backprop.198See the note on :meth:`require_grad` for details.199""")200    # TODO(beam2d): Write about backprop_id argument.201    _docs.set_doc(202        ndarray.is_grad_required,203        """is_grad_required()204Returns ``True`` if the gradient will be set after backprop.205See the note on :meth:`require_grad` for details.206""")207    _docs.set_doc(208        ndarray.item,209        """item()210Copies an element of an array to a standard Python scalar and returns it.211Returns:212    z:213        A copy of the specified element of the array as a suitable Python214        scalar.215.. seealso:: :func:`numpy.item`216""")217    _docs.set_doc(218        ndarray.max,219        """max(axis=None, keepdims=False)220Returns the maximum along a given axis.221See :func:`chainerx.amax` for the full documentation.222""")223    _docs.set_doc(224        ndarray.min,225        """min(axis=None, keepdims=False)226Returns the minimum along a given axis.227See :func:`chainerx.amin` for the full documentation.228""")229    # TODO(beam2d): Write about backprop_id argument.230    _docs.set_doc(231        ndarray.require_grad,232        """require_grad()233Declares that a gradient for this array will be made available after backprop.234Once calling this method, any operations applied to this array are recorded for235later backprop. After backprop, the :attr:`grad` attribute holds the gradient236array.237.. note::238    ChainerX distinguishes *gradient requirements* and *backprop requirements*239    strictly. They are strongly related, but different concepts as follows.240    - *Gradient requirement* indicates that the gradient array should be made241      available after backprop. This attribute **is not propagated** through242      any operations. It implicates the backprop requirement.243    - *Backprop requirement* indicates that the gradient should be propagated244      through the array during backprop. This attribute **is propagated**245      through differentiable operations.246    :meth:`require_grad` sets the gradient requirement flag. If you need to247    extract the gradient after backprop, you have to call :meth:`require_grad`248    on the array even if the array is an intermediate result of differentiable249    computations.250Returns:251    ~chainerx.ndarray: ``self``252""")253    _docs.set_doc(254        ndarray.reshape,255        """reshape(newshape)256Creates an array with a new shape and the same data.257See :func:`chainerx.reshape` for the full documentation.258""")259    _docs.set_doc(260        ndarray.set_grad,261        """set_grad(grad)262Sets a gradient to the array.263This method overwrites the gradient with a given array.264Args:265    grad (~chainerx.ndarray): New gradient array.266""")267    _docs.set_doc(268        ndarray.squeeze,269        """squeeze(axis=None)270Removes size-one axes from an array.271See :func:`chainerx.squeeze` for the full documentation.272""")273    _docs.set_doc(274        ndarray.swapaxes,275        """swapaxes(axis1, axis2)276Interchange two axes of an array..277See :func:`chainerx.swapaxes` for the full documentation.278""")279    _docs.set_doc(280        ndarray.repeat,281        """repeat(repeats, axis=None)282Constructs an array by repeating a given array.283See :func:`chainerx.repeats` for the full documentation.284""")285    _docs.set_doc(286        ndarray.sum,287        """sum(axis=None, keepdims=False)288Returns the sum of an array along given axes.289See :func:`chainerx.sum` for the full documentation.290""")291    _docs.set_doc(292        ndarray.take,293        """take(indices, axis)294Takes elements from the array along an axis.295See :func:`chainerx.take` for the full documentation.296""")297    _docs.set_doc(298        ndarray.to_device,299        """to_device(device, index=None)300Transfers the array to the specified device.301Args:302    device (~chainerx.Device or str): Device to which the array is transferred,303        or a backend name. If it is a backend name, ``index`` should also be304        specified.305    index (int): Index of the device for the backend specified by ``device``.306Returns:307    ~chainerx.ndarray:308        An array on the target device.309        If the original array is already on the device, it is a view of that.310        Otherwise, it is a copy of the array on the target device.311""")312    _docs.set_doc(313        ndarray.transpose,314        """transpose(axes=None)315Creates a view of an array with permutated axes.316See :func:`chainerx.transpose` for the full documentation.317""")318    _docs.set_doc(319        ndarray.view,320        """view()321Returns a view of the array.322The returned array shares the underlying buffer, though it has a different323identity as a Python object....

device.py

Source:device.py

1import chainerx2from chainerx import _docs3def _set_docs_device():4    Device = chainerx.Device5    _docs.set_doc(6        Device,7        """Represents a physical computing unit.8""")9    _docs.set_doc(10        Device.synchronize,11        """Synchronizes the device.12""")13    _docs.set_doc(14        Device.name,15        """Device name.16It is the backend name and the device index concatenated with a colon, e.g.17``native:0``.18Returns:19    str: Device name.20""")21    _docs.set_doc(22        Device.backend,23        """Backend to which this device belongs.24Returns:25    ~chainerx.Backend: Backend object.26""")27    _docs.set_doc(28        Device.context,29        """Context to which this device belongs.30Returns:31    ~chainerx.Context: Context object.32""")33    _docs.set_doc(34        Device.index,35        """Index of this device.36Returns:37    int: Index of this device.38""")39def set_docs():40    _set_docs_device()41    _docs.set_doc(42        chainerx.get_device,43        """get_device(*device)44Returns a device specified by the arguments.45If the argument is a single :class:`~chainerx.Device` instance, it's simply46returned.47Otherwise, there are three ways to specify a device:48.. testcode::49    # Specify a backend name and a device index separately.50    chainerx.get_device('native', 0)51    # Specify a backend name and a device index in a single string.52    chainerx.get_device('native:0')53    # Specify only a backend name. In this case device index 0 is chosen.54    chainerx.get_device('native')55Returns:56    ~chainerx.Device: Device object.57""")58    _docs.set_doc(59        chainerx.get_default_device,60        """get_default_device()61Returns the default device associated with the current thread.62Returns:63    ~chainerx.Device: The default device.64.. seealso::65    * :func:`chainerx.set_default_device`66    * :func:`chainerx.using_device`67""")68    _docs.set_doc(69        chainerx.set_default_device,70        """set_default_device(device)71Sets the given device as the default device of the current thread.72Args:73    device (~chainerx.Device or str): Device object or device name to set as74        the default device.75.. seealso::76    * :func:`chainerx.get_default_device`77    * :func:`chainerx.using_device`78""")79    _docs.set_doc(80        chainerx.using_device,81        """using_device(device)82Creates a context manager to temporarily set the default device.83Args:84    device (~chainerx.Device or str): Device object or device name to set as85        the default device during the context. See :data:`chainerx.Device.name`86        for the specification of device names.87.. seealso::88    * :func:`chainerx.get_default_device`89    * :func:`chainerx.set_default_device`...

Automation Testing Tutorials

Learn to execute automation testing from scratch with LambdaTest Learning Hub. Right from setting up the prerequisites to run your first automation test, to following best practices and diving deeper into advanced test scenarios. LambdaTest Learning Hubs compile a list of step-by-step guides to help you be proficient with different test automation frameworks i.e. Selenium, Cypress, TestNG etc.