Name	Size	Mode	Actions
common_types.cpython-36.pyc	990	0644	editdlrm
cpp.cpython-36.pyc	3460	0644	editdlrm
functional.cpython-36.pyc	161648	0644	editdlrm
grad.cpython-36.pyc	11840	0644	editdlrm
init.cpython-36.pyc	17551	0644	editdlrm
parameter.cpython-36.pyc	7568	0644	editdlrm
_reduction.cpython-36.pyc	1244	0644	editdlrm
__init__.cpython-36.pyc	2002	0644	editdlrm

3 �Eg�@srdd
lZddl
mZ
ddlmZ
Gdd�dej�ZGdd�d�Zdd �ZGd d�dee�Z Gdd �d eej�Z d
S)�N)
�_disabled_torch_function_impl)
�OrderedDictcs>eZ
dZd
Zddd�
Zdd�Z�f
dd �Zd d�ZeZ �Z S) � ParameteraJA kind of Tensor that is to be considered a module parameter. Parameters are :class:`~torch.Tensor` subclasses, that have a very special property when used with :class:`Module` s - when they're assigned as Module attributes they are automatically added to the list of its parameters, and will appear e.g. in :meth:`~Module.parameters` iterator. Assigning a Tensor doesn't have such effect. This is because one might want to cache some temporary state, like last hidden state of the RNN, in the model. If there was no such class as :class:`Parameter`, these temporaries would get registered too. Args: data (Tensor): parameter tensor. requires_grad (bool, optional): if the parameter requires gradient. See :ref:`locally-disable-grad-doc` for more details. Default: `True` NTcCs"|
dkrtj
g�
}
tjj||
|�S)
N)�torch�tensor�Tensor�_make_subclass)�cls�data� requires_grad�r�>/usr/local/lib64/python3.6/site-packages/torch/nn/parameter.py�__new__s


zParameter.__new__cCsHt|�
|
kr|
t|�
St
|�
|jjtjd
�
|j�}||
t|�
<|SdS)N)
Z memory_format)�id�typer �clonerZpreserve_formatr)�self�memo�resultrrr �__deepcopy__s



zParameter.__deepcopy__c

sd
tt
|�j�S)NzParameter containing: )�superr�__repr__)
r)
� __class__rr r$s
zParameter.__repr__cCstj
j|j|jt�ffS)
N)r�_utilsZ_rebuild_parameterr rr)r�protorrr � __reduce_ex__'s
zParameter.__reduce_ex__)NT)�__name__� __module__�__qualname__�__doc__rrrrr�__torch_function__� __classcell__rr)
rr rs
rcs�eZ
dZejjejjejjejjejj ejj ejjejjejj ejjejjejjejjejjejjejgZddd�
Zedd��
Zdd�Zdd �Zd d�Zefd
f�f
dd � �
Z�ZS)�UninitializedTensorMixinNcCs>|d
kr|jj
}|d
kr |jj}tj|
||d�|_|j|_d
S)a~Create a Parameter or Tensor with the same properties of the uninitialized one. Given a shape, it materializes a parameter in the same device and with the same `dtype` as the current one or the specified ones in the arguments. Args: shape : (tuple): the shape for the materialized tensor. device (:class:`torch.device`): the desired device of the parameters and buffers in this module. Optional. dtype (:class:`torch.dtype`): the desired floating point type of the floating point parameters and buffers in this module. Optional. N)�device�dtype)r r#r$r�empty� cls_to_becomer)r�shaper#r$rrr �materializeEs 




z$UninitializedTensorMixin.materializec

Cstd
�
�
dS)Na
Can't access the shape of an uninitialized parameter or buffer. This error usually happens in `load_state_dict` when trying to load an uninitialized parameter into an initialized one. Call `forward` to initialize the parameters before accessing their attributes.)
�RuntimeError)
rrrr r'Ys
zUninitializedTensorMixin.shapec

Cstd
�
�
dS)Nz�Can't share memory on an uninitialized parameter or buffer. Call `forward` to initialize the parameters before calling `module.share_memory()`.)
r))
rrrr � share_memory_as

z&UninitializedTensorMixin.share_memory_c

Csd
|jj
�d�S)N�
<�
>)rr)
rrrr rgs
z!UninitializedTensorMixin.__repr__cCs|j|j
f
fS)
N)rr)rrrrr rjs
z&UninitializedTensorMixin.__reduce_ex__csJ|
|jks|
j
jd
kr4|dkr"i}t�j|
|||�Stdj|
|j��
�
dS)Nzmethod-wrappera
Attempted to use an uninitialized parameter in {}. This error happens when you are using a `LazyModule` or explicitly manipulating `torch.nn.parameter.{}` objects. When using LazyModules Call `forward` with a dummy batch to initialize the parameters before calling torch functions)�_allowed_methodsrrrr � ValueError�format)r �func�types�args�kwargs)
rrr r qs




z+UninitializedTensorMixin.__torch_function__)NN)rrrrr�__hash__�sizeZcopy_Zis_floating_pointZhalf�float�double�char�short�int�longZcuda�cpuZtoZ get_deviceZ!_has_compatible_shallow_copy_typer-r(�propertyr'r*rr�classmethodr r!rr)
rr r"1s.














 
r"c

Cs t|t
�S)
N)� isinstancer")
�paramrrr �is_lazy�s
rAc@s$eZ
dZd
ZeZddd�
dd�ZdS)�UninitializedParametera�A parameter that is not initialized. Unitialized Parameters are a a special case of :class:`torch.nn.Parameter` where the shape of the data is still unknown. Unlike a :class:`torch.nn.Parameter`, uninitialized parameters hold no data and attempting to access some properties, like their shape, will throw a runtime error. The only operations that can be performed on a uninitialized parameter are changing its datatype, moving it to a different device and converting it to a regular :class:`torch.nn.Parameter`. The default device or dtype to use when the parameter is materialized can be set during construction using e.g. ``device='cuda'``. TN)
�returncCs(||d
�}tj
gf
|�
}tjj|||
�S)N)r#r$)rrrr)r rr#r$�factory_kwargsr rrr r�s


zUninitializedParameter.__new__)TNN)rrrrrr&rrrrr rB�srBc@s&eZ
dZd
ZejZddd�
dd�ZdS)�UninitializedBuffera�A buffer that is not initialized. Unitialized Buffer is a a special case of :class:`torch.Tensor` where the shape of the data is still unknown. Unlike a :class:`torch.Tensor`, uninitialized parameters hold no data and attempting to access some properties, like their shape, will throw a runtime error. The only operations that can be performed on a uninitialized parameter are changing its datatype, moving it to a different device and converting it to a regular :class:`torch.Tensor`. The default device or dtype to use when the buffer is materialized can be set during construction using e.g. ``device='cuda'``. FN)
rCcCs(||d
�}tj
gf
|�
}tjj|||
�S)N)r#r$)rrrr)r rr#r$rDr rrr r�s


zUninitializedBuffer.__new__)FNN)rrrrrrr&rrrrr rE�srE)rZtorch._Cr�collectionsrrrr"rArBrErrrr �
s

+P