jK) d dl Z d dlZd dlmZ d dlmZ d dlmZmZm Z m
Z
m Z m Z m
Z
d dlmZ ddlmZmZmZ ddlmZ G d d
e ZdS ) N)
namedtuple)Path)AnyCallableDictListOptionalTupleUnion)Image )extract_archiveiterable_to_strverify_str_arg)
VisionDatasetc e Zd ZdZ edg d Zg eddddddd d
ed d ddddd d
ed
dddddd d
eddddddd d
eddddddd d
eddddddd d eddddddd d eddddd ddd ed d d dd ddd ed d!ddd dd d" ed#d$ddd dd d% ed&d'dd(dddd) ed*d+dd(dddd, ed-d.dd(dddd/ ed0d1dd(ddd d2 ed3d4dd(ddd d5 ed6d7dd(ddd d8 ed9d:dd;dddd< ed=d>dd;ddd d< ed?d@dd;ddddA edBdCdd;ddddD edEdFd dGddddH edIdJd!dGddddK edLdMd$dLddddN edOdPd'dQdd ddR edSdTd+dQdd ddU edVdWd.dXdd ddY edZd[d1dXdd dd\ ed]d^d4dXdd dd_ ed`daddXdd d db edcddddXdd d de edfdgd7dXdd ddh edidjd:dXdd ddk edldmd>dXdd ddn edodpdpdXddd dY Z ddteee f duedvedwee
e ef dxe e dye e dze e d{dsf fd|
Z
d}ed{eeef fd~Zd{efdZd{efdZded{eeef fdZdvedwed{efdZ xZS )
Cityscapesaw `Cityscapes `_ Dataset.
Args:
root (str or ``pathlib.Path``): Root directory of dataset where directory ``leftImg8bit``
and ``gtFine`` or ``gtCoarse`` are located.
split (string, optional): The image split to use, ``train``, ``test`` or ``val`` if mode="fine"
otherwise ``train``, ``train_extra`` or ``val``
mode (string, optional): The quality mode to use, ``fine`` or ``coarse``
target_type (string or list, optional): Type of target to use, ``instance``, ``semantic``, ``polygon``
or ``color``. Can also be a list to output a tuple with all specified target types.
transform (callable, optional): A function/transform that takes in a PIL image
and returns a transformed version. E.g, ``transforms.RandomCrop``
target_transform (callable, optional): A function/transform that takes in the
target and transforms it.
transforms (callable, optional): A function/transform that takes input sample and its target as entry
and returns a transformed version.
Examples:
Get semantic segmentation target
.. code-block:: python
dataset = Cityscapes('./data/cityscapes', split='train', mode='fine',
target_type='semantic')
img, smnt = dataset[0]
Get multiple targets
.. code-block:: python
dataset = Cityscapes('./data/cityscapes', split='train', mode='fine',
target_type=['instance', 'color', 'polygon'])
img, (inst, col, poly) = dataset[0]
Validate on the "coarse" set
.. code-block:: python
dataset = Cityscapes('./data/cityscapes', split='val', mode='coarse',
target_type='semantic')
img, smnt = dataset[0]
CityscapesClass)nameidtrain_idcategory category_id
has_instancesignore_in_evalcolor unlabeledr voidFT)r r r z ego vehicler
zrectification border z
out of roi static dynamic )o J r ground )Q r r* road flat) @ r. sidewalk ) # parking ) z
rail track
) building construction)F rA rA wall )f rD fence
) rI z
guard rail ) rK bridge )r< d rO tunnel )r< x Z pole object)rI rI rI polegroup z
traffic light )r7 r8 z traffic sign ) r\ r
vegetation nature)k r3 terrain ) rd sky )rA rK person human)r\ r[ < rider )r r r car vehicle)r r ra truck )r r rA bus )r rl rO caravan )r r rS trailerrZ )r r n train )r P rO
motorcycle )r r r; bicycle! )w r? r~ z
license platefineinstanceNrootsplitmode target_type transformtarget_transform
transformsreturnc
t |||| |dk rdnd| _ t j | j d| | _ t j | j | j | | _ || _ || _
g | _ g | _ t |dd |dk rd}nd}d } | ||t | } t |d
|| t! |t" s|g| _ d | j D t j | j r%t j | j s<|d k r&t j | j d
}
n%t j | j d }
| j dk r.t j | j | j d } n8| j dk r-t j | j | j d } t j |
rLt j | r-t) |
| j t) | | j nt+ d t j | j D ]=} t j | j | }
t j | j | }t j |
D ]}g }| j D ]}}d |
d d | | j | }| t j || ~| j t j |
| | j | ܐ?d S )Nr gtFinegtCoarse leftImg8bitr )r coarse)rz testval)rz train_extrar zOUnknown value '{}' for argument split if mode is '{}'. Valid values are {{{}}}.r c 0 g | ]}t |d d S )r )r semanticpolygonr )r ).0values PC:\Projects\Historical\venv\Lib\site-packages\torchvision/datasets/cityscapes.py
z'Cityscapes.__init__.. s5
5-1] ^ ^
r zleftImg8bit_trainextra.zipz leftImg8bit_trainvaltest.zipz_trainvaltest.zipz.zip) from_pathto_pathzDataset not found or incomplete. Please make sure all required folders for the specified "split" and "mode" are inside the "root" directoryz{}_{} _leftImg8bitr )super__init__r ospathjoinr
images_dir targets_dirr r imagestargetsr formatr
isinstancelistisdirisfiler RuntimeErrorlistdir_get_target_suffixappend)selfr r r r r r r valid_modesmsg
image_dir_ziptarget_dir_zipcityimg_dir
target_dir file_name target_typest target_name __class__s r r zCityscapes.__init__i s z96FGGG $HHJ ',,ty-GG 7<< 49eDD&
tV%7888 6>>2KK9K_jjok&B&BCCug{C888+t,, - +}D
)
w}}T_-- RW]]4CS5T5T
%% " TY8T U U
" TY8V W W
yH$$!# diDI9X9X9X!Y!Yj((!# diDI9K9K9K!L!Lw~~m,,
1O1O
-KKKK .$)LLLLL"T
Jt// 2 2Dgll4?D99G d&6==J Z 0 0
2
2 ! ) O OA")..!77:DRR$ 1c61eT?SS$ b#vq%WW$
B>1eULYY$ A~q%XX$ Q5% YY $ b#~q%__ $ "c>1eT?[[!$" "c>1eT>ZZ#$$ AxE5/RR%$& Rh5$XX'$( Q!UE>ZZ)$* AxE5-XX+$, b!Xq%WW-$. 2q(Aue_UU/$0 r2ua~NN1$2 "b'1dE=QQ3$4 R!T5+NN5$6 r2y!T5+NN7$8 RAtUJOO9$: r2y!T5,OO;$< 2sIq$jQQ=$> 2sIq$kRR?$@ RAtULQQA$B b"iD%UUC$D 2r9aumTTE$F RAudKXXG$GR -7(,/3)-D2 D2CI D2 D2 D2
49c>* D2 H%
D2 #8,D2 X&D2
D2 D2 D2 D2 D2 D2L sCx 6 8C 8 8 8 8s tCH~
+s + + + + + + + + + +r r )r r collectionsr pathlibr typingr r r r r r
r PILr utilsr r r visionr r r r r r s " " " " " " D D D D D D D D D D D D D D D D D D C C C C C C C C C C ! ! ! ! ! !Q+ Q+ Q+ Q+ Q+ Q+ Q+ Q+ Q+ Q+r