+ j}.^RIHtHt^RIt^RIt^RIt^RIHtHt^RIH t ^RI H t ^RI H t ]PPRRl4tRR R llt!R R 4tR RltRRlt]P(P*RRl4t]P(P*RRl4t]P(P*RRl4t!RR]P24tR#))OptionalUnionN)nnTensor)box_area)_log_api_usage_once) roi_aligncRV^8dQhR\R\\,R\/#)levelsunmerged_resultsreturn)rlist)formats"X/home/luca/screen-checking/.venv/lib/python3.14/site-packages/torchvision/ops/poolers.py __annotate__rs%vf&c V^,pVPVPrC\P!VP ^4VP ^4VP ^4VP ^43W4R7p\ \ V44Fp\P!W8H4^,PR^^^4pVPVP ^4W,P ^4W,P ^4W,P ^44pVP^WqV,4pK V#)rdtypedevice) rrtorchzerossizerangelenwhereviewexpandscatter)r r first_resultrrreslevelindexs&& r_onnx_merge_levelsr&s #A&L && (;(;6 ++ Q**1-|/@/@/C\EVEVWXEYZbg Cs+,- FO,Q/44RAqA JJqM  # ( ( +  # ( ( +  # ( ( +   kk!UU$;<. Jrc TV^8dQhR\R\R\R\R\/#r k_mink_maxcanonical_scalecanonical_levelepsintfloat)rs"rrr%sALL L LL L  Lrc\WW#V4#N) LevelMapper)r)r*r+r,r-s&&&&&rinitLevelMapperr4%s u_s KKrcLa]tRt^/toRtRV3RlRlltV3RlRltRtVtR#) r3zDetermine which FPN level each RoI in a set of RoIs should map to based on the heuristic in the FPN paper. Args: k_min (int) k_max (int) canonical_scale (int) canonical_level (int) eps (float) c 8<V^8dQhRS[RS[RS[RS[RS[/#r(r.)r __classdict__s"rrLevelMapper.__annotate__;s=         rcBWnW nW0nW@nWPnR#r2)r)r*s0lvl0r-)selfr)r*r+r,r-s&&&&&&r__init__LevelMapper.__init__;s  !# rc6<V^8dQhRS[S[,RS[/#)r boxlistsrrr)rr7s"rrr8Is# J Jf J& Jrc  |\P!\P!VUu.uFp\V4NK up44p\P!VP \P !W0P, 4,\P!VPVPR7,4p\P!W@PVPR7pVP\P4VP, P\P4#uupi)z$ Args: boxlists (list[BoxList]) r)minmax)rsqrtcatrfloorr;log2r:tensorr-rclampr)r*toint64)r<r@boxlists target_lvlss&& r__call__LevelMapper.__call__Is JJuyy8!L8(7"38!LM Nkk$))ejjWW.E"E UYU]U]efelelHm"mn kk+::4::N u{{+djj8<) __name__ __module__ __qualname____firstlineno____doc__r=rQ__static_attributes____classdictcell__)r7s@rr3r3/s     J Jrr3cFV^8dQhR\\,R\/#)r boxesrrA)rs"rrrWs$v,6rcf\P!V^R7pVPVPr2\P!\ V4UUu.uF4wrE\P !VR,WC\P VR7NK6 upp^R7p\P!Wa.^R7pV#uuppi)r)dim)rlayoutr):NNN:NN)rrGrr enumerate full_likestrided)r^ concat_boxesrribidsroiss& r_convert_to_roi_formatrkWs99U*L ''););E ))dmnsdtudt\`\]51%--PV Wdtu  C 99c(a 0D K vs:B- cRV^8dQhR\R\\,R\/#)r feature original_sizer)rrr/r0)rs"rrrbs%&curc:VPRRp.p\W!4FswrE\V4\V4, p^\\P!V4P 4P 44,pVPV4Ku V^,#)r N)shapezipr0rrJrIroundappend)rmrnrpossible_scaless1s2 approx_scalescales&& r _infer_scalerzbs == D#%Od*Ry59, U5<< 5::<BBDEEu%+ 1 rc V^8dQhR\\,R\\\\3,,R\R\R\\\,\ 3,/#)r features image_shapesr+r,r)rrtupler/r0r3)rs"rrrnsS6l*.uS#X*?RUhk 4; #$rcZV'g \R4h^p^pVF)p\V^,V4p\V^,V4pK+ WE3pVUu.uFp\W4NK p p\P!\P !V ^,\P R74P4)p \P!\P !V R,\P R74P4)p \\V 4\V 4VVR7p W3#uupi)zimages list should not be emptyrC)r+r,r) ValueErrorrErzrrIrJfloat32itemr4r/) r|r}r+r,max_xmax_yrqoriginal_input_shapefeatscaleslvl_minlvl_max map_levelss &&&& r _setup_scalesrms :;; E EE!He$E!He$">CK L84l468F Lzz%,,vay FGLLNNGzz%,,vbzGHMMOOG  G  G '' J  MsD(cV^8dQhR\\\3,R\\,R\\,/#)r x featmap_namesr)dictstrrr)rs"rrrs2T#v+&tCyT&\rcn.pVP4Fwr4W19gK VPV4K V#r2)itemsrt)rr x_filteredkvs&& r _filter_inputrs5J     a  rcV^8dQhR\\,R\\,R\\,R\R\\\,,R\\ ,R\/#)r rr^ output_sizesampling_ratiormapperr)rrr/rr0r3)rs"rrrsqPPV P <PcP P T%[ ! P [ ! P Prc BVeVf \R4h\V4p\V4pV^8Xd\V^,VVV^,VR7#V!V4p\V4p V^,P^,p V^,P V^,P r\P!V V 3V,V V R7p .p\\W44Fwpwpp\P!W8H4^,pVV,p\VVVVVR7p\P!4'd#VPVPV 44K~VPV P 4V V&K \P!4'd \!W4p V #)a Args: x_filtered (List[Tensor]): List of input tensors. boxes (List[Tensor[N, 4]]): boxes to be used to perform the pooling operation, in (x1, y1, x2, y2) format and in the image reference size, not the feature map reference. The coordinate must satisfy ``0 <= x1 < x2`` and ``0 <= y1 < y2``. output_size (Union[List[Tuple[int, int]], List[int]]): size of the output sampling_ratio (int): sampling ratio for ROIAlign scales (Optional[List[float]]): If None, scales will be automatically inferred. Default value is None. mapper (Optional[LevelMapper]): If none, mapper will be automatically inferred. Default value is None. Returns: result (Tensor) z$scales and mapper should not be None)r spatial_scalerr)rrrkr rqrrrrrcrrr torchvision _is_tracingrtrLr&)rr^rrrr num_levelsrjr num_rois num_channelsrrresulttracing_resultsr$per_level_featurery idx_in_levelrois_per_levelresult_idx_in_levels&&&&&& r_multiscale_roi_alignrs,~?@@ZJ !% (DQ qM # ))   E]F4yHa=&&q)LqM''A)=)=6 [[       FO-6s:7N-O))!5{{6?3A6 l+'  #)    " " $ $  " "#6#9#9%#@ A$7#9#9&,,#GF< -.P0  #F< Mrcaa]tRt^toRtR]]],,R]],/t R^R^/V3RlV3Rlllt V3RlR lt V3R lR lt R t VtV;t#) MultiScaleRoIAligna  Multi-scale RoIAlign pooling, which is useful for detection with or without FPN. It infers the scale of the pooling via the heuristics specified in eq. 1 of the `Feature Pyramid Network paper `_. They keyword-only parameters ``canonical_scale`` and ``canonical_level`` correspond respectively to ``224`` and ``k0=4`` in eq. 1, and have the following meaning: ``canonical_level`` is the target level of the pyramid from which to pool a region of interest with ``w x h = canonical_scale x canonical_scale``. Args: featmap_names (List[str]): the names of the feature maps that will be used for the pooling. output_size (List[Tuple[int, int]] or List[int]): output size for the pooled region sampling_ratio (int): sampling ratio for ROIAlign canonical_scale (int, optional): canonical_scale for LevelMapper canonical_level (int, optional): canonical_level for LevelMapper Examples:: >>> m = torchvision.ops.MultiScaleRoIAlign(['feat1', 'feat3'], 3, 2) >>> i = OrderedDict() >>> i['feat1'] = torch.rand(1, 5, 64, 64) >>> i['feat2'] = torch.rand(1, 5, 32, 32) # this feature won't be used in the pooling >>> i['feat3'] = torch.rand(1, 5, 16, 16) >>> # create some random bounding boxes >>> boxes = torch.rand(6, 4) * 256; boxes[:, 2:] += boxes[:, :2] >>> # original image size, before computing the feature maps >>> image_sizes = [(512, 512)] >>> output = m(i, [boxes], image_sizes) >>> print(output.shape) >>> torch.Size([6, 5, 3, 3]) rrr+r,c <V^8dQhRS[S[,RS[S[S[S[,S[S[,3,RS[RS[RS[/#)r rrrr+r,)rrrr/r~)rr7s"rrMultiScaleRoIAlign.__annotate__ sU//Cy/3c DI56/ /  //rc<\SV`4\V4\V\4'dW"3pWnW0n\V4VnRVn RVn W@n WPn R#r2) superr=r isinstancer/rrr~rrrr+r,)r<rrrr+r, __class__s&&&&$$rr=MultiScaleRoIAlign.__init__ s` D! k3 ' '&4K*, - ..rc ~<V^8dQhRS[S[S[3,RS[S[,RS[S[S[S[3,,RS[/#)r rr^r}r)rrrrr~r/)rr7s"rrr!sM   V   F|  5c?+    rc 0\WP4pVPeVPf/\ WCVP VP 4wVnVn\VVVPVPVPVP4#)a Args: x (OrderedDict[Tensor]): feature maps for each level. They are assumed to have all the same number of channels, but they can have different sizes. boxes (List[Tensor[N, 4]]): boxes to be used to perform the pooling operation, in (x1, y1, x2, y2) format and in the image reference size, not the feature map reference. The coordinate must satisfy ``0 <= x1 < x2`` and ``0 <= y1 < y2``. image_shapes (List[Tuple[height, width]]): the sizes of each image before they have been fed to a CNN to obtain feature maps. This allows us to infer the scale factor for each one of the levels to be pooled. Returns: result (Tensor) ) rrrrrr+r,rrr)r<rr^r}rs&&&& rforwardMultiScaleRoIAlign.forward!s&#1&8&89 ;; $//"9+8$*>*>@T@T, (DK%         KK OO   rc <V^8dQhRS[/#)r r)r)rr7s"rrrCs  # rcVPP RVP RVP RVP R2#)z(featmap_names=z, output_size=z, sampling_ratio=))rrVrrr)r<s&r__repr__MultiScaleRoIAlign.__repr__CsM~~&&'t7I7I6JK++,,=d>Q>Q=RRS U r)r,r+rrrrr)rVrWrXrYrZrrr0r3__annotations__r=rrr[r\ __classcell__)rr7s@@rrrsa!F $u+!6 h{F[\O/ # / !//*   D   rrrS)typingrrrtorch.fxrrrtorchvision.ops.boxesrutilsrr jitunusedr&r4r3rkrzfxwraprrrModulerrrrs" *' &L%J%JP6PPfa a r