kh% d dl mZ d dlmZmZmZmZ d dlZd dlmZm Z ddl
m Z ddl m
Z
ddlmZ d d
lmZmZmZ d d lmZ d d lmZmZmZ g d
Z G d dej6 Z G d dej6 Z ddedZ G d de Z e ede j@ f ddddee d e!d ed e fd Z"y) )partial)AnyCallableListOptionalN)nnTensor )Conv2dNormActivation)ImageClassification)_log_api_usage_once )register_modelWeights WeightsEnum)_IMAGENET_CATEGORIES)_make_divisible_ovewrite_named_paramhandle_legacy_interface) MobileNetV2MobileNet_V2_Weights mobilenet_v2c p e Zd Z d dededededeedej f ddf fd
Zd
e de fd Z
xZ S )
InvertedResidualNinpoupstride expand_ratio
norm_layer.returnc 4 t | || _ |dvrt d| |t j
}t
t ||z }| j dk( xr ||k( | _ g }|dk7 r-|j t ||d|t j |j t |||||t j t j ||dddd || g t j | | _ || _ |dkD | _ y ) N)r r
z#stride should be 1 or 2 instead of r kernel_sizer activation_layer)r groupsr r$ r F)bias)super__init__r
ValueErrorr BatchNorm2dintrounduse_res_connectappendr ReLU6extendConv2d
Sequentialconv out_channels_is_cn) selfr r r r r
hidden_dimlayers __class__s Z/var/www/teggl/fontify/venv/lib/python3.12/site-packages/torchvision/models/mobilenetv2.pyr( zInvertedResidual.__init__ s B6( KL L Js\ 123
#{{a / >C3J "$ 1 MM$S*!PZmomumuv
% !% )%'XX
*c1a?3
MM6* qj xc d | j r|| j | z S | j | S N)r- r3 r6 r< s r: forwardzInvertedResidual.forward< s, tyy|# #99Q< r; r> ) __name__
__module__ __qualname__r+ r r r Moduler( r r@
__classcell__r9 s @r: r r sg sw&!&! &!*-&!=@&!NVW_`cegenen`nWoNp&!
&!P F r; r c e Zd Z ddededeeee dedeedej f deedej f d ed
df fd
Z
d e d
e fd
Z d e d
e fdZ
xZS )r N num_classes
width_multinverted_residual_setting
round_nearestblock.r dropoutr c r t | t | |t }|t j
}d}d} |g dg dg dg dg dg d g d
g}t
| d k( st
|d d k7 rt d
| t ||z | }t | t d| z | | _
t d|d|t j g}
|D ]M \ } } }
}t | |z | }t |
D ]) }|d k( r|nd}|
j ||||| | |}+ O |
j t || j d|t j t j |
| _ t j t j" | t j$ | j | | _ | j) D ]l }t+ |t j, rbt j. j1 |j2 d |j4 Vt j. j7 |j4 t+ |t j
t j8 f rSt j. j; |j2 t j. j7 |j4 t+ |t j$ st j. j= |j2 d d t j. j7 |j4 o y)aw
MobileNet V2 main class
Args:
num_classes (int): Number of classes
width_mult (float): Width multiplier - adjusts number of channels in each layer by this amount
inverted_residual_setting: Network structure
round_nearest (int): Round the number of channels in each layer to be a multiple of this number
Set to 1 to turn off rounding
block: Module specifying inverted residual building block for mobilenet
norm_layer: Module specifying the normalization layer to use
dropout (float): The droupout probability
N i )r r r ) r
r
)rQ rO r
)rQ @ r
)rQ ` rS r )rQ rS r
)rQ i@ r r r rU zGinverted_residual_setting should be non-empty or a 4-element list, got ?rS r
)r r r$ r )r r r" )pfan_out)modeg{Gz?) r' r( r
r r r* lenr) r max last_channelr r/ ranger. r2 featuresDropoutLinear
classifiermodules
isinstancer1 initkaiming_normal_weightr& zeros_ GroupNormones_normal_)r6 rH rI rJ rK rL r rM
input_channelr^ r` tcnsoutput_channelir mr9 s r: r( zMobileNetV2.__init__D s 0 D! =$E J
$ , ) % ( )Q .#6OPQ6R2SWX2X YZsYtu
(
(BMR
+L3sJ;O,OQ^ _ M!
egemem n%
4 /JAq!Q ,Q^] KN1X
/ 1f!m^VZ[hr st .
/ / t 0 0aJikiqiq
x0
-- JJ ! IId'' 5
'A!RYY'''y'A66%GGNN166*A =>
ahh'qvv&Aryy)!T2qvv&
'r; r< c | j | }t j j |d }t j
|d }| j
| }|S )Nr r r )r` r
functionaladaptive_avg_pool2dtorchflattenrc r? s r:
_forward_implzMobileNetV2._forward_impl sK
MM! MM - -a 8 MM!Q OOA r; c $ | j | S r> )r{ r? s r: r@ zMobileNetV2.forward s !!!$$r; )i rX N NNg?)rA rB rC r+ floatr r r r rD r( r r{ r@ rE rF s @r: r r C s ?C 489= ]']' ]' $,DcO#< ]'
]' bii 01
]' Xc299n 56]' ]'
]'~v & % %F %r; r iz5 rv )
num_paramsmin_size
categoriesc e Zd Z ed eed i edddddid d
d d
Z ed eedd i edddddid ddd
ZeZ y)r z=https://download.pytorch.org/models/mobilenet_v2-b0353104.pth ) crop_sizezQhttps://github.com/pytorch/vision/tree/main/references/classification#mobilenetv2z ImageNet-1Kgx&1Q@gMV@)zacc@1zacc@5g$C?g\( +@zXThese weights reproduce closely the results of the paper using a simple training recipe.)recipe_metrics_ops
_file_size_docs)url
transformsmetaz=https://download.pytorch.org/models/mobilenet_v2-7ebf99e0.pth )r resize_sizezHhttps://github.com/pytorch/vision/issues/3995#new-recipe-with-reg-tuningg`" R@gS㥛V@gV-2+@a$
These weights improve upon the results of the original paper by using a modified version of TorchVision's
`new training recipe
`_.
N)
rA rB rC r r r _COMMON_META
IMAGENET1K_V1
IMAGENET1K_V2DEFAULT r; r: r r s K.#>
i # # s
M" K.#3O
` # #
M* Gr; r
pretrained)weightsT)r progressr r kwargsr c t j | } | #t |dt | j d t di |}| "|j
| j |d |S )a MobileNetV2 architecture from the `MobileNetV2: Inverted Residuals and Linear
Bottlenecks `_ paper.
Args:
weights (:class:`~torchvision.models.MobileNet_V2_Weights`, optional): The
pretrained weights to use. See
:class:`~torchvision.models.MobileNet_V2_Weights` below for
more details, and possible values. By default, no pre-trained
weights are used.
progress (bool, optional): If True, displays a progress bar of the
download to stderr. Default is True.
**kwargs: parameters passed to the ``torchvision.models.mobilenetv2.MobileNetV2``
base class. Please refer to the `source code
`_
for more details about this class.
.. autoclass:: torchvision.models.MobileNet_V2_Weights
:members:
rH r T)r
check_hashr )r verifyr r\ r r load_state_dictget_state_dict)r r r models r: r r sk 0 #))'2G fmS l9S5TU !& !E
g 4 4hSW 4 XY Lr; )# functoolsr typingr r r r ry r r ops.miscr transforms._presetsr utilsr
_apir r r _metar _utilsr r r __all__rD r r r r r boolr r r; r: r s 0 0 + 5 ' 6 6 ' S S B- ryy - `k%")) k%^ & ' ; ' T ,0D0R0R!ST15 -. AE X[ U r;