<
   ir_version: 10,
   opset_import: ["" : 23],
   producer_name: "pytorch"
>
main_graph (uint8[batch,448,448,3] image) => (float[batch,1024] image_embedding) 
   <
      float[batch,1024,56,56] add_1072
      float[batch,1024,56,56] add_1148
      float[batch,1024,56,56] add_1224
      float[batch,1024,56,56] add_1300
      float[batch,1024,56,56] add_1376
      float[batch,512,112,112] add_140
      float[batch,1024,56,56] add_1452
      float[batch,2048,28,28] add_1548
      float[batch,2048,28,28] add_1624
      float[batch,2048,28,28] add_1700
      float[batch,2048,28,28] add_1776
      float[batch,2048,28,28] add_1852
      float[batch,2048,28,28] add_1928
      float[batch,2048,28,28] add_2004
      float[batch,2048,28,28] add_2080
      float[batch,2048,28,28] add_2156
      float[batch,512,112,112] add_216
      float[batch,2048,28,28] add_2232
      float[batch,2048,28,28] add_2308
      float[batch,2048,28,28] add_2384
      float[batch,2048,28,28] add_2460
      float[batch,2048,28,28] add_2536
      float[batch,2048,28,28] add_2612
      float[batch,2048,28,28] add_2688
      float[batch,2048,28,28] add_2764
      float[batch,2048,28,28] add_2840
      float[batch,2048,28,28] add_2916
      float[batch,512,112,112] add_292
      float[batch,2048,28,28] add_2992
      float[batch,2048,28,28] add_3068
      float[batch,2048,28,28] add_3144
      float[batch,2048,28,28] add_3220
      float[batch,2048,28,28] add_3296
      float[batch,2048,28,28] add_3372
      float[batch,2048,28,28] add_3448
      float[batch,2048,28,28] add_3524
      float[batch,2048,28,28] add_3600
      float[batch,2048,28,28] add_3676
      float[batch,2048,28,28] add_3752
      float[batch,2048,28,28] add_3828
      float[batch,1024,56,56] add_388
      float[batch,2048,28,28] add_3904
      float[batch,2048,28,28] add_3980
      float[batch,2048,28,28] add_4056
      float[batch,2048,28,28] add_4132
      float[batch,2048,28,28] add_4208
      float[batch,4096,14,14] add_4304
      float[batch,4096,14,14] add_4380
      float[batch,4096,14,14] add_4456
      float[batch,4096,14,14] add_4532
      float[batch,4096,14,14] add_4608
      float[batch,1024,56,56] add_464
      float[batch,4096,14,14] add_4684
      float[batch,4096,14,14] add_4760
      float[batch,4096,14,14] add_4836
      float[batch,4096,14,14] add_4912
      float[batch,4096,14,14] add_4988
      float[197,batch,4096] add_5018
      float[batch,1024,56,56] add_540
      float[batch,1024,56,56] add_616
      float[batch,1024,56,56] add_692
      float[batch,1024,56,56] add_768
      float[batch,1024,56,56] add_844
      float[batch,1024,56,56] add_920
      float[batch,1024,56,56] add_996
      float[batch,128,112,112] avg_pool2d
      float[batch,256,56,56] avg_pool2d_2
      float[batch,512,56,56] avg_pool2d_3
      float[batch,512,28,28] avg_pool2d_4
      float[batch,1024,28,28] avg_pool2d_5
      float[batch,1024,14,14] avg_pool2d_6
      float[batch,2048,14,14] avg_pool2d_7
      float[197,batch,4096] cat
      float[batch,1] clamp_min
      float[batch,64,224,224] getitem
      float[batch,1024,56,56] getitem_102
      float[batch,256,56,56] getitem_105
      float[batch,256,56,56] getitem_108
      float[batch,1024,56,56] getitem_111
      float[batch,256,56,56] getitem_114
      float[batch,256,56,56] getitem_117
      float[batch,128,112,112] getitem_12
      float[batch,1024,56,56] getitem_120
      float[batch,256,56,56] getitem_123
      float[batch,256,56,56] getitem_126
      float[batch,1024,56,56] getitem_129
      float[batch,256,56,56] getitem_132
      float[batch,256,56,56] getitem_135
      float[batch,1024,56,56] getitem_138
      float[batch,256,56,56] getitem_141
      float[batch,256,56,56] getitem_144
      float[batch,1024,56,56] getitem_147
      float[batch,512,112,112] getitem_15
      float[batch,256,56,56] getitem_150
      float[batch,256,56,56] getitem_153
      float[batch,1024,56,56] getitem_156
      float[batch,256,56,56] getitem_159
      float[batch,256,56,56] getitem_162
      float[batch,1024,56,56] getitem_165
      float[batch,256,56,56] getitem_168
      float[batch,256,56,56] getitem_171
      float[batch,1024,56,56] getitem_174
      float[batch,512,56,56] getitem_177
      float[batch,512,112,112] getitem_18
      float[batch,512,56,56] getitem_180
      float[batch,2048,28,28] getitem_183
      float[batch,2048,28,28] getitem_186
      float[batch,512,28,28] getitem_189
      float[batch,512,28,28] getitem_192
      float[batch,2048,28,28] getitem_195
      float[batch,512,28,28] getitem_198
      float[batch,512,28,28] getitem_201
      float[batch,2048,28,28] getitem_204
      float[batch,512,28,28] getitem_207
      float[batch,128,112,112] getitem_21
      float[batch,512,28,28] getitem_210
      float[batch,2048,28,28] getitem_213
      float[batch,512,28,28] getitem_216
      float[batch,512,28,28] getitem_219
      float[batch,2048,28,28] getitem_222
      float[batch,512,28,28] getitem_225
      float[batch,512,28,28] getitem_228
      float[batch,2048,28,28] getitem_231
      float[batch,512,28,28] getitem_234
      float[batch,512,28,28] getitem_237
      float[batch,128,112,112] getitem_24
      float[batch,2048,28,28] getitem_240
      float[batch,512,28,28] getitem_243
      float[batch,512,28,28] getitem_246
      float[batch,2048,28,28] getitem_249
      float[batch,512,28,28] getitem_252
      float[batch,512,28,28] getitem_255
      float[batch,2048,28,28] getitem_258
      float[batch,512,28,28] getitem_261
      float[batch,512,28,28] getitem_264
      float[batch,2048,28,28] getitem_267
      float[batch,512,112,112] getitem_27
      float[batch,512,28,28] getitem_270
      float[batch,512,28,28] getitem_273
      float[batch,2048,28,28] getitem_276
      float[batch,512,28,28] getitem_279
      float[batch,512,28,28] getitem_282
      float[batch,2048,28,28] getitem_285
      float[batch,512,28,28] getitem_288
      float[batch,512,28,28] getitem_291
      float[batch,2048,28,28] getitem_294
      float[batch,512,28,28] getitem_297
      float[batch,64,224,224] getitem_3
      float[batch,128,112,112] getitem_30
      float[batch,512,28,28] getitem_300
      float[batch,2048,28,28] getitem_303
      float[batch,512,28,28] getitem_306
      float[batch,512,28,28] getitem_309
      float[batch,2048,28,28] getitem_312
      float[batch,512,28,28] getitem_315
      float[batch,512,28,28] getitem_318
      float[batch,2048,28,28] getitem_321
      float[batch,512,28,28] getitem_324
      float[batch,512,28,28] getitem_327
      float[batch,128,112,112] getitem_33
      float[batch,2048,28,28] getitem_330
      float[batch,512,28,28] getitem_333
      float[batch,512,28,28] getitem_336
      float[batch,2048,28,28] getitem_339
      float[batch,512,28,28] getitem_342
      float[batch,512,28,28] getitem_345
      float[batch,2048,28,28] getitem_348
      float[batch,512,28,28] getitem_351
      float[batch,512,28,28] getitem_354
      float[batch,2048,28,28] getitem_357
      float[batch,512,112,112] getitem_36
      float[batch,512,28,28] getitem_360
      float[batch,512,28,28] getitem_363
      float[batch,2048,28,28] getitem_366
      float[batch,512,28,28] getitem_369
      float[batch,512,28,28] getitem_372
      float[batch,2048,28,28] getitem_375
      float[batch,512,28,28] getitem_378
      float[batch,512,28,28] getitem_381
      float[batch,2048,28,28] getitem_384
      float[batch,512,28,28] getitem_387
      float[batch,256,112,112] getitem_39
      float[batch,512,28,28] getitem_390
      float[batch,2048,28,28] getitem_393
      float[batch,512,28,28] getitem_396
      float[batch,512,28,28] getitem_399
      float[batch,2048,28,28] getitem_402
      float[batch,512,28,28] getitem_405
      float[batch,512,28,28] getitem_408
      float[batch,2048,28,28] getitem_411
      float[batch,512,28,28] getitem_414
      float[batch,512,28,28] getitem_417
      float[batch,256,112,112] getitem_42
      float[batch,2048,28,28] getitem_420
      float[batch,512,28,28] getitem_423
      float[batch,512,28,28] getitem_426
      float[batch,2048,28,28] getitem_429
      float[batch,512,28,28] getitem_432
      float[batch,512,28,28] getitem_435
      float[batch,2048,28,28] getitem_438
      float[batch,512,28,28] getitem_441
      float[batch,512,28,28] getitem_444
      float[batch,2048,28,28] getitem_447
      float[batch,1024,56,56] getitem_45
      float[batch,512,28,28] getitem_450
      float[batch,512,28,28] getitem_453
      float[batch,2048,28,28] getitem_456
      float[batch,512,28,28] getitem_459
      float[batch,512,28,28] getitem_462
      float[batch,2048,28,28] getitem_465
      float[batch,512,28,28] getitem_468
      float[batch,512,28,28] getitem_471
      float[batch,2048,28,28] getitem_474
      float[batch,512,28,28] getitem_477
      float[batch,1024,56,56] getitem_48
      float[batch,512,28,28] getitem_480
      float[batch,2048,28,28] getitem_483
      float[batch,512,28,28] getitem_486
      float[batch,512,28,28] getitem_489
      float[batch,2048,28,28] getitem_492
      float[batch,512,28,28] getitem_495
      float[batch,512,28,28] getitem_498
      float[batch,2048,28,28] getitem_501
      float[batch,1024,28,28] getitem_504
      float[batch,1024,28,28] getitem_507
      float[batch,256,56,56] getitem_51
      float[batch,4096,14,14] getitem_510
      float[batch,4096,14,14] getitem_513
      float[batch,1024,14,14] getitem_516
      float[batch,1024,14,14] getitem_519
      float[batch,4096,14,14] getitem_522
      float[batch,1024,14,14] getitem_525
      float[batch,1024,14,14] getitem_528
      float[batch,4096,14,14] getitem_531
      float[batch,1024,14,14] getitem_534
      float[batch,1024,14,14] getitem_537
      float[batch,256,56,56] getitem_54
      float[batch,4096,14,14] getitem_540
      float[batch,1024,14,14] getitem_543
      float[batch,1024,14,14] getitem_546
      float[batch,4096,14,14] getitem_549
      float[batch,1024,14,14] getitem_552
      float[batch,1024,14,14] getitem_555
      float[batch,4096,14,14] getitem_558
      float[batch,1024,14,14] getitem_561
      float[batch,1024,14,14] getitem_564
      float[batch,4096,14,14] getitem_567
      float[batch,1024,56,56] getitem_57
      float[batch,1024,14,14] getitem_570
      float[batch,1024,14,14] getitem_573
      float[batch,4096,14,14] getitem_576
      float[batch,1024,14,14] getitem_579
      float[batch,1024,14,14] getitem_582
      float[batch,4096,14,14] getitem_585
      float[batch,1024,14,14] getitem_588
      float[batch,1024,14,14] getitem_591
      float[batch,4096,14,14] getitem_594
      float[batch,128,224,224] getitem_6
      float[batch,256,56,56] getitem_60
      float[batch,256,56,56] getitem_63
      float[batch,1024,56,56] getitem_66
      float[batch,256,56,56] getitem_69
      float[batch,256,56,56] getitem_72
      float[batch,1024,56,56] getitem_75
      float[batch,256,56,56] getitem_78
      float[batch,256,56,56] getitem_81
      float[batch,1024,56,56] getitem_84
      float[batch,256,56,56] getitem_87
      float[batch,128,112,112] getitem_9
      float[batch,256,56,56] getitem_90
      float[batch,1024,56,56] getitem_93
      float[batch,256,56,56] getitem_96
      float[batch,256,56,56] getitem_99
      float[batch,3,448,448] image_chw
      float[batch,448,448,3] image_f32
      float[batch,448,448,3] image_shifted
      float[batch,1] linalg_vector_norm
      float[1,batch,4096] linear
      float[197,batch,4096] linear_1
      float[197,batch,4096] linear_2
      float[batch,1024] linear_3
      float[1,batch,4096] mean
      float[1,batch,4096] node_scaled_dot_product_attention_q_row
      float[196,batch,4096] permute_1
      float[1,batch,64,64] permute_2
      float[batch,64,224,224] relu
      float[batch,64,224,224] relu_1
      float[batch,128,112,112] relu_10
      float[batch,512,28,28] relu_100
      float[batch,2048,28,28] relu_101
      float[batch,512,28,28] relu_102
      float[batch,512,28,28] relu_103
      float[batch,2048,28,28] relu_104
      float[batch,512,28,28] relu_105
      float[batch,512,28,28] relu_106
      float[batch,2048,28,28] relu_107
      float[batch,512,28,28] relu_108
      float[batch,512,28,28] relu_109
      float[batch,512,112,112] relu_11
      float[batch,2048,28,28] relu_110
      float[batch,512,28,28] relu_111
      float[batch,512,28,28] relu_112
      float[batch,2048,28,28] relu_113
      float[batch,512,28,28] relu_114
      float[batch,512,28,28] relu_115
      float[batch,2048,28,28] relu_116
      float[batch,512,28,28] relu_117
      float[batch,512,28,28] relu_118
      float[batch,2048,28,28] relu_119
      float[batch,256,112,112] relu_12
      float[batch,512,28,28] relu_120
      float[batch,512,28,28] relu_121
      float[batch,2048,28,28] relu_122
      float[batch,512,28,28] relu_123
      float[batch,512,28,28] relu_124
      float[batch,2048,28,28] relu_125
      float[batch,512,28,28] relu_126
      float[batch,512,28,28] relu_127
      float[batch,2048,28,28] relu_128
      float[batch,512,28,28] relu_129
      float[batch,256,112,112] relu_13
      float[batch,512,28,28] relu_130
      float[batch,2048,28,28] relu_131
      float[batch,512,28,28] relu_132
      float[batch,512,28,28] relu_133
      float[batch,2048,28,28] relu_134
      float[batch,512,28,28] relu_135
      float[batch,512,28,28] relu_136
      float[batch,2048,28,28] relu_137
      float[batch,512,28,28] relu_138
      float[batch,512,28,28] relu_139
      float[batch,1024,56,56] relu_14
      float[batch,2048,28,28] relu_140
      float[batch,512,28,28] relu_141
      float[batch,512,28,28] relu_142
      float[batch,2048,28,28] relu_143
      float[batch,512,28,28] relu_144
      float[batch,512,28,28] relu_145
      float[batch,2048,28,28] relu_146
      float[batch,512,28,28] relu_147
      float[batch,512,28,28] relu_148
      float[batch,2048,28,28] relu_149
      float[batch,256,56,56] relu_15
      float[batch,512,28,28] relu_150
      float[batch,512,28,28] relu_151
      float[batch,2048,28,28] relu_152
      float[batch,512,28,28] relu_153
      float[batch,512,28,28] relu_154
      float[batch,2048,28,28] relu_155
      float[batch,512,28,28] relu_156
      float[batch,512,28,28] relu_157
      float[batch,2048,28,28] relu_158
      float[batch,512,28,28] relu_159
      float[batch,256,56,56] relu_16
      float[batch,512,28,28] relu_160
      float[batch,2048,28,28] relu_161
      float[batch,512,28,28] relu_162
      float[batch,512,28,28] relu_163
      float[batch,2048,28,28] relu_164
      float[batch,1024,28,28] relu_165
      float[batch,1024,28,28] relu_166
      float[batch,4096,14,14] relu_167
      float[batch,1024,14,14] relu_168
      float[batch,1024,14,14] relu_169
      float[batch,1024,56,56] relu_17
      float[batch,4096,14,14] relu_170
      float[batch,1024,14,14] relu_171
      float[batch,1024,14,14] relu_172
      float[batch,4096,14,14] relu_173
      float[batch,1024,14,14] relu_174
      float[batch,1024,14,14] relu_175
      float[batch,4096,14,14] relu_176
      float[batch,1024,14,14] relu_177
      float[batch,1024,14,14] relu_178
      float[batch,4096,14,14] relu_179
      float[batch,256,56,56] relu_18
      float[batch,1024,14,14] relu_180
      float[batch,1024,14,14] relu_181
      float[batch,4096,14,14] relu_182
      float[batch,1024,14,14] relu_183
      float[batch,1024,14,14] relu_184
      float[batch,4096,14,14] relu_185
      float[batch,1024,14,14] relu_186
      float[batch,1024,14,14] relu_187
      float[batch,4096,14,14] relu_188
      float[batch,1024,14,14] relu_189
      float[batch,256,56,56] relu_19
      float[batch,1024,14,14] relu_190
      float[batch,4096,14,14] relu_191
      float[batch,1024,14,14] relu_192
      float[batch,1024,14,14] relu_193
      float[batch,4096,14,14] relu_194
      float[batch,128,224,224] relu_2
      float[batch,1024,56,56] relu_20
      float[batch,256,56,56] relu_21
      float[batch,256,56,56] relu_22
      float[batch,1024,56,56] relu_23
      float[batch,256,56,56] relu_24
      float[batch,256,56,56] relu_25
      float[batch,1024,56,56] relu_26
      float[batch,256,56,56] relu_27
      float[batch,256,56,56] relu_28
      float[batch,1024,56,56] relu_29
      float[batch,128,112,112] relu_3
      float[batch,256,56,56] relu_30
      float[batch,256,56,56] relu_31
      float[batch,1024,56,56] relu_32
      float[batch,256,56,56] relu_33
      float[batch,256,56,56] relu_34
      float[batch,1024,56,56] relu_35
      float[batch,256,56,56] relu_36
      float[batch,256,56,56] relu_37
      float[batch,1024,56,56] relu_38
      float[batch,256,56,56] relu_39
      float[batch,128,112,112] relu_4
      float[batch,256,56,56] relu_40
      float[batch,1024,56,56] relu_41
      float[batch,256,56,56] relu_42
      float[batch,256,56,56] relu_43
      float[batch,1024,56,56] relu_44
      float[batch,256,56,56] relu_45
      float[batch,256,56,56] relu_46
      float[batch,1024,56,56] relu_47
      float[batch,256,56,56] relu_48
      float[batch,256,56,56] relu_49
      float[batch,512,112,112] relu_5
      float[batch,1024,56,56] relu_50
      float[batch,256,56,56] relu_51
      float[batch,256,56,56] relu_52
      float[batch,1024,56,56] relu_53
      float[batch,256,56,56] relu_54
      float[batch,256,56,56] relu_55
      float[batch,1024,56,56] relu_56
      float[batch,512,56,56] relu_57
      float[batch,512,56,56] relu_58
      float[batch,2048,28,28] relu_59
      float[batch,128,112,112] relu_6
      float[batch,512,28,28] relu_60
      float[batch,512,28,28] relu_61
      float[batch,2048,28,28] relu_62
      float[batch,512,28,28] relu_63
      float[batch,512,28,28] relu_64
      float[batch,2048,28,28] relu_65
      float[batch,512,28,28] relu_66
      float[batch,512,28,28] relu_67
      float[batch,2048,28,28] relu_68
      float[batch,512,28,28] relu_69
      float[batch,128,112,112] relu_7
      float[batch,512,28,28] relu_70
      float[batch,2048,28,28] relu_71
      float[batch,512,28,28] relu_72
      float[batch,512,28,28] relu_73
      float[batch,2048,28,28] relu_74
      float[batch,512,28,28] relu_75
      float[batch,512,28,28] relu_76
      float[batch,2048,28,28] relu_77
      float[batch,512,28,28] relu_78
      float[batch,512,28,28] relu_79
      float[batch,512,112,112] relu_8
      float[batch,2048,28,28] relu_80
      float[batch,512,28,28] relu_81
      float[batch,512,28,28] relu_82
      float[batch,2048,28,28] relu_83
      float[batch,512,28,28] relu_84
      float[batch,512,28,28] relu_85
      float[batch,2048,28,28] relu_86
      float[batch,512,28,28] relu_87
      float[batch,512,28,28] relu_88
      float[batch,2048,28,28] relu_89
      float[batch,128,112,112] relu_9
      float[batch,512,28,28] relu_90
      float[batch,512,28,28] relu_91
      float[batch,2048,28,28] relu_92
      float[batch,512,28,28] relu_93
      float[batch,512,28,28] relu_94
      float[batch,2048,28,28] relu_95
      float[batch,512,28,28] relu_96
      float[batch,512,28,28] relu_97
      float[batch,2048,28,28] relu_98
      float[batch,512,28,28] relu_99
      float[batch,64,1,64] scaled_dot_product_attention
      float[batch,1024] select
      float[4096] split_split_0
      float[4096] split_split_1
      float[4096] split_split_2
      float[unk__1,1,64] transpose
      float[unk__1,197,64] transpose_1
      float[unk__1,197,64] transpose_2
      float[197,1,4096] unsqueeze
      float[1,batch,4096] val_7
      float[197,batch,4096] val_8
      float[197,batch,4096] val_9
      float[1,batch,1024] view_10
      float[batch,4096,196] view_2
      float[1,unk__1,64] view_3
      float[197,unk__1,64] view_4
      float[197,unk__1,64] view_5
      float[batch,64,1,64] view_6
      float[batch,64,197,64] view_7
      float[batch,64,197,64] view_8
      float[batch,4096] view_9
   >
{
   [pre_cast] image_f32 = Cast <to: int = 1> (image)
   [pre_shift] image_shifted = Sub (image_f32, image_shift)
   [pre_nhwc_to_nchw] image_chw = Transpose <perm: ints = [0, 3, 1, 2]> (image_shifted)
   getitem = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [2, 2]> (image_chw, "visual.conv1.weight", "visual.conv1.weight_bias")
   [node_relu] relu = Relu (getitem)
   getitem_3 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu, "visual.conv2.weight", "visual.conv2.weight_bias")
   relu_1 = Relu (getitem_3)
   getitem_6 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_1, "visual.conv3.weight", "visual.conv3.weight_bias")
   relu_2 = Relu (getitem_6)
   [node_avg_pool2d] avg_pool2d = AveragePool <auto_pad: string = "NOTSET", ceil_mode: int = 0, count_include_pad: int = 1, kernel_shape: ints = [2, 2], pads: ints = [0, 0, 0, 0], strides: ints = [2, 2]> (relu_2)
   getitem_9 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (avg_pool2d, "visual.layer1.0.conv1.weight", "visual.layer1.0.conv1.weight_bias")
   relu_3 = Relu (getitem_9)
   getitem_12 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_3, "visual.layer1.0.conv2.weight", "visual.layer1.0.conv2.weight_bias")
   relu_4 = Relu (getitem_12)
   getitem_15 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_4, "visual.layer1.0.conv3.weight", "visual.layer1.0.conv3.weight_bias")
   getitem_18 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (avg_pool2d, "visual.layer1.0.downsample.0.weight", "visual.layer1.0.downsample.0.weight_bias")
   add_140 = Add (getitem_15, getitem_18)
   relu_5 = Relu (add_140)
   getitem_21 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_5, "visual.layer1.1.conv1.weight", "visual.layer1.1.conv1.weight_bias")
   relu_6 = Relu (getitem_21)
   getitem_24 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_6, "visual.layer1.1.conv2.weight", "visual.layer1.1.conv2.weight_bias")
   relu_7 = Relu (getitem_24)
   getitem_27 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_7, "visual.layer1.1.conv3.weight", "visual.layer1.1.conv3.weight_bias")
   add_216 = Add (getitem_27, relu_5)
   relu_8 = Relu (add_216)
   getitem_30 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_8, "visual.layer1.2.conv1.weight", "visual.layer1.2.conv1.weight_bias")
   relu_9 = Relu (getitem_30)
   getitem_33 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_9, "visual.layer1.2.conv2.weight", "visual.layer1.2.conv2.weight_bias")
   relu_10 = Relu (getitem_33)
   getitem_36 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_10, "visual.layer1.2.conv3.weight", "visual.layer1.2.conv3.weight_bias")
   add_292 = Add (getitem_36, relu_8)
   relu_11 = Relu (add_292)
   getitem_39 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_11, "visual.layer2.0.conv1.weight", "visual.layer2.0.conv1.weight_bias")
   relu_12 = Relu (getitem_39)
   getitem_42 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_12, "visual.layer2.0.conv2.weight", "visual.layer2.0.conv2.weight_bias")
   relu_13 = Relu (getitem_42)
   avg_pool2d_2 = AveragePool <auto_pad: string = "NOTSET", ceil_mode: int = 0, count_include_pad: int = 1, kernel_shape: ints = [2, 2], pads: ints = [0, 0, 0, 0], strides: ints = [2, 2]> (relu_13)
   getitem_45 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (avg_pool2d_2, "visual.layer2.0.conv3.weight", "visual.layer2.0.conv3.weight_bias")
   avg_pool2d_3 = AveragePool <auto_pad: string = "NOTSET", ceil_mode: int = 0, count_include_pad: int = 1, kernel_shape: ints = [2, 2], pads: ints = [0, 0, 0, 0], strides: ints = [2, 2]> (relu_11)
   getitem_48 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (avg_pool2d_3, "visual.layer2.0.downsample.0.weight", "visual.layer2.0.downsample.0.weight_bias")
   add_388 = Add (getitem_45, getitem_48)
   relu_14 = Relu (add_388)
   getitem_51 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_14, "visual.layer2.1.conv1.weight", "visual.layer2.1.conv1.weight_bias")
   relu_15 = Relu (getitem_51)
   getitem_54 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_15, "visual.layer2.1.conv2.weight", "visual.layer2.1.conv2.weight_bias")
   relu_16 = Relu (getitem_54)
   getitem_57 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_16, "visual.layer2.1.conv3.weight", "visual.layer2.1.conv3.weight_bias")
   add_464 = Add (getitem_57, relu_14)
   relu_17 = Relu (add_464)
   getitem_60 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_17, "visual.layer2.2.conv1.weight", "visual.layer2.2.conv1.weight_bias")
   relu_18 = Relu (getitem_60)
   getitem_63 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_18, "visual.layer2.2.conv2.weight", "visual.layer2.2.conv2.weight_bias")
   relu_19 = Relu (getitem_63)
   getitem_66 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_19, "visual.layer2.2.conv3.weight", "visual.layer2.2.conv3.weight_bias")
   add_540 = Add (getitem_66, relu_17)
   relu_20 = Relu (add_540)
   getitem_69 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_20, "visual.layer2.3.conv1.weight", "visual.layer2.3.conv1.weight_bias")
   relu_21 = Relu (getitem_69)
   getitem_72 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_21, "visual.layer2.3.conv2.weight", "visual.layer2.3.conv2.weight_bias")
   relu_22 = Relu (getitem_72)
   getitem_75 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_22, "visual.layer2.3.conv3.weight", "visual.layer2.3.conv3.weight_bias")
   add_616 = Add (getitem_75, relu_20)
   relu_23 = Relu (add_616)
   getitem_78 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_23, "visual.layer2.4.conv1.weight", "visual.layer2.4.conv1.weight_bias")
   relu_24 = Relu (getitem_78)
   getitem_81 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_24, "visual.layer2.4.conv2.weight", "visual.layer2.4.conv2.weight_bias")
   relu_25 = Relu (getitem_81)
   getitem_84 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_25, "visual.layer2.4.conv3.weight", "visual.layer2.4.conv3.weight_bias")
   add_692 = Add (getitem_84, relu_23)
   relu_26 = Relu (add_692)
   getitem_87 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_26, "visual.layer2.5.conv1.weight", "visual.layer2.5.conv1.weight_bias")
   relu_27 = Relu (getitem_87)
   getitem_90 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_27, "visual.layer2.5.conv2.weight", "visual.layer2.5.conv2.weight_bias")
   relu_28 = Relu (getitem_90)
   getitem_93 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_28, "visual.layer2.5.conv3.weight", "visual.layer2.5.conv3.weight_bias")
   add_768 = Add (getitem_93, relu_26)
   relu_29 = Relu (add_768)
   getitem_96 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_29, "visual.layer2.6.conv1.weight", "visual.layer2.6.conv1.weight_bias")
   relu_30 = Relu (getitem_96)
   getitem_99 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_30, "visual.layer2.6.conv2.weight", "visual.layer2.6.conv2.weight_bias")
   relu_31 = Relu (getitem_99)
   getitem_102 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_31, "visual.layer2.6.conv3.weight", "visual.layer2.6.conv3.weight_bias")
   add_844 = Add (getitem_102, relu_29)
   relu_32 = Relu (add_844)
   getitem_105 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_32, "visual.layer2.7.conv1.weight", "visual.layer2.7.conv1.weight_bias")
   relu_33 = Relu (getitem_105)
   getitem_108 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_33, "visual.layer2.7.conv2.weight", "visual.layer2.7.conv2.weight_bias")
   relu_34 = Relu (getitem_108)
   getitem_111 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_34, "visual.layer2.7.conv3.weight", "visual.layer2.7.conv3.weight_bias")
   add_920 = Add (getitem_111, relu_32)
   relu_35 = Relu (add_920)
   getitem_114 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_35, "visual.layer2.8.conv1.weight", "visual.layer2.8.conv1.weight_bias")
   relu_36 = Relu (getitem_114)
   getitem_117 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_36, "visual.layer2.8.conv2.weight", "visual.layer2.8.conv2.weight_bias")
   relu_37 = Relu (getitem_117)
   getitem_120 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_37, "visual.layer2.8.conv3.weight", "visual.layer2.8.conv3.weight_bias")
   add_996 = Add (getitem_120, relu_35)
   relu_38 = Relu (add_996)
   getitem_123 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_38, "visual.layer2.9.conv1.weight", "visual.layer2.9.conv1.weight_bias")
   relu_39 = Relu (getitem_123)
   getitem_126 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_39, "visual.layer2.9.conv2.weight", "visual.layer2.9.conv2.weight_bias")
   relu_40 = Relu (getitem_126)
   getitem_129 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_40, "visual.layer2.9.conv3.weight", "visual.layer2.9.conv3.weight_bias")
   add_1072 = Add (getitem_129, relu_38)
   relu_41 = Relu (add_1072)
   getitem_132 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_41, "visual.layer2.10.conv1.weight", "visual.layer2.10.conv1.weight_bias")
   relu_42 = Relu (getitem_132)
   getitem_135 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_42, "visual.layer2.10.conv2.weight", "visual.layer2.10.conv2.weight_bias")
   relu_43 = Relu (getitem_135)
   getitem_138 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_43, "visual.layer2.10.conv3.weight", "visual.layer2.10.conv3.weight_bias")
   add_1148 = Add (getitem_138, relu_41)
   relu_44 = Relu (add_1148)
   getitem_141 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_44, "visual.layer2.11.conv1.weight", "visual.layer2.11.conv1.weight_bias")
   relu_45 = Relu (getitem_141)
   getitem_144 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_45, "visual.layer2.11.conv2.weight", "visual.layer2.11.conv2.weight_bias")
   relu_46 = Relu (getitem_144)
   getitem_147 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_46, "visual.layer2.11.conv3.weight", "visual.layer2.11.conv3.weight_bias")
   add_1224 = Add (getitem_147, relu_44)
   relu_47 = Relu (add_1224)
   getitem_150 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_47, "visual.layer2.12.conv1.weight", "visual.layer2.12.conv1.weight_bias")
   relu_48 = Relu (getitem_150)
   getitem_153 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_48, "visual.layer2.12.conv2.weight", "visual.layer2.12.conv2.weight_bias")
   relu_49 = Relu (getitem_153)
   getitem_156 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_49, "visual.layer2.12.conv3.weight", "visual.layer2.12.conv3.weight_bias")
   add_1300 = Add (getitem_156, relu_47)
   relu_50 = Relu (add_1300)
   getitem_159 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_50, "visual.layer2.13.conv1.weight", "visual.layer2.13.conv1.weight_bias")
   relu_51 = Relu (getitem_159)
   getitem_162 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_51, "visual.layer2.13.conv2.weight", "visual.layer2.13.conv2.weight_bias")
   relu_52 = Relu (getitem_162)
   getitem_165 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_52, "visual.layer2.13.conv3.weight", "visual.layer2.13.conv3.weight_bias")
   add_1376 = Add (getitem_165, relu_50)
   relu_53 = Relu (add_1376)
   getitem_168 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_53, "visual.layer2.14.conv1.weight", "visual.layer2.14.conv1.weight_bias")
   relu_54 = Relu (getitem_168)
   getitem_171 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_54, "visual.layer2.14.conv2.weight", "visual.layer2.14.conv2.weight_bias")
   relu_55 = Relu (getitem_171)
   getitem_174 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_55, "visual.layer2.14.conv3.weight", "visual.layer2.14.conv3.weight_bias")
   add_1452 = Add (getitem_174, relu_53)
   relu_56 = Relu (add_1452)
   getitem_177 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_56, "visual.layer3.0.conv1.weight", "visual.layer3.0.conv1.weight_bias")
   relu_57 = Relu (getitem_177)
   getitem_180 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_57, "visual.layer3.0.conv2.weight", "visual.layer3.0.conv2.weight_bias")
   relu_58 = Relu (getitem_180)
   avg_pool2d_4 = AveragePool <auto_pad: string = "NOTSET", ceil_mode: int = 0, count_include_pad: int = 1, kernel_shape: ints = [2, 2], pads: ints = [0, 0, 0, 0], strides: ints = [2, 2]> (relu_58)
   getitem_183 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (avg_pool2d_4, "visual.layer3.0.conv3.weight", "visual.layer3.0.conv3.weight_bias")
   avg_pool2d_5 = AveragePool <auto_pad: string = "NOTSET", ceil_mode: int = 0, count_include_pad: int = 1, kernel_shape: ints = [2, 2], pads: ints = [0, 0, 0, 0], strides: ints = [2, 2]> (relu_56)
   getitem_186 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (avg_pool2d_5, "visual.layer3.0.downsample.0.weight", "visual.layer3.0.downsample.0.weight_bias")
   add_1548 = Add (getitem_183, getitem_186)
   relu_59 = Relu (add_1548)
   getitem_189 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_59, "visual.layer3.1.conv1.weight", "visual.layer3.1.conv1.weight_bias")
   relu_60 = Relu (getitem_189)
   getitem_192 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_60, "visual.layer3.1.conv2.weight", "visual.layer3.1.conv2.weight_bias")
   relu_61 = Relu (getitem_192)
   getitem_195 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_61, "visual.layer3.1.conv3.weight", "visual.layer3.1.conv3.weight_bias")
   add_1624 = Add (getitem_195, relu_59)
   relu_62 = Relu (add_1624)
   getitem_198 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_62, "visual.layer3.2.conv1.weight", "visual.layer3.2.conv1.weight_bias")
   relu_63 = Relu (getitem_198)
   getitem_201 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_63, "visual.layer3.2.conv2.weight", "visual.layer3.2.conv2.weight_bias")
   relu_64 = Relu (getitem_201)
   getitem_204 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_64, "visual.layer3.2.conv3.weight", "visual.layer3.2.conv3.weight_bias")
   add_1700 = Add (getitem_204, relu_62)
   relu_65 = Relu (add_1700)
   getitem_207 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_65, "visual.layer3.3.conv1.weight", "visual.layer3.3.conv1.weight_bias")
   relu_66 = Relu (getitem_207)
   getitem_210 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_66, "visual.layer3.3.conv2.weight", "visual.layer3.3.conv2.weight_bias")
   relu_67 = Relu (getitem_210)
   getitem_213 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_67, "visual.layer3.3.conv3.weight", "visual.layer3.3.conv3.weight_bias")
   add_1776 = Add (getitem_213, relu_65)
   relu_68 = Relu (add_1776)
   getitem_216 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_68, "visual.layer3.4.conv1.weight", "visual.layer3.4.conv1.weight_bias")
   relu_69 = Relu (getitem_216)
   getitem_219 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_69, "visual.layer3.4.conv2.weight", "visual.layer3.4.conv2.weight_bias")
   relu_70 = Relu (getitem_219)
   getitem_222 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_70, "visual.layer3.4.conv3.weight", "visual.layer3.4.conv3.weight_bias")
   add_1852 = Add (getitem_222, relu_68)
   relu_71 = Relu (add_1852)
   getitem_225 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_71, "visual.layer3.5.conv1.weight", "visual.layer3.5.conv1.weight_bias")
   relu_72 = Relu (getitem_225)
   getitem_228 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_72, "visual.layer3.5.conv2.weight", "visual.layer3.5.conv2.weight_bias")
   relu_73 = Relu (getitem_228)
   getitem_231 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_73, "visual.layer3.5.conv3.weight", "visual.layer3.5.conv3.weight_bias")
   add_1928 = Add (getitem_231, relu_71)
   relu_74 = Relu (add_1928)
   getitem_234 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_74, "visual.layer3.6.conv1.weight", "visual.layer3.6.conv1.weight_bias")
   relu_75 = Relu (getitem_234)
   getitem_237 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_75, "visual.layer3.6.conv2.weight", "visual.layer3.6.conv2.weight_bias")
   relu_76 = Relu (getitem_237)
   getitem_240 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_76, "visual.layer3.6.conv3.weight", "visual.layer3.6.conv3.weight_bias")
   add_2004 = Add (getitem_240, relu_74)
   relu_77 = Relu (add_2004)
   getitem_243 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_77, "visual.layer3.7.conv1.weight", "visual.layer3.7.conv1.weight_bias")
   relu_78 = Relu (getitem_243)
   getitem_246 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_78, "visual.layer3.7.conv2.weight", "visual.layer3.7.conv2.weight_bias")
   relu_79 = Relu (getitem_246)
   getitem_249 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_79, "visual.layer3.7.conv3.weight", "visual.layer3.7.conv3.weight_bias")
   add_2080 = Add (getitem_249, relu_77)
   relu_80 = Relu (add_2080)
   getitem_252 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_80, "visual.layer3.8.conv1.weight", "visual.layer3.8.conv1.weight_bias")
   relu_81 = Relu (getitem_252)
   getitem_255 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_81, "visual.layer3.8.conv2.weight", "visual.layer3.8.conv2.weight_bias")
   relu_82 = Relu (getitem_255)
   getitem_258 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_82, "visual.layer3.8.conv3.weight", "visual.layer3.8.conv3.weight_bias")
   add_2156 = Add (getitem_258, relu_80)
   relu_83 = Relu (add_2156)
   getitem_261 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_83, "visual.layer3.9.conv1.weight", "visual.layer3.9.conv1.weight_bias")
   relu_84 = Relu (getitem_261)
   getitem_264 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_84, "visual.layer3.9.conv2.weight", "visual.layer3.9.conv2.weight_bias")
   relu_85 = Relu (getitem_264)
   getitem_267 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_85, "visual.layer3.9.conv3.weight", "visual.layer3.9.conv3.weight_bias")
   add_2232 = Add (getitem_267, relu_83)
   relu_86 = Relu (add_2232)
   getitem_270 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_86, "visual.layer3.10.conv1.weight", "visual.layer3.10.conv1.weight_bias")
   relu_87 = Relu (getitem_270)
   getitem_273 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_87, "visual.layer3.10.conv2.weight", "visual.layer3.10.conv2.weight_bias")
   relu_88 = Relu (getitem_273)
   getitem_276 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_88, "visual.layer3.10.conv3.weight", "visual.layer3.10.conv3.weight_bias")
   add_2308 = Add (getitem_276, relu_86)
   relu_89 = Relu (add_2308)
   getitem_279 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_89, "visual.layer3.11.conv1.weight", "visual.layer3.11.conv1.weight_bias")
   relu_90 = Relu (getitem_279)
   getitem_282 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_90, "visual.layer3.11.conv2.weight", "visual.layer3.11.conv2.weight_bias")
   relu_91 = Relu (getitem_282)
   getitem_285 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_91, "visual.layer3.11.conv3.weight", "visual.layer3.11.conv3.weight_bias")
   add_2384 = Add (getitem_285, relu_89)
   relu_92 = Relu (add_2384)
   getitem_288 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_92, "visual.layer3.12.conv1.weight", "visual.layer3.12.conv1.weight_bias")
   relu_93 = Relu (getitem_288)
   getitem_291 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_93, "visual.layer3.12.conv2.weight", "visual.layer3.12.conv2.weight_bias")
   relu_94 = Relu (getitem_291)
   getitem_294 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_94, "visual.layer3.12.conv3.weight", "visual.layer3.12.conv3.weight_bias")
   add_2460 = Add (getitem_294, relu_92)
   relu_95 = Relu (add_2460)
   getitem_297 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_95, "visual.layer3.13.conv1.weight", "visual.layer3.13.conv1.weight_bias")
   relu_96 = Relu (getitem_297)
   getitem_300 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_96, "visual.layer3.13.conv2.weight", "visual.layer3.13.conv2.weight_bias")
   relu_97 = Relu (getitem_300)
   getitem_303 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_97, "visual.layer3.13.conv3.weight", "visual.layer3.13.conv3.weight_bias")
   add_2536 = Add (getitem_303, relu_95)
   relu_98 = Relu (add_2536)
   getitem_306 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_98, "visual.layer3.14.conv1.weight", "visual.layer3.14.conv1.weight_bias")
   relu_99 = Relu (getitem_306)
   getitem_309 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_99, "visual.layer3.14.conv2.weight", "visual.layer3.14.conv2.weight_bias")
   relu_100 = Relu (getitem_309)
   getitem_312 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_100, "visual.layer3.14.conv3.weight", "visual.layer3.14.conv3.weight_bias")
   add_2612 = Add (getitem_312, relu_98)
   relu_101 = Relu (add_2612)
   getitem_315 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_101, "visual.layer3.15.conv1.weight", "visual.layer3.15.conv1.weight_bias")
   relu_102 = Relu (getitem_315)
   getitem_318 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_102, "visual.layer3.15.conv2.weight", "visual.layer3.15.conv2.weight_bias")
   relu_103 = Relu (getitem_318)
   getitem_321 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_103, "visual.layer3.15.conv3.weight", "visual.layer3.15.conv3.weight_bias")
   add_2688 = Add (getitem_321, relu_101)
   relu_104 = Relu (add_2688)
   getitem_324 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_104, "visual.layer3.16.conv1.weight", "visual.layer3.16.conv1.weight_bias")
   relu_105 = Relu (getitem_324)
   getitem_327 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_105, "visual.layer3.16.conv2.weight", "visual.layer3.16.conv2.weight_bias")
   relu_106 = Relu (getitem_327)
   getitem_330 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_106, "visual.layer3.16.conv3.weight", "visual.layer3.16.conv3.weight_bias")
   add_2764 = Add (getitem_330, relu_104)
   relu_107 = Relu (add_2764)
   getitem_333 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_107, "visual.layer3.17.conv1.weight", "visual.layer3.17.conv1.weight_bias")
   relu_108 = Relu (getitem_333)
   getitem_336 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_108, "visual.layer3.17.conv2.weight", "visual.layer3.17.conv2.weight_bias")
   relu_109 = Relu (getitem_336)
   getitem_339 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_109, "visual.layer3.17.conv3.weight", "visual.layer3.17.conv3.weight_bias")
   add_2840 = Add (getitem_339, relu_107)
   relu_110 = Relu (add_2840)
   getitem_342 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_110, "visual.layer3.18.conv1.weight", "visual.layer3.18.conv1.weight_bias")
   relu_111 = Relu (getitem_342)
   getitem_345 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_111, "visual.layer3.18.conv2.weight", "visual.layer3.18.conv2.weight_bias")
   relu_112 = Relu (getitem_345)
   getitem_348 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_112, "visual.layer3.18.conv3.weight", "visual.layer3.18.conv3.weight_bias")
   add_2916 = Add (getitem_348, relu_110)
   relu_113 = Relu (add_2916)
   getitem_351 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_113, "visual.layer3.19.conv1.weight", "visual.layer3.19.conv1.weight_bias")
   relu_114 = Relu (getitem_351)
   getitem_354 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_114, "visual.layer3.19.conv2.weight", "visual.layer3.19.conv2.weight_bias")
   relu_115 = Relu (getitem_354)
   getitem_357 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_115, "visual.layer3.19.conv3.weight", "visual.layer3.19.conv3.weight_bias")
   add_2992 = Add (getitem_357, relu_113)
   relu_116 = Relu (add_2992)
   getitem_360 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_116, "visual.layer3.20.conv1.weight", "visual.layer3.20.conv1.weight_bias")
   relu_117 = Relu (getitem_360)
   getitem_363 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_117, "visual.layer3.20.conv2.weight", "visual.layer3.20.conv2.weight_bias")
   relu_118 = Relu (getitem_363)
   getitem_366 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_118, "visual.layer3.20.conv3.weight", "visual.layer3.20.conv3.weight_bias")
   add_3068 = Add (getitem_366, relu_116)
   relu_119 = Relu (add_3068)
   getitem_369 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_119, "visual.layer3.21.conv1.weight", "visual.layer3.21.conv1.weight_bias")
   relu_120 = Relu (getitem_369)
   getitem_372 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_120, "visual.layer3.21.conv2.weight", "visual.layer3.21.conv2.weight_bias")
   relu_121 = Relu (getitem_372)
   getitem_375 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_121, "visual.layer3.21.conv3.weight", "visual.layer3.21.conv3.weight_bias")
   add_3144 = Add (getitem_375, relu_119)
   relu_122 = Relu (add_3144)
   getitem_378 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_122, "visual.layer3.22.conv1.weight", "visual.layer3.22.conv1.weight_bias")
   relu_123 = Relu (getitem_378)
   getitem_381 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_123, "visual.layer3.22.conv2.weight", "visual.layer3.22.conv2.weight_bias")
   relu_124 = Relu (getitem_381)
   getitem_384 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_124, "visual.layer3.22.conv3.weight", "visual.layer3.22.conv3.weight_bias")
   add_3220 = Add (getitem_384, relu_122)
   relu_125 = Relu (add_3220)
   getitem_387 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_125, "visual.layer3.23.conv1.weight", "visual.layer3.23.conv1.weight_bias")
   relu_126 = Relu (getitem_387)
   getitem_390 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_126, "visual.layer3.23.conv2.weight", "visual.layer3.23.conv2.weight_bias")
   relu_127 = Relu (getitem_390)
   getitem_393 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_127, "visual.layer3.23.conv3.weight", "visual.layer3.23.conv3.weight_bias")
   add_3296 = Add (getitem_393, relu_125)
   relu_128 = Relu (add_3296)
   getitem_396 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_128, "visual.layer3.24.conv1.weight", "visual.layer3.24.conv1.weight_bias")
   relu_129 = Relu (getitem_396)
   getitem_399 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_129, "visual.layer3.24.conv2.weight", "visual.layer3.24.conv2.weight_bias")
   relu_130 = Relu (getitem_399)
   getitem_402 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_130, "visual.layer3.24.conv3.weight", "visual.layer3.24.conv3.weight_bias")
   add_3372 = Add (getitem_402, relu_128)
   relu_131 = Relu (add_3372)
   getitem_405 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_131, "visual.layer3.25.conv1.weight", "visual.layer3.25.conv1.weight_bias")
   relu_132 = Relu (getitem_405)
   getitem_408 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_132, "visual.layer3.25.conv2.weight", "visual.layer3.25.conv2.weight_bias")
   relu_133 = Relu (getitem_408)
   getitem_411 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_133, "visual.layer3.25.conv3.weight", "visual.layer3.25.conv3.weight_bias")
   add_3448 = Add (getitem_411, relu_131)
   relu_134 = Relu (add_3448)
   getitem_414 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_134, "visual.layer3.26.conv1.weight", "visual.layer3.26.conv1.weight_bias")
   relu_135 = Relu (getitem_414)
   getitem_417 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_135, "visual.layer3.26.conv2.weight", "visual.layer3.26.conv2.weight_bias")
   relu_136 = Relu (getitem_417)
   getitem_420 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_136, "visual.layer3.26.conv3.weight", "visual.layer3.26.conv3.weight_bias")
   add_3524 = Add (getitem_420, relu_134)
   relu_137 = Relu (add_3524)
   getitem_423 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_137, "visual.layer3.27.conv1.weight", "visual.layer3.27.conv1.weight_bias")
   relu_138 = Relu (getitem_423)
   getitem_426 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_138, "visual.layer3.27.conv2.weight", "visual.layer3.27.conv2.weight_bias")
   relu_139 = Relu (getitem_426)
   getitem_429 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_139, "visual.layer3.27.conv3.weight", "visual.layer3.27.conv3.weight_bias")
   add_3600 = Add (getitem_429, relu_137)
   relu_140 = Relu (add_3600)
   getitem_432 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_140, "visual.layer3.28.conv1.weight", "visual.layer3.28.conv1.weight_bias")
   relu_141 = Relu (getitem_432)
   getitem_435 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_141, "visual.layer3.28.conv2.weight", "visual.layer3.28.conv2.weight_bias")
   relu_142 = Relu (getitem_435)
   getitem_438 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_142, "visual.layer3.28.conv3.weight", "visual.layer3.28.conv3.weight_bias")
   add_3676 = Add (getitem_438, relu_140)
   relu_143 = Relu (add_3676)
   getitem_441 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_143, "visual.layer3.29.conv1.weight", "visual.layer3.29.conv1.weight_bias")
   relu_144 = Relu (getitem_441)
   getitem_444 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_144, "visual.layer3.29.conv2.weight", "visual.layer3.29.conv2.weight_bias")
   relu_145 = Relu (getitem_444)
   getitem_447 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_145, "visual.layer3.29.conv3.weight", "visual.layer3.29.conv3.weight_bias")
   add_3752 = Add (getitem_447, relu_143)
   relu_146 = Relu (add_3752)
   getitem_450 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_146, "visual.layer3.30.conv1.weight", "visual.layer3.30.conv1.weight_bias")
   relu_147 = Relu (getitem_450)
   getitem_453 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_147, "visual.layer3.30.conv2.weight", "visual.layer3.30.conv2.weight_bias")
   relu_148 = Relu (getitem_453)
   getitem_456 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_148, "visual.layer3.30.conv3.weight", "visual.layer3.30.conv3.weight_bias")
   add_3828 = Add (getitem_456, relu_146)
   relu_149 = Relu (add_3828)
   getitem_459 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_149, "visual.layer3.31.conv1.weight", "visual.layer3.31.conv1.weight_bias")
   relu_150 = Relu (getitem_459)
   getitem_462 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_150, "visual.layer3.31.conv2.weight", "visual.layer3.31.conv2.weight_bias")
   relu_151 = Relu (getitem_462)
   getitem_465 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_151, "visual.layer3.31.conv3.weight", "visual.layer3.31.conv3.weight_bias")
   add_3904 = Add (getitem_465, relu_149)
   relu_152 = Relu (add_3904)
   getitem_468 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_152, "visual.layer3.32.conv1.weight", "visual.layer3.32.conv1.weight_bias")
   relu_153 = Relu (getitem_468)
   getitem_471 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_153, "visual.layer3.32.conv2.weight", "visual.layer3.32.conv2.weight_bias")
   relu_154 = Relu (getitem_471)
   getitem_474 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_154, "visual.layer3.32.conv3.weight", "visual.layer3.32.conv3.weight_bias")
   add_3980 = Add (getitem_474, relu_152)
   relu_155 = Relu (add_3980)
   getitem_477 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_155, "visual.layer3.33.conv1.weight", "visual.layer3.33.conv1.weight_bias")
   relu_156 = Relu (getitem_477)
   getitem_480 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_156, "visual.layer3.33.conv2.weight", "visual.layer3.33.conv2.weight_bias")
   relu_157 = Relu (getitem_480)
   getitem_483 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_157, "visual.layer3.33.conv3.weight", "visual.layer3.33.conv3.weight_bias")
   add_4056 = Add (getitem_483, relu_155)
   relu_158 = Relu (add_4056)
   getitem_486 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_158, "visual.layer3.34.conv1.weight", "visual.layer3.34.conv1.weight_bias")
   relu_159 = Relu (getitem_486)
   getitem_489 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_159, "visual.layer3.34.conv2.weight", "visual.layer3.34.conv2.weight_bias")
   relu_160 = Relu (getitem_489)
   getitem_492 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_160, "visual.layer3.34.conv3.weight", "visual.layer3.34.conv3.weight_bias")
   add_4132 = Add (getitem_492, relu_158)
   relu_161 = Relu (add_4132)
   getitem_495 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_161, "visual.layer3.35.conv1.weight", "visual.layer3.35.conv1.weight_bias")
   relu_162 = Relu (getitem_495)
   getitem_498 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_162, "visual.layer3.35.conv2.weight", "visual.layer3.35.conv2.weight_bias")
   relu_163 = Relu (getitem_498)
   getitem_501 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_163, "visual.layer3.35.conv3.weight", "visual.layer3.35.conv3.weight_bias")
   add_4208 = Add (getitem_501, relu_161)
   relu_164 = Relu (add_4208)
   getitem_504 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_164, "visual.layer4.0.conv1.weight", "visual.layer4.0.conv1.weight_bias")
   relu_165 = Relu (getitem_504)
   getitem_507 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_165, "visual.layer4.0.conv2.weight", "visual.layer4.0.conv2.weight_bias")
   relu_166 = Relu (getitem_507)
   avg_pool2d_6 = AveragePool <auto_pad: string = "NOTSET", ceil_mode: int = 0, count_include_pad: int = 1, kernel_shape: ints = [2, 2], pads: ints = [0, 0, 0, 0], strides: ints = [2, 2]> (relu_166)
   getitem_510 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (avg_pool2d_6, "visual.layer4.0.conv3.weight", "visual.layer4.0.conv3.weight_bias")
   avg_pool2d_7 = AveragePool <auto_pad: string = "NOTSET", ceil_mode: int = 0, count_include_pad: int = 1, kernel_shape: ints = [2, 2], pads: ints = [0, 0, 0, 0], strides: ints = [2, 2]> (relu_164)
   getitem_513 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (avg_pool2d_7, "visual.layer4.0.downsample.0.weight", "visual.layer4.0.downsample.0.weight_bias")
   add_4304 = Add (getitem_510, getitem_513)
   relu_167 = Relu (add_4304)
   getitem_516 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_167, "visual.layer4.1.conv1.weight", "visual.layer4.1.conv1.weight_bias")
   relu_168 = Relu (getitem_516)
   getitem_519 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_168, "visual.layer4.1.conv2.weight", "visual.layer4.1.conv2.weight_bias")
   relu_169 = Relu (getitem_519)
   getitem_522 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_169, "visual.layer4.1.conv3.weight", "visual.layer4.1.conv3.weight_bias")
   add_4380 = Add (getitem_522, relu_167)
   relu_170 = Relu (add_4380)
   getitem_525 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_170, "visual.layer4.2.conv1.weight", "visual.layer4.2.conv1.weight_bias")
   relu_171 = Relu (getitem_525)
   getitem_528 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_171, "visual.layer4.2.conv2.weight", "visual.layer4.2.conv2.weight_bias")
   relu_172 = Relu (getitem_528)
   getitem_531 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_172, "visual.layer4.2.conv3.weight", "visual.layer4.2.conv3.weight_bias")
   add_4456 = Add (getitem_531, relu_170)
   relu_173 = Relu (add_4456)
   getitem_534 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_173, "visual.layer4.3.conv1.weight", "visual.layer4.3.conv1.weight_bias")
   relu_174 = Relu (getitem_534)
   getitem_537 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_174, "visual.layer4.3.conv2.weight", "visual.layer4.3.conv2.weight_bias")
   relu_175 = Relu (getitem_537)
   getitem_540 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_175, "visual.layer4.3.conv3.weight", "visual.layer4.3.conv3.weight_bias")
   add_4532 = Add (getitem_540, relu_173)
   relu_176 = Relu (add_4532)
   getitem_543 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_176, "visual.layer4.4.conv1.weight", "visual.layer4.4.conv1.weight_bias")
   relu_177 = Relu (getitem_543)
   getitem_546 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_177, "visual.layer4.4.conv2.weight", "visual.layer4.4.conv2.weight_bias")
   relu_178 = Relu (getitem_546)
   getitem_549 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_178, "visual.layer4.4.conv3.weight", "visual.layer4.4.conv3.weight_bias")
   add_4608 = Add (getitem_549, relu_176)
   relu_179 = Relu (add_4608)
   getitem_552 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_179, "visual.layer4.5.conv1.weight", "visual.layer4.5.conv1.weight_bias")
   relu_180 = Relu (getitem_552)
   getitem_555 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_180, "visual.layer4.5.conv2.weight", "visual.layer4.5.conv2.weight_bias")
   relu_181 = Relu (getitem_555)
   getitem_558 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_181, "visual.layer4.5.conv3.weight", "visual.layer4.5.conv3.weight_bias")
   add_4684 = Add (getitem_558, relu_179)
   relu_182 = Relu (add_4684)
   getitem_561 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_182, "visual.layer4.6.conv1.weight", "visual.layer4.6.conv1.weight_bias")
   relu_183 = Relu (getitem_561)
   getitem_564 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_183, "visual.layer4.6.conv2.weight", "visual.layer4.6.conv2.weight_bias")
   relu_184 = Relu (getitem_564)
   getitem_567 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_184, "visual.layer4.6.conv3.weight", "visual.layer4.6.conv3.weight_bias")
   add_4760 = Add (getitem_567, relu_182)
   relu_185 = Relu (add_4760)
   getitem_570 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_185, "visual.layer4.7.conv1.weight", "visual.layer4.7.conv1.weight_bias")
   relu_186 = Relu (getitem_570)
   getitem_573 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_186, "visual.layer4.7.conv2.weight", "visual.layer4.7.conv2.weight_bias")
   relu_187 = Relu (getitem_573)
   getitem_576 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_187, "visual.layer4.7.conv3.weight", "visual.layer4.7.conv3.weight_bias")
   add_4836 = Add (getitem_576, relu_185)
   relu_188 = Relu (add_4836)
   getitem_579 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_188, "visual.layer4.8.conv1.weight", "visual.layer4.8.conv1.weight_bias")
   relu_189 = Relu (getitem_579)
   getitem_582 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_189, "visual.layer4.8.conv2.weight", "visual.layer4.8.conv2.weight_bias")
   relu_190 = Relu (getitem_582)
   getitem_585 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_190, "visual.layer4.8.conv3.weight", "visual.layer4.8.conv3.weight_bias")
   add_4912 = Add (getitem_585, relu_188)
   relu_191 = Relu (add_4912)
   getitem_588 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_191, "visual.layer4.9.conv1.weight", "visual.layer4.9.conv1.weight_bias")
   relu_192 = Relu (getitem_588)
   getitem_591 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [1, 1, 1, 1], strides: ints = [1, 1]> (relu_192, "visual.layer4.9.conv2.weight", "visual.layer4.9.conv2.weight_bias")
   relu_193 = Relu (getitem_591)
   getitem_594 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_193, "visual.layer4.9.conv3.weight", "visual.layer4.9.conv3.weight_bias")
   add_4988 = Add (getitem_594, relu_191)
   relu_194 = Relu (add_4988)
   view_2 = Reshape <allowzero: int = 1> (relu_194, view_2_target)
   permute_1 = Transpose <perm: ints = [2, 0, 1]> (view_2)
   [node_mean] mean = ReduceMean <keepdims: int = 1, noop_with_empty_axes: int = 0> (permute_1, val_0)
   [node_cat] cat = Concat <axis: int = 0> (mean, permute_1)
   [node_unsqueeze] unsqueeze = Unsqueeze ("visual.attnpool.positional_embedding", val_3)
   add_5018 = Add (cat, unsqueeze)
   split_split_0, split_split_1, split_split_2 = Split <axis: int = 0, num_outputs: int = 3> (cat_1)
   node_scaled_dot_product_attention_q_row = Slice (add_5018, val_0, val_3, val_0)
   val_7 = MatMul (node_scaled_dot_product_attention_q_row, val_4)
   [node_linear] linear = Add (val_7, split_split_0)
   val_8 = MatMul (add_5018, val_5)
   linear_1 = Add (val_8, split_split_1)
   val_9 = MatMul (add_5018, val_6)
   linear_2 = Add (val_9, split_split_2)
   view_3 = Reshape <allowzero: int = 1> (linear, node_scaled_dot_product_attention_q_unpack_1)
   [node_transpose] transpose = Transpose <perm: ints = [1, 0, 2]> (view_3)
   view_4 = Reshape <allowzero: int = 1> (linear_1, view_4_target)
   transpose_1 = Transpose <perm: ints = [1, 0, 2]> (view_4)
   view_5 = Reshape <allowzero: int = 1> (linear_2, view_4_target)
   transpose_2 = Transpose <perm: ints = [1, 0, 2]> (view_5)
   view_6 = Reshape <allowzero: int = 1> (transpose, node_scaled_dot_product_attention_q_pack_1)
   view_7 = Reshape <allowzero: int = 1> (transpose_1, view_7_target)
   view_8 = Reshape <allowzero: int = 1> (transpose_2, view_7_target)
   [node_scaled_dot_product_attention] scaled_dot_product_attention = Attention <is_causal: int = 0, qk_matmul_output_mode: int = 0, softcap: float = 0> (view_6, view_7, view_8)
   permute_2 = Transpose <perm: ints = [2, 0, 1, 3]> (scaled_dot_product_attention)
   view_9 = Reshape <allowzero: int = 1> (permute_2, view_9_target)
   linear_3 = Gemm <alpha: float = 1, beta: float = 1, transA: int = 0, transB: int = 1> (view_9, "visual.attnpool.c_proj.weight", "visual.attnpool.c_proj.bias")
   view_10 = Reshape <allowzero: int = 1> (linear_3, node_scaled_dot_product_attention_out_1)
   select = Squeeze (view_10, val_0)
   [node_linalg_vector_norm] linalg_vector_norm = ReduceL2 <keepdims: int = 1, noop_with_empty_axes: int = 0> (select, val_1)
   [node_clamp_min] clamp_min = Clip (linalg_vector_norm, val_2)
   [node_div] image_embedding = Div (select, clamp_min)
}

weights:
cat_1 FLOAT[12288] f3bc9fd4a15d
image_shift FLOAT[3] 2f7a50e604ad
node_scaled_dot_product_attention_out_1 INT64[3] 7162728d1394
node_scaled_dot_product_attention_q_pack_1 INT64[4] dc6cfe24d500
node_scaled_dot_product_attention_q_unpack_1 INT64[3] cfe34a386daf
val_0 INT64[1] af5570f5a181
val_1 INT64[1] 12a3ae445661
val_2 FLOAT[] 6708d9be4956
val_3 INT64[1] 7c9fa136d441
val_4 FLOAT[4096,4096] a174248b2b19
val_5 FLOAT[4096,4096] 644df4b9439c
val_6 FLOAT[4096,4096] 7db20657060c
view_2_target INT64[3] 105ca240ffa2
view_4_target INT64[3] 73e5306dc236
view_7_target INT64[4] ad96207b8c08
view_9_target INT64[2] 96fb1613a890
visual.attnpool.c_proj.bias FLOAT[1024] 05f2bbc5ed03
visual.attnpool.c_proj.weight FLOAT[1024,4096] df14b8517bbf
visual.attnpool.positional_embedding FLOAT[197,4096] 6013dc5d2caa
visual.conv1.weight FLOAT[64,3,3,3] 310cd617cb81
visual.conv1.weight_bias FLOAT[64] 4e04008e909a
visual.conv2.weight FLOAT[64,64,3,3] 9f7b74dc85d8
visual.conv2.weight_bias FLOAT[64] 881654deb8f9
visual.conv3.weight FLOAT[128,64,3,3] dc643ff73edb
visual.conv3.weight_bias FLOAT[128] 0df01c31767e
visual.layer1.0.conv1.weight FLOAT[128,128,1,1] 21f7d7382001
visual.layer1.0.conv1.weight_bias FLOAT[128] 17b23d18c49f
visual.layer1.0.conv2.weight FLOAT[128,128,3,3] 53ea940dda32
visual.layer1.0.conv2.weight_bias FLOAT[128] ebd2d80a0beb
visual.layer1.0.conv3.weight FLOAT[512,128,1,1] fd936b7f6bea
visual.layer1.0.conv3.weight_bias FLOAT[512] b221ff1254d3
visual.layer1.0.downsample.0.weight FLOAT[512,128,1,1] 781ade65f808
visual.layer1.0.downsample.0.weight_bias FLOAT[512] 809ae89b6e30
visual.layer1.1.conv1.weight FLOAT[128,512,1,1] 3202f500c5ae
visual.layer1.1.conv1.weight_bias FLOAT[128] 5d8d1c15a3dd
visual.layer1.1.conv2.weight FLOAT[128,128,3,3] c3599a080bfd
visual.layer1.1.conv2.weight_bias FLOAT[128] d2b9e4dd182a
visual.layer1.1.conv3.weight FLOAT[512,128,1,1] 2df3d01506b3
visual.layer1.1.conv3.weight_bias FLOAT[512] 1cac115d608d
visual.layer1.2.conv1.weight FLOAT[128,512,1,1] a3649e19481c
visual.layer1.2.conv1.weight_bias FLOAT[128] 9eb9f024cf5a
visual.layer1.2.conv2.weight FLOAT[128,128,3,3] 15a71672d6c5
visual.layer1.2.conv2.weight_bias FLOAT[128] 08b3958ed811
visual.layer1.2.conv3.weight FLOAT[512,128,1,1] db2515da1b39
visual.layer1.2.conv3.weight_bias FLOAT[512] 2b5cd041a14e
visual.layer2.0.conv1.weight FLOAT[256,512,1,1] 64520745c178
visual.layer2.0.conv1.weight_bias FLOAT[256] ebc82c87edc5
visual.layer2.0.conv2.weight FLOAT[256,256,3,3] 13ba3086d0ef
visual.layer2.0.conv2.weight_bias FLOAT[256] e486736c1ddb
visual.layer2.0.conv3.weight FLOAT[1024,256,1,1] fbe7c9590f8b
visual.layer2.0.conv3.weight_bias FLOAT[1024] 0f7eae8c5541
visual.layer2.0.downsample.0.weight FLOAT[1024,512,1,1] 8a8383aa8bfc
visual.layer2.0.downsample.0.weight_bias FLOAT[1024] 851e43282ea2
visual.layer2.1.conv1.weight FLOAT[256,1024,1,1] dbb9879c4227
visual.layer2.1.conv1.weight_bias FLOAT[256] 5f86864604a2
visual.layer2.1.conv2.weight FLOAT[256,256,3,3] 798c8d258778
visual.layer2.1.conv2.weight_bias FLOAT[256] c58bf30be273
visual.layer2.1.conv3.weight FLOAT[1024,256,1,1] 40189856d0e5
visual.layer2.1.conv3.weight_bias FLOAT[1024] b3678960fab4
visual.layer2.10.conv1.weight FLOAT[256,1024,1,1] 9be93450a313
visual.layer2.10.conv1.weight_bias FLOAT[256] 42bb9972b8d0
visual.layer2.10.conv2.weight FLOAT[256,256,3,3] bd95c8c4996a
visual.layer2.10.conv2.weight_bias FLOAT[256] 4560cf297220
visual.layer2.10.conv3.weight FLOAT[1024,256,1,1] ad4b5fc7529a
visual.layer2.10.conv3.weight_bias FLOAT[1024] 0f6ab91b0613
visual.layer2.11.conv1.weight FLOAT[256,1024,1,1] 1ddc0f3f86a0
visual.layer2.11.conv1.weight_bias FLOAT[256] 325bd608dd43
visual.layer2.11.conv2.weight FLOAT[256,256,3,3] a1698c2e1fee
visual.layer2.11.conv2.weight_bias FLOAT[256] 68dfe94e8a7a
visual.layer2.11.conv3.weight FLOAT[1024,256,1,1] 364b3b5f43b0
visual.layer2.11.conv3.weight_bias FLOAT[1024] 7229c27b4a41
visual.layer2.12.conv1.weight FLOAT[256,1024,1,1] 266e8c37bc80
visual.layer2.12.conv1.weight_bias FLOAT[256] 2af87281952b
visual.layer2.12.conv2.weight FLOAT[256,256,3,3] 309119502dc0
visual.layer2.12.conv2.weight_bias FLOAT[256] e6e62b389412
visual.layer2.12.conv3.weight FLOAT[1024,256,1,1] 832d0663f98a
visual.layer2.12.conv3.weight_bias FLOAT[1024] 3e5c7237eb5a
visual.layer2.13.conv1.weight FLOAT[256,1024,1,1] 8f54f6a4e24f
visual.layer2.13.conv1.weight_bias FLOAT[256] 3cba797363a6
visual.layer2.13.conv2.weight FLOAT[256,256,3,3] 7e3a4995b7c9
visual.layer2.13.conv2.weight_bias FLOAT[256] faa025f5b832
visual.layer2.13.conv3.weight FLOAT[1024,256,1,1] d0b69d876976
visual.layer2.13.conv3.weight_bias FLOAT[1024] d44a60ca00d9
visual.layer2.14.conv1.weight FLOAT[256,1024,1,1] 78a5e9d35daa
visual.layer2.14.conv1.weight_bias FLOAT[256] ead13664c089
visual.layer2.14.conv2.weight FLOAT[256,256,3,3] fb1a1236eaa2
visual.layer2.14.conv2.weight_bias FLOAT[256] 4ac7ea49634f
visual.layer2.14.conv3.weight FLOAT[1024,256,1,1] 5d222f064da6
visual.layer2.14.conv3.weight_bias FLOAT[1024] 2878e9215521
visual.layer2.2.conv1.weight FLOAT[256,1024,1,1] 79a1c0c7e85b
visual.layer2.2.conv1.weight_bias FLOAT[256] f9346ee3d55f
visual.layer2.2.conv2.weight FLOAT[256,256,3,3] db0a137fa711
visual.layer2.2.conv2.weight_bias FLOAT[256] 9154512dcb74
visual.layer2.2.conv3.weight FLOAT[1024,256,1,1] 1856029aa798
visual.layer2.2.conv3.weight_bias FLOAT[1024] 86db72314112
visual.layer2.3.conv1.weight FLOAT[256,1024,1,1] 7fabef5b844c
visual.layer2.3.conv1.weight_bias FLOAT[256] e95f0a6f7d3d
visual.layer2.3.conv2.weight FLOAT[256,256,3,3] cee7416f44bd
visual.layer2.3.conv2.weight_bias FLOAT[256] 6316705d42d6
visual.layer2.3.conv3.weight FLOAT[1024,256,1,1] 594c53a9e603
visual.layer2.3.conv3.weight_bias FLOAT[1024] 060fb2e3c71e
visual.layer2.4.conv1.weight FLOAT[256,1024,1,1] c635fc69eb75
visual.layer2.4.conv1.weight_bias FLOAT[256] 569e709716b6
visual.layer2.4.conv2.weight FLOAT[256,256,3,3] bde836fe800a
visual.layer2.4.conv2.weight_bias FLOAT[256] 76b372cf0349
visual.layer2.4.conv3.weight FLOAT[1024,256,1,1] c38d01674759
visual.layer2.4.conv3.weight_bias FLOAT[1024] 77ff98bd1f67
visual.layer2.5.conv1.weight FLOAT[256,1024,1,1] 8943bcbfe512
visual.layer2.5.conv1.weight_bias FLOAT[256] 3b2d7e7314c8
visual.layer2.5.conv2.weight FLOAT[256,256,3,3] 785aec4c720a
visual.layer2.5.conv2.weight_bias FLOAT[256] 95e27259c32f
visual.layer2.5.conv3.weight FLOAT[1024,256,1,1] 103f99d9edc3
visual.layer2.5.conv3.weight_bias FLOAT[1024] 38b6a5a502a2
visual.layer2.6.conv1.weight FLOAT[256,1024,1,1] 9a8fd4cb9751
visual.layer2.6.conv1.weight_bias FLOAT[256] 139b7c86c99d
visual.layer2.6.conv2.weight FLOAT[256,256,3,3] 5af0092c386a
visual.layer2.6.conv2.weight_bias FLOAT[256] 85d2cc2835a6
visual.layer2.6.conv3.weight FLOAT[1024,256,1,1] b7f53a702837
visual.layer2.6.conv3.weight_bias FLOAT[1024] 2ddbdd7745e0
visual.layer2.7.conv1.weight FLOAT[256,1024,1,1] 9b925b4d3ef4
visual.layer2.7.conv1.weight_bias FLOAT[256] a2c2710aba53
visual.layer2.7.conv2.weight FLOAT[256,256,3,3] da6b9495e6d9
visual.layer2.7.conv2.weight_bias FLOAT[256] ff7cbc192298
visual.layer2.7.conv3.weight FLOAT[1024,256,1,1] eaf979a682c3
visual.layer2.7.conv3.weight_bias FLOAT[1024] c57a8f89cef9
visual.layer2.8.conv1.weight FLOAT[256,1024,1,1] b745da828d19
visual.layer2.8.conv1.weight_bias FLOAT[256] 37f60cf96ad3
visual.layer2.8.conv2.weight FLOAT[256,256,3,3] f14a573becc6
visual.layer2.8.conv2.weight_bias FLOAT[256] 0cf289fdbd3a
visual.layer2.8.conv3.weight FLOAT[1024,256,1,1] 2c1dc7e6c216
visual.layer2.8.conv3.weight_bias FLOAT[1024] 784fb3a24397
visual.layer2.9.conv1.weight FLOAT[256,1024,1,1] e404448e70a3
visual.layer2.9.conv1.weight_bias FLOAT[256] 2a87230099a2
visual.layer2.9.conv2.weight FLOAT[256,256,3,3] 305803625a4b
visual.layer2.9.conv2.weight_bias FLOAT[256] e79c367e4603
visual.layer2.9.conv3.weight FLOAT[1024,256,1,1] 121c430ac208
visual.layer2.9.conv3.weight_bias FLOAT[1024] 8deebad2bb54
visual.layer3.0.conv1.weight FLOAT[512,1024,1,1] 794bf934252c
visual.layer3.0.conv1.weight_bias FLOAT[512] aab8eefd278a
visual.layer3.0.conv2.weight FLOAT[512,512,3,3] a67a2d887157
visual.layer3.0.conv2.weight_bias FLOAT[512] 1b2cbd09ccfc
visual.layer3.0.conv3.weight FLOAT[2048,512,1,1] e70019420aa6
visual.layer3.0.conv3.weight_bias FLOAT[2048] 6ae6376c208f
visual.layer3.0.downsample.0.weight FLOAT[2048,1024,1,1] 950088c3070c
visual.layer3.0.downsample.0.weight_bias FLOAT[2048] b57ea2dd8c31
visual.layer3.1.conv1.weight FLOAT[512,2048,1,1] 4b868ef6efe3
visual.layer3.1.conv1.weight_bias FLOAT[512] 5b062a0be3eb
visual.layer3.1.conv2.weight FLOAT[512,512,3,3] 904448cc4e92
visual.layer3.1.conv2.weight_bias FLOAT[512] 25d808df91c9
visual.layer3.1.conv3.weight FLOAT[2048,512,1,1] 20e25abba10d
visual.layer3.1.conv3.weight_bias FLOAT[2048] 6a80f0ac3ed9
visual.layer3.10.conv1.weight FLOAT[512,2048,1,1] 388576e29c80
visual.layer3.10.conv1.weight_bias FLOAT[512] 5d9f5b00ac80
visual.layer3.10.conv2.weight FLOAT[512,512,3,3] 974d92978dda
visual.layer3.10.conv2.weight_bias FLOAT[512] 5bf5b5f376b7
visual.layer3.10.conv3.weight FLOAT[2048,512,1,1] 9e8d7a07bbd0
visual.layer3.10.conv3.weight_bias FLOAT[2048] 202995baed92
visual.layer3.11.conv1.weight FLOAT[512,2048,1,1] e0fd748bfb62
visual.layer3.11.conv1.weight_bias FLOAT[512] 9f1b2c0bbfc5
visual.layer3.11.conv2.weight FLOAT[512,512,3,3] be7616421bc5
visual.layer3.11.conv2.weight_bias FLOAT[512] 3444741a5ee4
visual.layer3.11.conv3.weight FLOAT[2048,512,1,1] 6c1367aea6a0
visual.layer3.11.conv3.weight_bias FLOAT[2048] 7637e7778cd9
visual.layer3.12.conv1.weight FLOAT[512,2048,1,1] 931037e771c0
visual.layer3.12.conv1.weight_bias FLOAT[512] db313553d771
visual.layer3.12.conv2.weight FLOAT[512,512,3,3] 5c57e34cd9e9
visual.layer3.12.conv2.weight_bias FLOAT[512] 01dd2a466512
visual.layer3.12.conv3.weight FLOAT[2048,512,1,1] 3812c77a5b7c
visual.layer3.12.conv3.weight_bias FLOAT[2048] 4f65f37e17ca
visual.layer3.13.conv1.weight FLOAT[512,2048,1,1] 1f5367a7321e
visual.layer3.13.conv1.weight_bias FLOAT[512] 4354c6323c3a
visual.layer3.13.conv2.weight FLOAT[512,512,3,3] 2ef44478312f
visual.layer3.13.conv2.weight_bias FLOAT[512] cf506bd1ae12
visual.layer3.13.conv3.weight FLOAT[2048,512,1,1] 523e5cfca035
visual.layer3.13.conv3.weight_bias FLOAT[2048] 2a6c4b987522
visual.layer3.14.conv1.weight FLOAT[512,2048,1,1] c6825072c856
visual.layer3.14.conv1.weight_bias FLOAT[512] 4e095004b9f7
visual.layer3.14.conv2.weight FLOAT[512,512,3,3] 84fad0c6d422
visual.layer3.14.conv2.weight_bias FLOAT[512] 75690882afee
visual.layer3.14.conv3.weight FLOAT[2048,512,1,1] 5b89042b6812
visual.layer3.14.conv3.weight_bias FLOAT[2048] 78fa32695b7f
visual.layer3.15.conv1.weight FLOAT[512,2048,1,1] 9fd3880040ac
visual.layer3.15.conv1.weight_bias FLOAT[512] 39051ea0614a
visual.layer3.15.conv2.weight FLOAT[512,512,3,3] a9025835fa48
visual.layer3.15.conv2.weight_bias FLOAT[512] 009cb407cb87
visual.layer3.15.conv3.weight FLOAT[2048,512,1,1] cf7fa76f173f
visual.layer3.15.conv3.weight_bias FLOAT[2048] ea7390df8b41
visual.layer3.16.conv1.weight FLOAT[512,2048,1,1] 697958abf8e1
visual.layer3.16.conv1.weight_bias FLOAT[512] 20cfe1ede88f
visual.layer3.16.conv2.weight FLOAT[512,512,3,3] 7788dd3cb529
visual.layer3.16.conv2.weight_bias FLOAT[512] e5ec0c8250fd
visual.layer3.16.conv3.weight FLOAT[2048,512,1,1] 91109f29ddb7
visual.layer3.16.conv3.weight_bias FLOAT[2048] 101f50363a31
visual.layer3.17.conv1.weight FLOAT[512,2048,1,1] 38e467bb151a
visual.layer3.17.conv1.weight_bias FLOAT[512] 1f7007ca67e7
visual.layer3.17.conv2.weight FLOAT[512,512,3,3] 5842a8e931f2
visual.layer3.17.conv2.weight_bias FLOAT[512] c3af54d2a5a7
visual.layer3.17.conv3.weight FLOAT[2048,512,1,1] a78fa816cfd5
visual.layer3.17.conv3.weight_bias FLOAT[2048] 8e57006bc2af
visual.layer3.18.conv1.weight FLOAT[512,2048,1,1] b358e9b97a8f
visual.layer3.18.conv1.weight_bias FLOAT[512] 600c15bf0673
visual.layer3.18.conv2.weight FLOAT[512,512,3,3] c07e06037afd
visual.layer3.18.conv2.weight_bias FLOAT[512] 0a93bf76c800
visual.layer3.18.conv3.weight FLOAT[2048,512,1,1] 4590601a199c
visual.layer3.18.conv3.weight_bias FLOAT[2048] fc6a12819019
visual.layer3.19.conv1.weight FLOAT[512,2048,1,1] f418275fa3e6
visual.layer3.19.conv1.weight_bias FLOAT[512] cccd7f97f904
visual.layer3.19.conv2.weight FLOAT[512,512,3,3] d8d38951d053
visual.layer3.19.conv2.weight_bias FLOAT[512] 52b1d4008769
visual.layer3.19.conv3.weight FLOAT[2048,512,1,1] 0b64267638f2
visual.layer3.19.conv3.weight_bias FLOAT[2048] 17a1f9407fb6
visual.layer3.2.conv1.weight FLOAT[512,2048,1,1] 91a394d14f43
visual.layer3.2.conv1.weight_bias FLOAT[512] 6a29a78625d2
visual.layer3.2.conv2.weight FLOAT[512,512,3,3] b068fbee327f
visual.layer3.2.conv2.weight_bias FLOAT[512] c845b48bbf4b
visual.layer3.2.conv3.weight FLOAT[2048,512,1,1] 805985fbada9
visual.layer3.2.conv3.weight_bias FLOAT[2048] 2fb4fd62d33a
visual.layer3.20.conv1.weight FLOAT[512,2048,1,1] f77f144d5254
visual.layer3.20.conv1.weight_bias FLOAT[512] 1a7619dd997a
visual.layer3.20.conv2.weight FLOAT[512,512,3,3] 5cc72ecef495
visual.layer3.20.conv2.weight_bias FLOAT[512] 0d509c2fa6b2
visual.layer3.20.conv3.weight FLOAT[2048,512,1,1] a8693510c225
visual.layer3.20.conv3.weight_bias FLOAT[2048] 36d0b5b8f483
visual.layer3.21.conv1.weight FLOAT[512,2048,1,1] 1505fe0d0e3e
visual.layer3.21.conv1.weight_bias FLOAT[512] 8a2b58114ad9
visual.layer3.21.conv2.weight FLOAT[512,512,3,3] 8a1ea506e8d7
visual.layer3.21.conv2.weight_bias FLOAT[512] 493778c13c70
visual.layer3.21.conv3.weight FLOAT[2048,512,1,1] ede588a0a6e5
visual.layer3.21.conv3.weight_bias FLOAT[2048] b6e0e2221e14
visual.layer3.22.conv1.weight FLOAT[512,2048,1,1] 9bd0fbb28615
visual.layer3.22.conv1.weight_bias FLOAT[512] 42254183df32
visual.layer3.22.conv2.weight FLOAT[512,512,3,3] 3db9e702c096
visual.layer3.22.conv2.weight_bias FLOAT[512] 6eb31466641a
visual.layer3.22.conv3.weight FLOAT[2048,512,1,1] afe8d1931628
visual.layer3.22.conv3.weight_bias FLOAT[2048] 84db815b1094
visual.layer3.23.conv1.weight FLOAT[512,2048,1,1] 71b4020f06ab
visual.layer3.23.conv1.weight_bias FLOAT[512] cbfee46e94be
visual.layer3.23.conv2.weight FLOAT[512,512,3,3] ee6535a63ddb
visual.layer3.23.conv2.weight_bias FLOAT[512] 5cb2bfd4879d
visual.layer3.23.conv3.weight FLOAT[2048,512,1,1] 81c56e457b90
visual.layer3.23.conv3.weight_bias FLOAT[2048] cdab1a138f90
visual.layer3.24.conv1.weight FLOAT[512,2048,1,1] 6ac80a44f280
visual.layer3.24.conv1.weight_bias FLOAT[512] 7f3a0d12916d
visual.layer3.24.conv2.weight FLOAT[512,512,3,3] 1f406c8d0729
visual.layer3.24.conv2.weight_bias FLOAT[512] c4937bd08d3e
visual.layer3.24.conv3.weight FLOAT[2048,512,1,1] 3bfdaab1269d
visual.layer3.24.conv3.weight_bias FLOAT[2048] 9412da15b169
visual.layer3.25.conv1.weight FLOAT[512,2048,1,1] a417f70a41a4
visual.layer3.25.conv1.weight_bias FLOAT[512] d66debcfc20c
visual.layer3.25.conv2.weight FLOAT[512,512,3,3] 3dc69d19ec1a
visual.layer3.25.conv2.weight_bias FLOAT[512] f54a9486d4e0
visual.layer3.25.conv3.weight FLOAT[2048,512,1,1] e8c090db91fc
visual.layer3.25.conv3.weight_bias FLOAT[2048] bffcbddac646
visual.layer3.26.conv1.weight FLOAT[512,2048,1,1] cae683b8b5ca
visual.layer3.26.conv1.weight_bias FLOAT[512] 795132d624e7
visual.layer3.26.conv2.weight FLOAT[512,512,3,3] 228a678bdadb
visual.layer3.26.conv2.weight_bias FLOAT[512] 1d34ea73a41e
visual.layer3.26.conv3.weight FLOAT[2048,512,1,1] e634164279fe
visual.layer3.26.conv3.weight_bias FLOAT[2048] e741d3024f60
visual.layer3.27.conv1.weight FLOAT[512,2048,1,1] d5d380daa404
visual.layer3.27.conv1.weight_bias FLOAT[512] 4dcfd83586f2
visual.layer3.27.conv2.weight FLOAT[512,512,3,3] fda7d0230944
visual.layer3.27.conv2.weight_bias FLOAT[512] f0c6c017d635
visual.layer3.27.conv3.weight FLOAT[2048,512,1,1] 64624c0b5034
visual.layer3.27.conv3.weight_bias FLOAT[2048] bba687e3e5b4
visual.layer3.28.conv1.weight FLOAT[512,2048,1,1] a751be1b3585
visual.layer3.28.conv1.weight_bias FLOAT[512] 560d4e3b10dd
visual.layer3.28.conv2.weight FLOAT[512,512,3,3] 805a8d4f47e1
visual.layer3.28.conv2.weight_bias FLOAT[512] 2313e64c9cd5
visual.layer3.28.conv3.weight FLOAT[2048,512,1,1] 0b650d22403a
visual.layer3.28.conv3.weight_bias FLOAT[2048] 99fa4962d2d9
visual.layer3.29.conv1.weight FLOAT[512,2048,1,1] f05432226c15
visual.layer3.29.conv1.weight_bias FLOAT[512] af8f76f7c9a9
visual.layer3.29.conv2.weight FLOAT[512,512,3,3] ea2896d9074c
visual.layer3.29.conv2.weight_bias FLOAT[512] 9ab2b4109cca
visual.layer3.29.conv3.weight FLOAT[2048,512,1,1] 253b26c60e71
visual.layer3.29.conv3.weight_bias FLOAT[2048] eb6a808c1c3b
visual.layer3.3.conv1.weight FLOAT[512,2048,1,1] ac23f3d45137
visual.layer3.3.conv1.weight_bias FLOAT[512] 80ff032e5710
visual.layer3.3.conv2.weight FLOAT[512,512,3,3] 512ef35889a5
visual.layer3.3.conv2.weight_bias FLOAT[512] af6600a507a5
visual.layer3.3.conv3.weight FLOAT[2048,512,1,1] 9054dd4d9cf4
visual.layer3.3.conv3.weight_bias FLOAT[2048] 2f699c16b995
visual.layer3.30.conv1.weight FLOAT[512,2048,1,1] 7cf0d6bf315e
visual.layer3.30.conv1.weight_bias FLOAT[512] 09a0c7b5e699
visual.layer3.30.conv2.weight FLOAT[512,512,3,3] 3bf9b01f794e
visual.layer3.30.conv2.weight_bias FLOAT[512] 5d29c6e75721
visual.layer3.30.conv3.weight FLOAT[2048,512,1,1] 0db4bb97f19f
visual.layer3.30.conv3.weight_bias FLOAT[2048] 5e2c8200134e
visual.layer3.31.conv1.weight FLOAT[512,2048,1,1] 4d0b79112a64
visual.layer3.31.conv1.weight_bias FLOAT[512] 9f6b44b453e1
visual.layer3.31.conv2.weight FLOAT[512,512,3,3] 5c6ad750e405
visual.layer3.31.conv2.weight_bias FLOAT[512] 7d6dae9ef449
visual.layer3.31.conv3.weight FLOAT[2048,512,1,1] 68d5d99f4762
visual.layer3.31.conv3.weight_bias FLOAT[2048] 202ede3bf5ba
visual.layer3.32.conv1.weight FLOAT[512,2048,1,1] 32ae8f0030b2
visual.layer3.32.conv1.weight_bias FLOAT[512] 58bbaaa321a4
visual.layer3.32.conv2.weight FLOAT[512,512,3,3] 5a511a5c16e5
visual.layer3.32.conv2.weight_bias FLOAT[512] f6c721296a0e
visual.layer3.32.conv3.weight FLOAT[2048,512,1,1] 2d28ba7c8aed
visual.layer3.32.conv3.weight_bias FLOAT[2048] 4dff9ce75f73
visual.layer3.33.conv1.weight FLOAT[512,2048,1,1] 3030ad037576
visual.layer3.33.conv1.weight_bias FLOAT[512] bb77671b5ed1
visual.layer3.33.conv2.weight FLOAT[512,512,3,3] 9c83a47873fa
visual.layer3.33.conv2.weight_bias FLOAT[512] fe7298eb8e38
visual.layer3.33.conv3.weight FLOAT[2048,512,1,1] 3a7ce3f6cba2
visual.layer3.33.conv3.weight_bias FLOAT[2048] b9f274013827
visual.layer3.34.conv1.weight FLOAT[512,2048,1,1] f8a130c2ca0f
visual.layer3.34.conv1.weight_bias FLOAT[512] 05f1dad6cc59
visual.layer3.34.conv2.weight FLOAT[512,512,3,3] 5febb4b342b2
visual.layer3.34.conv2.weight_bias FLOAT[512] 1ce2baf7ce12
visual.layer3.34.conv3.weight FLOAT[2048,512,1,1] fb58ab295e1b
visual.layer3.34.conv3.weight_bias FLOAT[2048] aed078a55b44
visual.layer3.35.conv1.weight FLOAT[512,2048,1,1] 2d9d0e0b2975
visual.layer3.35.conv1.weight_bias FLOAT[512] efbd2e069706
visual.layer3.35.conv2.weight FLOAT[512,512,3,3] 1037a980d174
visual.layer3.35.conv2.weight_bias FLOAT[512] 0c06f696ad39
visual.layer3.35.conv3.weight FLOAT[2048,512,1,1] 11d4c1ffc0a3
visual.layer3.35.conv3.weight_bias FLOAT[2048] ff808bf38ce7
visual.layer3.4.conv1.weight FLOAT[512,2048,1,1] 6b7d3cae6202
visual.layer3.4.conv1.weight_bias FLOAT[512] 7131b5370230
visual.layer3.4.conv2.weight FLOAT[512,512,3,3] 45f669603c5f
visual.layer3.4.conv2.weight_bias FLOAT[512] b5179c860150
visual.layer3.4.conv3.weight FLOAT[2048,512,1,1] 685ffb513bb9
visual.layer3.4.conv3.weight_bias FLOAT[2048] 060256ddf9f2
visual.layer3.5.conv1.weight FLOAT[512,2048,1,1] 372a2d91e8f4
visual.layer3.5.conv1.weight_bias FLOAT[512] 7878f6ab112f
visual.layer3.5.conv2.weight FLOAT[512,512,3,3] cf151f7f7ec4
visual.layer3.5.conv2.weight_bias FLOAT[512] c9f67f812ec1
visual.layer3.5.conv3.weight FLOAT[2048,512,1,1] 056d92550320
visual.layer3.5.conv3.weight_bias FLOAT[2048] da491cb684b2
visual.layer3.6.conv1.weight FLOAT[512,2048,1,1] 3c7a87676844
visual.layer3.6.conv1.weight_bias FLOAT[512] 79f0c164981e
visual.layer3.6.conv2.weight FLOAT[512,512,3,3] fbe95e51a226
visual.layer3.6.conv2.weight_bias FLOAT[512] 31f246cffaa2
visual.layer3.6.conv3.weight FLOAT[2048,512,1,1] 43f3621c5697
visual.layer3.6.conv3.weight_bias FLOAT[2048] 633d75bc552e
visual.layer3.7.conv1.weight FLOAT[512,2048,1,1] 1feac3d10f45
visual.layer3.7.conv1.weight_bias FLOAT[512] 8a4d8edf8f83
visual.layer3.7.conv2.weight FLOAT[512,512,3,3] f99854884824
visual.layer3.7.conv2.weight_bias FLOAT[512] 424a5fea0139
visual.layer3.7.conv3.weight FLOAT[2048,512,1,1] 15a07b8e84de
visual.layer3.7.conv3.weight_bias FLOAT[2048] 6f7a9f502ef9
visual.layer3.8.conv1.weight FLOAT[512,2048,1,1] 9f0c1dac6924
visual.layer3.8.conv1.weight_bias FLOAT[512] 9df2e78273ac
visual.layer3.8.conv2.weight FLOAT[512,512,3,3] 9670f85ede83
visual.layer3.8.conv2.weight_bias FLOAT[512] a17f9c063720
visual.layer3.8.conv3.weight FLOAT[2048,512,1,1] d7586f865198
visual.layer3.8.conv3.weight_bias FLOAT[2048] 6223cd427733
visual.layer3.9.conv1.weight FLOAT[512,2048,1,1] 09cdfcec57b7
visual.layer3.9.conv1.weight_bias FLOAT[512] 1fe3ae64aa3f
visual.layer3.9.conv2.weight FLOAT[512,512,3,3] f2ca43108f30
visual.layer3.9.conv2.weight_bias FLOAT[512] 8773855d2ef4
visual.layer3.9.conv3.weight FLOAT[2048,512,1,1] fb986b8c2ade
visual.layer3.9.conv3.weight_bias FLOAT[2048] 39e44825ec27
visual.layer4.0.conv1.weight FLOAT[1024,2048,1,1] 6d8c5d17d1e1
visual.layer4.0.conv1.weight_bias FLOAT[1024] fc02834e5e95
visual.layer4.0.conv2.weight FLOAT[1024,1024,3,3] cc7ebd31f7b7
visual.layer4.0.conv2.weight_bias FLOAT[1024] 503a5f32e7d0
visual.layer4.0.conv3.weight FLOAT[4096,1024,1,1] fa5a19a1c50e
visual.layer4.0.conv3.weight_bias FLOAT[4096] d7d50bd61f27
visual.layer4.0.downsample.0.weight FLOAT[4096,2048,1,1] 91776c08c91c
visual.layer4.0.downsample.0.weight_bias FLOAT[4096] 7fb659b9bc52
visual.layer4.1.conv1.weight FLOAT[1024,4096,1,1] 3755b6d1b8a2
visual.layer4.1.conv1.weight_bias FLOAT[1024] 1d62627d9727
visual.layer4.1.conv2.weight FLOAT[1024,1024,3,3] 32917bde6312
visual.layer4.1.conv2.weight_bias FLOAT[1024] ac4f5df17bdf
visual.layer4.1.conv3.weight FLOAT[4096,1024,1,1] 1653db0a9e0b
visual.layer4.1.conv3.weight_bias FLOAT[4096] 570ccf7470e3
visual.layer4.2.conv1.weight FLOAT[1024,4096,1,1] faff8a876f6e
visual.layer4.2.conv1.weight_bias FLOAT[1024] 5b7f7c52ba99
visual.layer4.2.conv2.weight FLOAT[1024,1024,3,3] abe25840348f
visual.layer4.2.conv2.weight_bias FLOAT[1024] 440858fad1c2
visual.layer4.2.conv3.weight FLOAT[4096,1024,1,1] 26f9d0fb5e2c
visual.layer4.2.conv3.weight_bias FLOAT[4096] 5807e792c042
visual.layer4.3.conv1.weight FLOAT[1024,4096,1,1] 076d885d4075
visual.layer4.3.conv1.weight_bias FLOAT[1024] 9fb704663292
visual.layer4.3.conv2.weight FLOAT[1024,1024,3,3] b775f158d0a9
visual.layer4.3.conv2.weight_bias FLOAT[1024] d49673b2b449
visual.layer4.3.conv3.weight FLOAT[4096,1024,1,1] 04d74042fcbe
visual.layer4.3.conv3.weight_bias FLOAT[4096] 79ecc53a869a
visual.layer4.4.conv1.weight FLOAT[1024,4096,1,1] b1a4654fa465
visual.layer4.4.conv1.weight_bias FLOAT[1024] 1fdd21109a7a
visual.layer4.4.conv2.weight FLOAT[1024,1024,3,3] 49e6b26616ec
visual.layer4.4.conv2.weight_bias FLOAT[1024] fc945654989b
visual.layer4.4.conv3.weight FLOAT[4096,1024,1,1] 933ae6253134
visual.layer4.4.conv3.weight_bias FLOAT[4096] f210b3a6ca52
visual.layer4.5.conv1.weight FLOAT[1024,4096,1,1] 33bd264341b8
visual.layer4.5.conv1.weight_bias FLOAT[1024] 02a15e2b5c90
visual.layer4.5.conv2.weight FLOAT[1024,1024,3,3] a80e4b7ef69c
visual.layer4.5.conv2.weight_bias FLOAT[1024] f541772d56c5
visual.layer4.5.conv3.weight FLOAT[4096,1024,1,1] 2927b3ba911e
visual.layer4.5.conv3.weight_bias FLOAT[4096] 7a1e92f279ea
visual.layer4.6.conv1.weight FLOAT[1024,4096,1,1] 728ffce7882e
visual.layer4.6.conv1.weight_bias FLOAT[1024] bcfac20625a8
visual.layer4.6.conv2.weight FLOAT[1024,1024,3,3] 6da00c0ac8df
visual.layer4.6.conv2.weight_bias FLOAT[1024] dc3f129f2839
visual.layer4.6.conv3.weight FLOAT[4096,1024,1,1] 3fa93b7781cc
visual.layer4.6.conv3.weight_bias FLOAT[4096] 0a06e5e64f72
visual.layer4.7.conv1.weight FLOAT[1024,4096,1,1] bf27a9fa2d54
visual.layer4.7.conv1.weight_bias FLOAT[1024] a48a1650414c
visual.layer4.7.conv2.weight FLOAT[1024,1024,3,3] 453d315493af
visual.layer4.7.conv2.weight_bias FLOAT[1024] be78be5d217c
visual.layer4.7.conv3.weight FLOAT[4096,1024,1,1] 2ac35fc20dd8
visual.layer4.7.conv3.weight_bias FLOAT[4096] 5e3e0f01c955
visual.layer4.8.conv1.weight FLOAT[1024,4096,1,1] 8de67855e436
visual.layer4.8.conv1.weight_bias FLOAT[1024] 505e79e4b296
visual.layer4.8.conv2.weight FLOAT[1024,1024,3,3] 6921fc595126
visual.layer4.8.conv2.weight_bias FLOAT[1024] c6329683d8f5
visual.layer4.8.conv3.weight FLOAT[4096,1024,1,1] 7c6752d68518
visual.layer4.8.conv3.weight_bias FLOAT[4096] 5aa7a8029105
visual.layer4.9.conv1.weight FLOAT[1024,4096,1,1] 2b56d6c1f22f
visual.layer4.9.conv1.weight_bias FLOAT[1024] b267fc0994ba
visual.layer4.9.conv2.weight FLOAT[1024,1024,3,3] 9b0346d0873a
visual.layer4.9.conv2.weight_bias FLOAT[1024] df869f60fba1
visual.layer4.9.conv3.weight FLOAT[4096,1024,1,1] 3b8419b91ef3
visual.layer4.9.conv3.weight_bias FLOAT[4096] a53731ae3157
