+ jU5^RIHt^RIHt^RIt^RIHuHt^RIHt!RR]4t !RR]P4t R R lt !R R ]P4t!R R]P4t!RR]P4t!RR]P4t!RR]P4t!RR]P4t!RR]P4t!RR]P4t!RR]P4t!RR]P4t!RR ]P4t!R!R"]P4t!R#R$]P4tR#)%)Any)BaseSparsifierN)nncPaa]tRt^ toV3RlV3RlltV3RlRltRtVtV;t#)ImplementedSparsifierc:<V^8dQhRS[S[S[3,RR/#)kwargsreturnN)dictstrr)format __classdict__s"g/home/luca/screen-checking/.venv/lib/python3.14/site-packages/torch/testing/_internal/common_pruning.py __annotate__"ImplementedSparsifier.__annotate__ s#**c3h*D*c (<\SV`VR7R#))defaultsN)super__init__)selfr __class__s&,rrImplementedSparsifier.__init__ s &)rcZ<V^8dQhRS[PRS[RS[S[S[3,RR/#)r module tensor_namer r N)rModuler r r)rrs"rrr s=KK"))K#KcSVhK\`Krc ^VPP^,P^&VPR,pVP R^4^,VR&R#)rzlinear1.weight step_countN)parametrizationsweightmaskstateget)rrrr linear_states&&&, r update_mask!ImplementedSparsifier.update_mask sN45&&q)..q1zz"23 %1%5%5lA%F%J \"r) __name__ __module__ __qualname____firstlineno__rr'__static_attributes____classdictcell__ __classcell__rrs@@rrr s!**KKKrrc@a]tRt^toRt]V3RlRl4tRtVtR#)MockSparseLinearz This class is a MockSparseLinear class to check convert functionality. It is the same as a normal Linear layer, except with a different type, as well as an additional from_dense method. c8<V^8dQhRS[PRR/#)r modr r3)rLinear)rrs"rrMockSparseLinear.__annotate__s RYY+=rc @V!VPVP4pV#)z ) in_features out_features)clsr5linears&& r from_denseMockSparseLinear.from_denses"S__%%' rr)N) r*r+r,r-__doc__ classmethodr=r.r/)rs@rr3r3s rr3cdV^8dQhR\PR\PR\/#)r subset_tensorsuperset_tensorr )torchTensorbool)rs"rrr"s*  5<< %,, SW rc^pVFDpV\V48d1\P!W1V,4'g V^, pK>KCR# R#)zO Checks to see if all rows in subset tensor are present in the superset tensor FT)lenrDequal)rBrCirows&& rrows_are_subsetrL"sH A#o&&;;sA$677Q rcTaa]tRt^2toRtV3RlV3RlltV3RlRltRtVtV;t #) SimpleLinearzModel with only Linear layers without biases, some wrapped in a Sequential, some following the Sequential. Used to test basic pruned Linear-Linear fusion.c<V^8dQhRR/#r r Nr))rrs"rrSimpleLinear.__annotate__6s44$4rc `<\SV`4\P!\P!^^RR7\P!^^RR7\P!^^RR74Vn\P!^^RR7Vn\P!^^ RR7VnR#)FbiasN)rrr Sequentialr6seqlinear1linear2rrs&rrSimpleLinear.__init__6sx == IIa ' IIa ' IIa '  yyAE2 yyBU3 rcN<V^8dQhRS[PRS[P/#r xr rDrE)rrs"rrrQ@#%,,rclVPV4pVPV4pVPV4pV#N)rWrXrYrr^s&&rforwardSimpleLinear.forward@. HHQK LLO LLOr)rXrYrW r*r+r,r-r?rrdr.r/r0r1s@@rrNrN2s$V44rrNcTaa]tRt^GtoRtV3RlV3RlltV3RlRltRtVtV;t #) LinearBiaszModel with only Linear layers, alternating layers with biases, wrapped in a Sequential. Used to test pruned Linear-Bias-Linear fusion.c<V^8dQhRR/#rPr))rrs"rrLinearBias.__annotate__Ks  $ rc H<\SV`4\P!\P!^^RR7\P!^^RR7\P!^^RR7\P!^^RR7\P!^^ RR74VnR#rSTrTFN)rrrrVr6rWrZs&rrLinearBias.__init__Ksn == IIa & IIa ' IIa & IIa & IIa% (  rcN<V^8dQhRS[PRS[P/#r]r_)rrs"rrrkUs#%,,rc(VPV4pV#rbrWrcs&&rrdLinearBias.forwardUs HHQKrrqrgr1s@@rririGs$O  rricTaa]tRt^ZtoRtV3RlV3RlltV3RlRltRtVtV;t #)LinearActivationzModel with only Linear layers, some with bias, some in a Sequential and some following. Activation functions modules in between each Linear in the Sequential, and each outside layer. Used to test pruned Linear(Bias)-Activation-Linear fusion.c<V^8dQhRR/#rPr))rrs"rrLinearActivation.__annotate___  $ rc <\SV`4\P!\P!^^RR7\P !4\P!^^RR7\P !4\P!^^RR74Vn\P!^^RR7Vn\P !4Vn \P!^^ RR7Vn \P !4Vn R#rm) rrrrVr6ReLUTanhrWrXact1rYact2rZs&rrLinearActivation.__init___s == IIa & GGI IIa ' GGI IIa &  yyAD1 GGI yyBU3 GGI rcN<V^8dQhRS[PRS[P/#r]r_)rrs"rrrvm#%,,rcVPV4pVPV4pVPV4pVPV4pVP V4pV#rb)rWrXr{rYr|rcs&&rrdLinearActivation.forwardmH HHQK LLO IIaL LLO IIaLr)r{r|rXrYrWrgr1s@@rrtrtZs$B  rrtcTaa]tRt^vtoRtV3RlV3RlltV3RlRltRtVtV;t #)LinearActivationFunctionalaModel with only Linear layers, some with bias, some in a Sequential and some following. Activation functions modules in between each Linear in the Sequential, and functional activations are called in between each outside layer. Used to test pruned Linear(Bias)-Activation-Linear fusion.c<V^8dQhRR/#rPr))rrs"rr'LinearActivationFunctional.__annotate__|rwrc  <\SV`4\P!\P!^^RR7\P !4\P!^^RR7\P !4\P!^^RR74Vn\P!^^RR7Vn\P!^^RR7Vn\P!^^ RR7Vn \P !4Vn R#rm) rrrrVr6ryrWrXrYlinear3r{rZs&rr#LinearActivationFunctional.__init__|s == IIa & GGI IIa ' GGI IIa &  yyAD1 yyAE2 yyBU3 GGI rcN<V^8dQhRS[PRS[P/#r]r_)rrs"rrrs#%,,rcVPV4pVPV4p\P!V4pVP V4p\P!V4pVP V4p\P!V4pV#rb)rWrXFrelurYrrcs&&rrd"LinearActivationFunctional.forwardsb HHQK LLO FF1I LLO FF1I LLO FF1Ir)r{rXrYrrWrgr1s@@rrrvs$B  rrcTaa]tRt^toRtV3RlV3RlltV3RlRltRtVtV;t #) SimpleConv2dzModel with only Conv2d layers, all without bias, some in a Sequential and some following. Used to test pruned Conv2d-Conv2d fusion.c<V^8dQhRR/#rPr))rrs"rrSimpleConv2d.__annotate__s;;$;rc @<\SV`4\P!\P!^^ ^^RR7\P!^ ^@^^RR74Vn\P!^@^0^^RR7Vn\P!^0^4^^RR7VnR#)FrTNrrrrVConv2drWconv2d1conv2d2rZs&rrSimpleConv2d.__init__sx == IIaQ . IIb"a / yyRAE: yyRAE: rcN<V^8dQhRS[PRS[P/#r]r_)rrs"rrrr`rclVPV4pVPV4pVPV4pV#rbrWrrrcs&&rrdSimpleConv2d.forwardrfrrrrWrgr1s@@rrrs#1;;rrcTaa]tRt^toRtV3RlV3RlltV3RlRltRtVtV;t #) Conv2dBiaszModel with only Conv2d layers, some with bias, some in a Sequential and some outside. Used to test pruned Conv2d-Bias-Conv2d fusion.c<V^8dQhRR/#rPr))rrs"rrConv2dBias.__annotate__s;;$;rc t<\SV`4\P!\P!^^ ^^RR7\P!^ ^ ^^RR7\P!^ ^@^^RR74Vn\P!^@^0^^RR7Vn\P!^0^4^^RR7VnR#rTrTFNrrZs&rrConv2dBias.__init__s == IIaQ - IIb"a . IIb"a /  yyRAD9 yyRAE: rcN<V^8dQhRS[PRS[P/#r]r_)rrs"rrrr`rclVPV4pVPV4pVPV4pV#rbrrcs&&rrdConv2dBias.forwardrfrrrgr1s@@rrrs#6;;rrcTaa]tRt^toRtV3RlV3RlltV3RlRltRtVtV;t #)Conv2dActivationa Model with only Conv2d layers, some with bias, some in a Sequential and some following. Activation function modules in between each Sequential layer, functional activations called in-between each outside layer. Used to test pruned Conv2d-Bias-Activation-Conv2d fusion.c<V^8dQhRR/#rPr))rrs"rrConv2dActivation.__annotate__s : :$ :rc<\SV`4\P!\P!^^ ^^RR7\P !4\P!^ ^@^^RR7\P !4\P!^@^@^^RR7\P !44Vn\P!^@^0^^RR7Vn\P!^0^4^^RR7Vn R#r) rrrrVrryrzrWrrrZs&rrConv2dActivation.__init__s == IIaQ - GGI IIb"a . GGI IIb"a / GGI  yyRAE: yyRAD9 rcN<V^8dQhRS[PRS[P/#r]r_)rrs"rrrrrcVPV4pVPV4p\P!V4pVP V4p\P !V4pV#rb)rWrrrrhardtanhrcs&&rrdConv2dActivation.forwardsH HHQK LLO FF1I LLO JJqMrrrgr1s@@rrrs$A : :rrcTaa]tRt^toRtV3RlV3RlltV3RlRltRtVtV;t #) Conv2dPadBiasaEModel with only Conv2d layers, all with bias and some with padding > 0, some in a Sequential and some following. Activation function modules in between each layer. Used to test that bias is propagated correctly in the special case of pruned Conv2d-Bias-(Activation)Conv2d fusion, when the second Conv2d layer has padding > 0.c<V^8dQhRR/#rPr))rrs"rrConv2dPadBias.__annotate__s$rc<\SV`4\P!\P!^^ ^^^RR7\P !4\P!^ ^ ^^RR7\P !4\P!^ ^ ^^^RR7\P !4\P!^ ^ ^^^RR7\P !4\P!^ ^@^^RR7\P !44 Vn\P!^@^0^^^RR7Vn\P !4Vn \P!^0^4^^^RR7Vn \P !4Vn R#)rT)paddingrUFrTN) rrrrVrryrzrWrr{rr|rZs&rrConv2dPadBias.__init__s == IIaQ14 8 GGI IIb"a / GGI IIb"aAD 9 GGI IIb"aAD 9 GGI IIb"a . GGI  yyRAqtD GGI yyRAqtD GGI rcN<V^8dQhRS[PRS[P/#r]r_)rrs"rrrrrcVPV4pVPV4pVPV4pVPV4pVP V4pV#rb)rWrr{rr|rcs&&rrdConv2dPadBias.forwardrr)r{r|rrrWrgr1s@@rrrs$c &rrcTaa]tRt^toRtV3RlV3RlltV3RlRltRtVtV;t #) Conv2dPoolzModel with only Conv2d layers, all with bias, some in a Sequential and some following. Activation function modules in between each layer, Pool2d modules in between each layer. Used to test pruned Conv2d-Pool2d-Conv2d fusion.c<V^8dQhRR/#rPr))rrs"rrConv2dPool.__annotate__sNN$Nrc <\SV`4\P!\P!^^ ^^RR7\P !^^^R7\P !4\P!^ ^@^^RR7\P!4\P!^^^R74Vn \P!^@^0^^RR7Vn \P !^^^R7Vn \P !4Vn \P!^0^4^^RR7Vn \P!^4^4^^RR7VnR#)rT kernel_sizerrUrstriderN)rrrrVr MaxPool2dryrz AvgPool2drWrmaxpoolaf1rconv2d3rZs&rrConv2dPool.__init__s == IIaAD A LLQq! < GGI IIb"!QT B GGI LLQq! <  yyRQM ||!QG 779yyRQM yyRQM rcN<V^8dQhRS[PRS[P/#r]r_)rrs"rrr#   %,, rc2VPV4pVPV4pVPV4pVPV4pVP V4p\ P !V^^^R7p\ P!V4pVPV4pV#r r) rWrrrrr avg_pool2drrrcs&&rrdConv2dPool.forwardsv HHQK LLO LLO HHQK LLO LL!Q ? FF1I LLOr)rrrrrrWrgr1s@@rrrs%8NN   rrcTaa]tRtRtoRtV3RlV3RlltV3RlRltRtVtV;t #)Conv2dPoolFlattenFunctionaliaModel with Conv2d layers, all with bias, some in a Sequential and some following, and then a Pool2d and a functional Flatten followed by a Linear layer. Activation functions and Pool2ds in between each layer also. Used to test pruned Conv2d-Pool2d-Flatten-Linear fusion.c<V^8dQhRR/#rPr))rrs"rr(Conv2dPoolFlattenFunctional.__annotate__#s//$/rc <\SV`4\P!\P!^^^^RR7\P !^^^R7\P !4\P!^^^^RR7\P!4\P!^^^R74Vn \P!^^^^RR7Vn \P !4Vn \P!^^ ^^RR7Vn \P!R4Vn\P!^ ^ RR7VnR#)rTrrrTN)rr)rrrrVrrryrzrrWrrrAdaptiveAvgPool2davg_poolr6fcrZs&rr$Conv2dPoolFlattenFunctional.__init__#s == IIa14 @ LLQq! < GGI IIa14 @ GGI LLQq! <  yyA1adK 779yyBAqtL ,,V4 ))B.rcN<V^8dQhRS[PRS[P/#r]r_)rrs"rrr3rrc4VPV4pVPV4p\P!V^^^R7pVP V4pVP V4pVP V4p\P!V^4pVPV4pV#r) rWrr max_pool2drrrrDflattenrrcs&&rrd#Conv2dPoolFlattenFunctional.forward3sz HHQK LLO LL!Q ? HHQK LLO MM!  MM!Q  GGAJr)rrrrrrWrgr1s@@rrrs$@ //   rrcTaa]tRtRtoRtV3RlV3RlltV3RlRltRtVtV;t #)Conv2dPoolFlatteni?a Model with Conv2d layers, all with bias, some in a Sequential and some following, and then a Pool2d and a Flatten module followed by a Linear layer. Activation functions and Pool2ds in between each layer also. Used to test pruned Conv2d-Pool2d-Flatten-Linear fusion.c<V^8dQhRR/#rPr))rrs"rrConv2dPoolFlatten.__annotate__Es//$/rc <\SV`4\P!\P!^^^^RR7\P !^^^R7\P !4\P!^^^^RR7\P!4\P!^^^R74Vn \P!^^^^RR7Vn \P !4Vn \P!^^ ^^RR7Vn \P!R4Vn\P!4Vn\P"!^,^ RR7VnR#)rTrrrTN)r r )rrrrVrrryrzrrWrrrrrFlattenrr6rrZs&rrConv2dPoolFlatten.__init__Es == IIa14 @ LLQq! < GGI IIa14 @ GGI LLQq! <  yyA1adK 779yyBAqtL ,,V4 zz| ))B.rcN<V^8dQhRS[PRS[P/#r]r_)rrs"rrrVrrc(VPV4pVPV4p\P!V^^^R7pVP V4pVP V4pVP V4pVPV4pVPV4pV#r) rWrrrrrrrrrcs&&rrdConv2dPoolFlatten.forwardVsw HHQK LLO LL!Q ? HHQK LLO MM!  LLO GGAJr)rrrrrrrWrgr1s@@rrr?s$@ //"   rrcTaa]tRtRtoRtV3RlV3RlltV3RlRltRtVtV;t #)LSTMLinearModelibzCContainer module with an encoder, a recurrent module, and a linear.c 6<V^8dQhRS[RS[RS[RS[RR/#r input_dim hidden_dim output_dim num_layersr Nint)rrs"rrLSTMLinearModel.__annotate__es5888*-8;>8LO8 8rc<\SV`4\P!WV4Vn\P !W#4VnR#rb)rrrLSTMlstmr6r<rrrrrrs&&&&&rrLSTMLinearModel.__init__es2 GGI:> ii 7 rcx<V^8dQhRS[PRS[S[PS[P3,/#)r inputr rDrEtuple)rrs"rrrls1U\\eELL%,,4N.OrcPVPV4wr#VPV4pWB3#rb)rr<)rroutput_hiddendecodeds&& rrdLSTMLinearModel.forwardls())E*++f%r)r<rrgr1s@@rrrbs!M88rrcTaa]tRtRtoRtV3RlV3RlltV3RlRltRtVtV;t #)LSTMLayerNormLinearModelirz9Container module with an LSTM, a LayerNorm, and a linear.c 6<V^8dQhRS[RS[RS[RS[RR/#rr)rrs"rr%LSTMLayerNormLinearModel.__annotate__us5888*-8;>8LO8 8rc<\SV`4\P!WV4Vn\P !V4Vn\P!W#4VnR#rb) rrrrr LayerNormnormr6r<rs&&&&&rr!LSTMLayerNormLinearModel.__init__usB GGI:> LL, ii 7 rcx<V^8dQhRS[PRS[S[PS[P3,/#r]r)rrs"rrr}s1% ell0J*KrcrVPV4wrVPV4pVPV4pW3#rb)rr r<)rr^r$s&& rrd LSTMLayerNormLinearModel.forward}s299Q< IIaL KKNxr)r<rr rgr1s@@rrrrs!C88rr)typingrtorch.ao.pruningrrDtorch.nn.functionalr functionalrrr6r3rLrrNrirtrrrrrrrrrrr)rrrs+ KNK ryy   299*&ryy8>299(*ryy8BIIDB"))D  F bii  ryyr