node1 -- Input0 -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=0.0, parameter_size=0.000 node4 -- Embedding(32320, 1024, padding_idx=0) -- forward_compute_time=0.140, backward_compute_time=25.538, activation_size=13107200.0, parameter_size=132382720.000 node5 -- EmuBidirLSTM( (bidir): LSTM(1024, 1024, bidirectional=True) (layer1): LSTM(1024, 1024) (layer2): LSTM(1024, 1024)) -- forward_compute_time=14.126, backward_compute_time=0.019, activation_size=26214400.0, parameter_size=67174400.000 node2 -- Input1 -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=0.0, parameter_size=0.000 node6 -- Dropout(p=0.2) -- forward_compute_time=0.186, backward_compute_time=0.285, activation_size=26214400.0, parameter_size=0.000 node7 -- LSTM(2048, 1024) -- forward_compute_time=10.298, backward_compute_time=18.777, activation_size=[13107200.0; 524288.0; 524288.0], parameter_size=50364416.000 node8 -- __getitem__(0) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node9 -- __getitem__(1) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=524288.0, parameter_size=0.000 node10 -- Dropout(p=0.2) -- forward_compute_time=0.124, backward_compute_time=0.159, activation_size=13107200.0, parameter_size=0.000 node11 -- LSTM(1024, 1024) -- forward_compute_time=6.887, backward_compute_time=13.549, activation_size=[13107200.0; 524288.0; 524288.0], parameter_size=33587200.000 node12 -- __getitem__(0) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node13 -- __getitem__(1) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=524288.0, parameter_size=0.000 node14 -- Add -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node15 -- Dropout(p=0.2) -- forward_compute_time=0.129, backward_compute_time=0.167, activation_size=13107200.0, parameter_size=0.000 node16 -- LSTM(1024, 1024) -- forward_compute_time=6.903, backward_compute_time=13.518, activation_size=[13107200.0; 524288.0; 524288.0], parameter_size=33587200.000 node17 -- __getitem__(0) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node18 -- __getitem__(1) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=524288.0, parameter_size=0.000 node19 -- Add -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node20 -- Dropout(p=0.2) -- forward_compute_time=0.122, backward_compute_time=0.160, activation_size=13107200.0, parameter_size=0.000 node21 -- LSTM(1024, 1024) -- forward_compute_time=6.889, backward_compute_time=13.594, activation_size=[13107200.0; 524288.0; 524288.0], parameter_size=33587200.000 node22 -- __getitem__(0) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node23 -- __getitem__(1) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=524288.0, parameter_size=0.000 node24 -- Add -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node25 -- Dropout(p=0.2) -- forward_compute_time=0.120, backward_compute_time=0.160, activation_size=13107200.0, parameter_size=0.000 node26 -- LSTM(1024, 1024) -- forward_compute_time=7.019, backward_compute_time=13.628, activation_size=[13107200.0; 524288.0; 524288.0], parameter_size=33587200.000 node27 -- __getitem__(0) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node28 -- __getitem__(1) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=524288.0, parameter_size=0.000 node29 -- Add -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node30 -- Dropout(p=0.2) -- forward_compute_time=0.120, backward_compute_time=0.164, activation_size=13107200.0, parameter_size=0.000 node31 -- LSTM(1024, 1024) -- forward_compute_time=6.991, backward_compute_time=13.614, activation_size=[13107200.0; 524288.0; 524288.0], parameter_size=33587200.000 node32 -- __getitem__(0) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node33 -- __getitem__(1) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=524288.0, parameter_size=0.000 node34 -- Add -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node35 -- Dropout(p=0.2) -- forward_compute_time=0.120, backward_compute_time=0.169, activation_size=13107200.0, parameter_size=0.000 node36 -- LSTM(1024, 1024) -- forward_compute_time=7.055, backward_compute_time=14.112, activation_size=[13107200.0; 524288.0; 524288.0], parameter_size=33587200.000 node37 -- __getitem__(0) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node38 -- __getitem__(1) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=524288.0, parameter_size=0.000 node39 -- Add -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node3 -- Input2 -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=0.0, parameter_size=0.000 node41 -- Embedding(32320, 1024, padding_idx=0) -- forward_compute_time=0.128, backward_compute_time=0.460, activation_size=13107200.0, parameter_size=132382720.000 node40 -- hidden -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=0.0, parameter_size=0.000 node42 -- __getitem__(0) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=0.0, parameter_size=0.000 node43 -- RecurrentAttention( (rnn): LSTM(1024, 1024) (attn): BahdanauAttention( (linear_q): Linear(in_features=1024, out_features=1024, bias=False) (linear_k): Linear(in_features=1024, out_features=1024, bias=False) (dropout): Dropout(p=0) ) (dropout): Dropout(p=0)) -- forward_compute_time=15.210, backward_compute_time=23.086, activation_size=[13107200.0; 524288.0; 524288.0; 13107200.0; 320000.0], parameter_size=41979904.000 node44 -- __getitem__(0) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node45 -- __getitem__(1) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=524288.0, parameter_size=0.000 node46 -- __getitem__(2) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=524288.0, parameter_size=0.000 node47 -- __getitem__(3) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node48 -- Dropout(p=0.2) -- forward_compute_time=0.104, backward_compute_time=0.262, activation_size=13107200.0, parameter_size=0.000 node49 -- Concat(2) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node50 -- __getitem__(1) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=0.0, parameter_size=0.000 node51 -- LSTM(2048, 1024) -- forward_compute_time=10.517, backward_compute_time=18.744, activation_size=[13107200.0; 524288.0; 524288.0], parameter_size=50364416.000 node52 -- __getitem__(0) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node53 -- __getitem__(1) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=524288.0, parameter_size=0.000 node54 -- Dropout(p=0.2) -- forward_compute_time=0.109, backward_compute_time=0.277, activation_size=13107200.0, parameter_size=0.000 node55 -- Concat(2) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node56 -- __getitem__(2) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=0.0, parameter_size=0.000 node57 -- LSTM(2048, 1024) -- forward_compute_time=10.553, backward_compute_time=18.771, activation_size=[13107200.0; 524288.0; 524288.0], parameter_size=50364416.000 node58 -- __getitem__(0) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node59 -- __getitem__(1) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=524288.0, parameter_size=0.000 node60 -- Add -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node61 -- Dropout(p=0.2) -- forward_compute_time=0.120, backward_compute_time=0.256, activation_size=13107200.0, parameter_size=0.000 node62 -- Concat(2) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node63 -- __getitem__(3) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=0.0, parameter_size=0.000 node64 -- LSTM(2048, 1024) -- forward_compute_time=10.725, backward_compute_time=18.802, activation_size=[13107200.0; 524288.0; 524288.0], parameter_size=50364416.000 node65 -- __getitem__(0) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node66 -- __getitem__(1) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=524288.0, parameter_size=0.000 node67 -- Add -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node68 -- Dropout(p=0.2) -- forward_compute_time=0.121, backward_compute_time=0.263, activation_size=13107200.0, parameter_size=0.000 node69 -- Concat(2) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node70 -- __getitem__(4) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=0.0, parameter_size=0.000 node71 -- LSTM(2048, 1024) -- forward_compute_time=10.643, backward_compute_time=18.863, activation_size=[13107200.0; 524288.0; 524288.0], parameter_size=50364416.000 node72 -- __getitem__(0) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node73 -- __getitem__(1) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=524288.0, parameter_size=0.000 node74 -- Add -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node75 -- Dropout(p=0.2) -- forward_compute_time=0.131, backward_compute_time=0.263, activation_size=13107200.0, parameter_size=0.000 node76 -- Concat(2) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node77 -- __getitem__(5) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=0.0, parameter_size=0.000 node78 -- LSTM(2048, 1024) -- forward_compute_time=10.651, backward_compute_time=19.036, activation_size=[13107200.0; 524288.0; 524288.0], parameter_size=50364416.000 node79 -- __getitem__(0) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node80 -- __getitem__(1) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=524288.0, parameter_size=0.000 node81 -- Add -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node82 -- Dropout(p=0.2) -- forward_compute_time=0.126, backward_compute_time=0.261, activation_size=13107200.0, parameter_size=0.000 node83 -- Concat(2) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node84 -- __getitem__(6) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=0.0, parameter_size=0.000 node85 -- LSTM(2048, 1024) -- forward_compute_time=10.659, backward_compute_time=19.145, activation_size=[13107200.0; 524288.0; 524288.0], parameter_size=50364416.000 node86 -- __getitem__(0) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node87 -- __getitem__(1) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=524288.0, parameter_size=0.000 node88 -- Add -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node89 -- Dropout(p=0.2) -- forward_compute_time=0.144, backward_compute_time=0.201, activation_size=13107200.0, parameter_size=0.000 node90 -- Concat(2) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node91 -- __getitem__(7) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=0.0, parameter_size=0.000 node92 -- LSTM(2048, 1024) -- forward_compute_time=10.611, backward_compute_time=66.214, activation_size=[13107200.0; 524288.0; 524288.0], parameter_size=50364416.000 node93 -- __getitem__(0) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node94 -- __getitem__(1) -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=524288.0, parameter_size=0.000 node95 -- Add -- forward_compute_time=0.000, backward_compute_time=0.000, activation_size=13107200.0, parameter_size=0.000 node96 -- Classifier( (classifier): Linear(in_features=1024, out_features=32320, bias=True)) -- forward_compute_time=24.782, backward_compute_time=5.373, activation_size=413696000.0, parameter_size=132512000.000 node1 -- node4 node4 -- node5 node2 -- node5 node5 -- node6 node6 -- node7 node7 -- node8 node7 -- node9 node8 -- node10 node10 -- node11 node11 -- node12 node11 -- node13 node12 -- node14 node8 -- node14 node14 -- node15 node15 -- node16 node16 -- node17 node16 -- node18 node17 -- node19 node14 -- node19 node19 -- node20 node20 -- node21 node21 -- node22 node21 -- node23 node22 -- node24 node19 -- node24 node24 -- node25 node25 -- node26 node26 -- node27 node26 -- node28 node27 -- node29 node24 -- node29 node29 -- node30 node30 -- node31 node31 -- node32 node31 -- node33 node32 -- node34 node29 -- node34 node34 -- node35 node35 -- node36 node36 -- node37 node36 -- node38 node37 -- node39 node34 -- node39 node3 -- node41 node40 -- node42 node41 -- node43 node42 -- node43 node39 -- node43 node2 -- node43 node43 -- node44 node43 -- node45 node43 -- node46 node43 -- node47 node44 -- node48 node48 -- node49 node46 -- node49 node40 -- node50 node49 -- node51 node50 -- node51 node51 -- node52 node51 -- node53 node52 -- node54 node54 -- node55 node46 -- node55 node40 -- node56 node55 -- node57 node56 -- node57 node57 -- node58 node57 -- node59 node58 -- node60 node52 -- node60 node60 -- node61 node61 -- node62 node46 -- node62 node40 -- node63 node62 -- node64 node63 -- node64 node64 -- node65 node64 -- node66 node65 -- node67 node60 -- node67 node67 -- node68 node68 -- node69 node46 -- node69 node40 -- node70 node69 -- node71 node70 -- node71 node71 -- node72 node71 -- node73 node72 -- node74 node67 -- node74 node74 -- node75 node75 -- node76 node46 -- node76 node40 -- node77 node76 -- node78 node77 -- node78 node78 -- node79 node78 -- node80 node79 -- node81 node74 -- node81 node81 -- node82 node82 -- node83 node46 -- node83 node40 -- node84 node83 -- node85 node84 -- node85 node85 -- node86 node85 -- node87 node86 -- node88 node81 -- node88 node88 -- node89 node89 -- node90 node46 -- node90 node40 -- node91 node90 -- node92 node91 -- node92 node92 -- node93 node92 -- node94 node93 -- node95 node88 -- node95 node95 -- node96