<
   ir_version: 10,
   opset_import: ["" : 23],
   producer_name: "pytorch"
>
main_graph (uint8[batch,288,288,3] image) => (float[batch,640] image_embedding) 
   <
      float[batch,1280,18,18] add_1016
      float[batch,1280,18,18] add_1092
      float[batch,1280,18,18] add_1168
      float[batch,1280,18,18] add_1244
      float[batch,1280,18,18] add_1320
      float[batch,1280,18,18] add_1396
      float[batch,320,72,72] add_140
      float[batch,1280,18,18] add_1472
      float[batch,1280,18,18] add_1548
      float[batch,1280,18,18] add_1624
      float[batch,2560,9,9] add_1720
      float[batch,2560,9,9] add_1796
      float[batch,2560,9,9] add_1872
      float[batch,2560,9,9] add_1948
      float[batch,2560,9,9] add_2024
      float[batch,2560,9,9] add_2100
      float[82,batch,2560] add_2130
      float[batch,320,72,72] add_216
      float[batch,320,72,72] add_292
      float[batch,320,72,72] add_368
      float[batch,640,36,36] add_464
      float[batch,640,36,36] add_540
      float[batch,640,36,36] add_616
      float[batch,640,36,36] add_692
      float[batch,640,36,36] add_768
      float[batch,640,36,36] add_844
      float[batch,1280,18,18] add_940
      float[batch,80,72,72] avg_pool2d
      float[batch,160,36,36] avg_pool2d_2
      float[batch,320,36,36] avg_pool2d_3
      float[batch,320,18,18] avg_pool2d_4
      float[batch,640,18,18] avg_pool2d_5
      float[batch,640,9,9] avg_pool2d_6
      float[batch,1280,9,9] avg_pool2d_7
      float[82,batch,2560] cat
      float[batch,1] clamp_min
      float[batch,40,144,144] getitem
      float[batch,640,36,36] getitem_102
      float[batch,320,36,36] getitem_105
      float[batch,320,36,36] getitem_108
      float[batch,1280,18,18] getitem_111
      float[batch,1280,18,18] getitem_114
      float[batch,320,18,18] getitem_117
      float[batch,80,72,72] getitem_12
      float[batch,320,18,18] getitem_120
      float[batch,1280,18,18] getitem_123
      float[batch,320,18,18] getitem_126
      float[batch,320,18,18] getitem_129
      float[batch,1280,18,18] getitem_132
      float[batch,320,18,18] getitem_135
      float[batch,320,18,18] getitem_138
      float[batch,1280,18,18] getitem_141
      float[batch,320,18,18] getitem_144
      float[batch,320,18,18] getitem_147
      float[batch,320,72,72] getitem_15
      float[batch,1280,18,18] getitem_150
      float[batch,320,18,18] getitem_153
      float[batch,320,18,18] getitem_156
      float[batch,1280,18,18] getitem_159
      float[batch,320,18,18] getitem_162
      float[batch,320,18,18] getitem_165
      float[batch,1280,18,18] getitem_168
      float[batch,320,18,18] getitem_171
      float[batch,320,18,18] getitem_174
      float[batch,1280,18,18] getitem_177
      float[batch,320,72,72] getitem_18
      float[batch,320,18,18] getitem_180
      float[batch,320,18,18] getitem_183
      float[batch,1280,18,18] getitem_186
      float[batch,320,18,18] getitem_189
      float[batch,320,18,18] getitem_192
      float[batch,1280,18,18] getitem_195
      float[batch,640,18,18] getitem_198
      float[batch,640,18,18] getitem_201
      float[batch,2560,9,9] getitem_204
      float[batch,2560,9,9] getitem_207
      float[batch,80,72,72] getitem_21
      float[batch,640,9,9] getitem_210
      float[batch,640,9,9] getitem_213
      float[batch,2560,9,9] getitem_216
      float[batch,640,9,9] getitem_219
      float[batch,640,9,9] getitem_222
      float[batch,2560,9,9] getitem_225
      float[batch,640,9,9] getitem_228
      float[batch,640,9,9] getitem_231
      float[batch,2560,9,9] getitem_234
      float[batch,640,9,9] getitem_237
      float[batch,80,72,72] getitem_24
      float[batch,640,9,9] getitem_240
      float[batch,2560,9,9] getitem_243
      float[batch,640,9,9] getitem_246
      float[batch,640,9,9] getitem_249
      float[batch,2560,9,9] getitem_252
      float[batch,320,72,72] getitem_27
      float[batch,40,144,144] getitem_3
      float[batch,80,72,72] getitem_30
      float[batch,80,72,72] getitem_33
      float[batch,320,72,72] getitem_36
      float[batch,80,72,72] getitem_39
      float[batch,80,72,72] getitem_42
      float[batch,320,72,72] getitem_45
      float[batch,160,72,72] getitem_48
      float[batch,160,72,72] getitem_51
      float[batch,640,36,36] getitem_54
      float[batch,640,36,36] getitem_57
      float[batch,80,144,144] getitem_6
      float[batch,160,36,36] getitem_60
      float[batch,160,36,36] getitem_63
      float[batch,640,36,36] getitem_66
      float[batch,160,36,36] getitem_69
      float[batch,160,36,36] getitem_72
      float[batch,640,36,36] getitem_75
      float[batch,160,36,36] getitem_78
      float[batch,160,36,36] getitem_81
      float[batch,640,36,36] getitem_84
      float[batch,160,36,36] getitem_87
      float[batch,80,72,72] getitem_9
      float[batch,160,36,36] getitem_90
      float[batch,640,36,36] getitem_93
      float[batch,160,36,36] getitem_96
      float[batch,160,36,36] getitem_99
      float[batch,3,288,288] image_chw
      float[batch,288,288,3] image_f32
      float[batch,288,288,3] image_shifted
      float[batch,1] linalg_vector_norm
      float[1,batch,2560] linear
      float[82,batch,2560] linear_1
      float[82,batch,2560] linear_2
      float[batch,640] linear_3
      float[1,batch,2560] mean
      float[1,batch,2560] node_scaled_dot_product_attention_q_row
      float[81,batch,2560] permute_1
      float[1,batch,40,64] permute_2
      float[batch,40,144,144] relu
      float[batch,40,144,144] relu_1
      float[batch,80,72,72] relu_10
      float[batch,320,72,72] relu_11
      float[batch,80,72,72] relu_12
      float[batch,80,72,72] relu_13
      float[batch,320,72,72] relu_14
      float[batch,160,72,72] relu_15
      float[batch,160,72,72] relu_16
      float[batch,640,36,36] relu_17
      float[batch,160,36,36] relu_18
      float[batch,160,36,36] relu_19
      float[batch,80,144,144] relu_2
      float[batch,640,36,36] relu_20
      float[batch,160,36,36] relu_21
      float[batch,160,36,36] relu_22
      float[batch,640,36,36] relu_23
      float[batch,160,36,36] relu_24
      float[batch,160,36,36] relu_25
      float[batch,640,36,36] relu_26
      float[batch,160,36,36] relu_27
      float[batch,160,36,36] relu_28
      float[batch,640,36,36] relu_29
      float[batch,80,72,72] relu_3
      float[batch,160,36,36] relu_30
      float[batch,160,36,36] relu_31
      float[batch,640,36,36] relu_32
      float[batch,320,36,36] relu_33
      float[batch,320,36,36] relu_34
      float[batch,1280,18,18] relu_35
      float[batch,320,18,18] relu_36
      float[batch,320,18,18] relu_37
      float[batch,1280,18,18] relu_38
      float[batch,320,18,18] relu_39
      float[batch,80,72,72] relu_4
      float[batch,320,18,18] relu_40
      float[batch,1280,18,18] relu_41
      float[batch,320,18,18] relu_42
      float[batch,320,18,18] relu_43
      float[batch,1280,18,18] relu_44
      float[batch,320,18,18] relu_45
      float[batch,320,18,18] relu_46
      float[batch,1280,18,18] relu_47
      float[batch,320,18,18] relu_48
      float[batch,320,18,18] relu_49
      float[batch,320,72,72] relu_5
      float[batch,1280,18,18] relu_50
      float[batch,320,18,18] relu_51
      float[batch,320,18,18] relu_52
      float[batch,1280,18,18] relu_53
      float[batch,320,18,18] relu_54
      float[batch,320,18,18] relu_55
      float[batch,1280,18,18] relu_56
      float[batch,320,18,18] relu_57
      float[batch,320,18,18] relu_58
      float[batch,1280,18,18] relu_59
      float[batch,80,72,72] relu_6
      float[batch,320,18,18] relu_60
      float[batch,320,18,18] relu_61
      float[batch,1280,18,18] relu_62
      float[batch,640,18,18] relu_63
      float[batch,640,18,18] relu_64
      float[batch,2560,9,9] relu_65
      float[batch,640,9,9] relu_66
      float[batch,640,9,9] relu_67
      float[batch,2560,9,9] relu_68
      float[batch,640,9,9] relu_69
      float[batch,80,72,72] relu_7
      float[batch,640,9,9] relu_70
      float[batch,2560,9,9] relu_71
      float[batch,640,9,9] relu_72
      float[batch,640,9,9] relu_73
      float[batch,2560,9,9] relu_74
      float[batch,640,9,9] relu_75
      float[batch,640,9,9] relu_76
      float[batch,2560,9,9] relu_77
      float[batch,640,9,9] relu_78
      float[batch,640,9,9] relu_79
      float[batch,320,72,72] relu_8
      float[batch,2560,9,9] relu_80
      float[batch,80,72,72] relu_9
      float[batch,40,1,64] scaled_dot_product_attention
      float[batch,640] select
      float[2560] split_split_0
      float[2560] split_split_1
      float[2560] split_split_2
      float[unk__1,1,64] transpose
      float[unk__1,82,64] transpose_1
      float[unk__1,82,64] transpose_2
      float[82,1,2560] unsqueeze
      float[1,batch,2560] val_7
      float[82,batch,2560] val_8
      float[82,batch,2560] val_9
      float[1,batch,640] view_10
      float[batch,2560,81] view_2
      float[1,unk__1,64] view_3
      float[82,unk__1,64] view_4
      float[82,unk__1,64] view_5
      float[batch,40,1,64] view_6
      float[batch,40,82,64] view_7
      float[batch,40,82,64] view_8
      float[batch,2560] 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.layer1.3.conv1.weight", "visual.layer1.3.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.layer1.3.conv2.weight", "visual.layer1.3.conv2.weight_bias")
   relu_13 = Relu (getitem_42)
   getitem_45 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_13, "visual.layer1.3.conv3.weight", "visual.layer1.3.conv3.weight_bias")
   add_368 = Add (getitem_45, relu_11)
   relu_14 = Relu (add_368)
   getitem_48 = 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.layer2.0.conv1.weight", "visual.layer2.0.conv1.weight_bias")
   relu_15 = Relu (getitem_48)
   getitem_51 = 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.layer2.0.conv2.weight", "visual.layer2.0.conv2.weight_bias")
   relu_16 = Relu (getitem_51)
   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_16)
   getitem_54 = 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_14)
   getitem_57 = 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_464 = Add (getitem_54, getitem_57)
   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.layer2.1.conv1.weight", "visual.layer2.1.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.layer2.1.conv2.weight", "visual.layer2.1.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.layer2.1.conv3.weight", "visual.layer2.1.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.layer2.2.conv1.weight", "visual.layer2.2.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.layer2.2.conv2.weight", "visual.layer2.2.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.layer2.2.conv3.weight", "visual.layer2.2.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.layer2.3.conv1.weight", "visual.layer2.3.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.layer2.3.conv2.weight", "visual.layer2.3.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.layer2.3.conv3.weight", "visual.layer2.3.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.layer2.4.conv1.weight", "visual.layer2.4.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.layer2.4.conv2.weight", "visual.layer2.4.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.layer2.4.conv3.weight", "visual.layer2.4.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.layer2.5.conv1.weight", "visual.layer2.5.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.layer2.5.conv2.weight", "visual.layer2.5.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.layer2.5.conv3.weight", "visual.layer2.5.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.layer3.0.conv1.weight", "visual.layer3.0.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.layer3.0.conv2.weight", "visual.layer3.0.conv2.weight_bias")
   relu_34 = Relu (getitem_108)
   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_34)
   getitem_111 = 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_32)
   getitem_114 = 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_940 = Add (getitem_111, getitem_114)
   relu_35 = Relu (add_940)
   getitem_117 = 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.layer3.1.conv1.weight", "visual.layer3.1.conv1.weight_bias")
   relu_36 = Relu (getitem_117)
   getitem_120 = 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.layer3.1.conv2.weight", "visual.layer3.1.conv2.weight_bias")
   relu_37 = Relu (getitem_120)
   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_37, "visual.layer3.1.conv3.weight", "visual.layer3.1.conv3.weight_bias")
   add_1016 = Add (getitem_123, relu_35)
   relu_38 = Relu (add_1016)
   getitem_126 = 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.layer3.2.conv1.weight", "visual.layer3.2.conv1.weight_bias")
   relu_39 = Relu (getitem_126)
   getitem_129 = 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.layer3.2.conv2.weight", "visual.layer3.2.conv2.weight_bias")
   relu_40 = Relu (getitem_129)
   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_40, "visual.layer3.2.conv3.weight", "visual.layer3.2.conv3.weight_bias")
   add_1092 = Add (getitem_132, relu_38)
   relu_41 = Relu (add_1092)
   getitem_135 = 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.layer3.3.conv1.weight", "visual.layer3.3.conv1.weight_bias")
   relu_42 = Relu (getitem_135)
   getitem_138 = 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.layer3.3.conv2.weight", "visual.layer3.3.conv2.weight_bias")
   relu_43 = Relu (getitem_138)
   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_43, "visual.layer3.3.conv3.weight", "visual.layer3.3.conv3.weight_bias")
   add_1168 = Add (getitem_141, relu_41)
   relu_44 = Relu (add_1168)
   getitem_144 = 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.layer3.4.conv1.weight", "visual.layer3.4.conv1.weight_bias")
   relu_45 = Relu (getitem_144)
   getitem_147 = 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.layer3.4.conv2.weight", "visual.layer3.4.conv2.weight_bias")
   relu_46 = Relu (getitem_147)
   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_46, "visual.layer3.4.conv3.weight", "visual.layer3.4.conv3.weight_bias")
   add_1244 = Add (getitem_150, relu_44)
   relu_47 = Relu (add_1244)
   getitem_153 = 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.layer3.5.conv1.weight", "visual.layer3.5.conv1.weight_bias")
   relu_48 = Relu (getitem_153)
   getitem_156 = 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.layer3.5.conv2.weight", "visual.layer3.5.conv2.weight_bias")
   relu_49 = Relu (getitem_156)
   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_49, "visual.layer3.5.conv3.weight", "visual.layer3.5.conv3.weight_bias")
   add_1320 = Add (getitem_159, relu_47)
   relu_50 = Relu (add_1320)
   getitem_162 = 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.layer3.6.conv1.weight", "visual.layer3.6.conv1.weight_bias")
   relu_51 = Relu (getitem_162)
   getitem_165 = 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.layer3.6.conv2.weight", "visual.layer3.6.conv2.weight_bias")
   relu_52 = Relu (getitem_165)
   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_52, "visual.layer3.6.conv3.weight", "visual.layer3.6.conv3.weight_bias")
   add_1396 = Add (getitem_168, relu_50)
   relu_53 = Relu (add_1396)
   getitem_171 = 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.layer3.7.conv1.weight", "visual.layer3.7.conv1.weight_bias")
   relu_54 = Relu (getitem_171)
   getitem_174 = 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.layer3.7.conv2.weight", "visual.layer3.7.conv2.weight_bias")
   relu_55 = Relu (getitem_174)
   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_55, "visual.layer3.7.conv3.weight", "visual.layer3.7.conv3.weight_bias")
   add_1472 = Add (getitem_177, relu_53)
   relu_56 = Relu (add_1472)
   getitem_180 = 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.layer3.8.conv1.weight", "visual.layer3.8.conv1.weight_bias")
   relu_57 = Relu (getitem_180)
   getitem_183 = 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.layer3.8.conv2.weight", "visual.layer3.8.conv2.weight_bias")
   relu_58 = Relu (getitem_183)
   getitem_186 = Conv <auto_pad: string = "NOTSET", dilations: ints = [1, 1], group: int = 1, pads: ints = [0, 0, 0, 0], strides: ints = [1, 1]> (relu_58, "visual.layer3.8.conv3.weight", "visual.layer3.8.conv3.weight_bias")
   add_1548 = Add (getitem_186, relu_56)
   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.layer3.9.conv1.weight", "visual.layer3.9.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.layer3.9.conv2.weight", "visual.layer3.9.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.layer3.9.conv3.weight", "visual.layer3.9.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.layer4.0.conv1.weight", "visual.layer4.0.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.layer4.0.conv2.weight", "visual.layer4.0.conv2.weight_bias")
   relu_64 = Relu (getitem_201)
   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_64)
   getitem_204 = 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_62)
   getitem_207 = 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_1720 = Add (getitem_204, getitem_207)
   relu_65 = Relu (add_1720)
   getitem_210 = 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.layer4.1.conv1.weight", "visual.layer4.1.conv1.weight_bias")
   relu_66 = Relu (getitem_210)
   getitem_213 = 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.layer4.1.conv2.weight", "visual.layer4.1.conv2.weight_bias")
   relu_67 = Relu (getitem_213)
   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_67, "visual.layer4.1.conv3.weight", "visual.layer4.1.conv3.weight_bias")
   add_1796 = Add (getitem_216, relu_65)
   relu_68 = Relu (add_1796)
   getitem_219 = 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.layer4.2.conv1.weight", "visual.layer4.2.conv1.weight_bias")
   relu_69 = Relu (getitem_219)
   getitem_222 = 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.layer4.2.conv2.weight", "visual.layer4.2.conv2.weight_bias")
   relu_70 = Relu (getitem_222)
   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_70, "visual.layer4.2.conv3.weight", "visual.layer4.2.conv3.weight_bias")
   add_1872 = Add (getitem_225, relu_68)
   relu_71 = Relu (add_1872)
   getitem_228 = 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.layer4.3.conv1.weight", "visual.layer4.3.conv1.weight_bias")
   relu_72 = Relu (getitem_228)
   getitem_231 = 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.layer4.3.conv2.weight", "visual.layer4.3.conv2.weight_bias")
   relu_73 = Relu (getitem_231)
   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_73, "visual.layer4.3.conv3.weight", "visual.layer4.3.conv3.weight_bias")
   add_1948 = Add (getitem_234, relu_71)
   relu_74 = Relu (add_1948)
   getitem_237 = 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.layer4.4.conv1.weight", "visual.layer4.4.conv1.weight_bias")
   relu_75 = Relu (getitem_237)
   getitem_240 = 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.layer4.4.conv2.weight", "visual.layer4.4.conv2.weight_bias")
   relu_76 = Relu (getitem_240)
   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_76, "visual.layer4.4.conv3.weight", "visual.layer4.4.conv3.weight_bias")
   add_2024 = Add (getitem_243, relu_74)
   relu_77 = Relu (add_2024)
   getitem_246 = 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.layer4.5.conv1.weight", "visual.layer4.5.conv1.weight_bias")
   relu_78 = Relu (getitem_246)
   getitem_249 = 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.layer4.5.conv2.weight", "visual.layer4.5.conv2.weight_bias")
   relu_79 = Relu (getitem_249)
   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_79, "visual.layer4.5.conv3.weight", "visual.layer4.5.conv3.weight_bias")
   add_2100 = Add (getitem_252, relu_77)
   relu_80 = Relu (add_2100)
   view_2 = Reshape <allowzero: int = 1> (relu_80, 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_2130 = 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_2130, 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_2130, val_5)
   linear_1 = Add (val_8, split_split_1)
   val_9 = MatMul (add_2130, 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[7680] 02169d887c20
image_shift FLOAT[3] 2f7a50e604ad
node_scaled_dot_product_attention_out_1 INT64[3] b21750a40eac
node_scaled_dot_product_attention_q_pack_1 INT64[4] 41b931e3a504
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[2560,2560] 5736c74beb1f
val_5 FLOAT[2560,2560] 807f60e16206
val_6 FLOAT[2560,2560] 333bff0cb9dd
view_2_target INT64[3] e683a94499f2
view_4_target INT64[3] f5e12c71e402
view_7_target INT64[4] 86d1b5aa033a
view_9_target INT64[2] 3025738d86fc
visual.attnpool.c_proj.bias FLOAT[640] 4d3b7acccb96
visual.attnpool.c_proj.weight FLOAT[640,2560] 6d7b8c46e6dc
visual.attnpool.positional_embedding FLOAT[82,2560] 4e5b4b84ec32
visual.conv1.weight FLOAT[40,3,3,3] 4a386e67fdd2
visual.conv1.weight_bias FLOAT[40] 6ad8e26f520b
visual.conv2.weight FLOAT[40,40,3,3] ac78538072a8
visual.conv2.weight_bias FLOAT[40] 377f0867cf6b
visual.conv3.weight FLOAT[80,40,3,3] 95a868052faa
visual.conv3.weight_bias FLOAT[80] af4be2ff044c
visual.layer1.0.conv1.weight FLOAT[80,80,1,1] f7afa8453060
visual.layer1.0.conv1.weight_bias FLOAT[80] 866d62b981bf
visual.layer1.0.conv2.weight FLOAT[80,80,3,3] 1310e8499d9d
visual.layer1.0.conv2.weight_bias FLOAT[80] 91860eb8af5e
visual.layer1.0.conv3.weight FLOAT[320,80,1,1] 9d8938f44c04
visual.layer1.0.conv3.weight_bias FLOAT[320] 589af9f55891
visual.layer1.0.downsample.0.weight FLOAT[320,80,1,1] 4db2473f8977
visual.layer1.0.downsample.0.weight_bias FLOAT[320] 7f8e2faad82a
visual.layer1.1.conv1.weight FLOAT[80,320,1,1] 0ccef24c1d1d
visual.layer1.1.conv1.weight_bias FLOAT[80] 4d9797922307
visual.layer1.1.conv2.weight FLOAT[80,80,3,3] 38cbb92174e8
visual.layer1.1.conv2.weight_bias FLOAT[80] 68e905e8d6c4
visual.layer1.1.conv3.weight FLOAT[320,80,1,1] c13440312ce1
visual.layer1.1.conv3.weight_bias FLOAT[320] fe14ec8ce307
visual.layer1.2.conv1.weight FLOAT[80,320,1,1] be2bd4d6a1ec
visual.layer1.2.conv1.weight_bias FLOAT[80] 522a0b846fa3
visual.layer1.2.conv2.weight FLOAT[80,80,3,3] cd372c100e00
visual.layer1.2.conv2.weight_bias FLOAT[80] 538bb51da6f1
visual.layer1.2.conv3.weight FLOAT[320,80,1,1] 1d3796a33fa0
visual.layer1.2.conv3.weight_bias FLOAT[320] efe66252832c
visual.layer1.3.conv1.weight FLOAT[80,320,1,1] 3aa3e6a35926
visual.layer1.3.conv1.weight_bias FLOAT[80] 83677b1c3ca8
visual.layer1.3.conv2.weight FLOAT[80,80,3,3] a6f6176464f2
visual.layer1.3.conv2.weight_bias FLOAT[80] 7f644086a61c
visual.layer1.3.conv3.weight FLOAT[320,80,1,1] aaa8c1b7386e
visual.layer1.3.conv3.weight_bias FLOAT[320] 221986494af2
visual.layer2.0.conv1.weight FLOAT[160,320,1,1] d3f082c1c0a9
visual.layer2.0.conv1.weight_bias FLOAT[160] 353cbbb2b24f
visual.layer2.0.conv2.weight FLOAT[160,160,3,3] 544b55967be2
visual.layer2.0.conv2.weight_bias FLOAT[160] 023b41e7a379
visual.layer2.0.conv3.weight FLOAT[640,160,1,1] 4f2153d63cca
visual.layer2.0.conv3.weight_bias FLOAT[640] 5d50beb52d55
visual.layer2.0.downsample.0.weight FLOAT[640,320,1,1] 91e4799418bb
visual.layer2.0.downsample.0.weight_bias FLOAT[640] dc8d9e6d4c06
visual.layer2.1.conv1.weight FLOAT[160,640,1,1] c151f9e73b96
visual.layer2.1.conv1.weight_bias FLOAT[160] cde450425808
visual.layer2.1.conv2.weight FLOAT[160,160,3,3] 025f68e48711
visual.layer2.1.conv2.weight_bias FLOAT[160] 2ca1b70ffad6
visual.layer2.1.conv3.weight FLOAT[640,160,1,1] 76e42669cdd6
visual.layer2.1.conv3.weight_bias FLOAT[640] 884817f9aede
visual.layer2.2.conv1.weight FLOAT[160,640,1,1] 30ec5fd76c17
visual.layer2.2.conv1.weight_bias FLOAT[160] d0362b26a192
visual.layer2.2.conv2.weight FLOAT[160,160,3,3] 7b688ff53813
visual.layer2.2.conv2.weight_bias FLOAT[160] ca81004391c8
visual.layer2.2.conv3.weight FLOAT[640,160,1,1] 839306ce4136
visual.layer2.2.conv3.weight_bias FLOAT[640] e4d7afce43e6
visual.layer2.3.conv1.weight FLOAT[160,640,1,1] d80e864a0a4c
visual.layer2.3.conv1.weight_bias FLOAT[160] 480235497d03
visual.layer2.3.conv2.weight FLOAT[160,160,3,3] 967c0966ecfd
visual.layer2.3.conv2.weight_bias FLOAT[160] 5fbd467bf6f7
visual.layer2.3.conv3.weight FLOAT[640,160,1,1] 106e262c4cc5
visual.layer2.3.conv3.weight_bias FLOAT[640] 19e0452cda66
visual.layer2.4.conv1.weight FLOAT[160,640,1,1] 55dd4a366f91
visual.layer2.4.conv1.weight_bias FLOAT[160] 5a211fd53822
visual.layer2.4.conv2.weight FLOAT[160,160,3,3] 332eb3a9b445
visual.layer2.4.conv2.weight_bias FLOAT[160] 889013bce33d
visual.layer2.4.conv3.weight FLOAT[640,160,1,1] 11f216f8878c
visual.layer2.4.conv3.weight_bias FLOAT[640] 369e31603c86
visual.layer2.5.conv1.weight FLOAT[160,640,1,1] 0c36b49d617f
visual.layer2.5.conv1.weight_bias FLOAT[160] 1175dc26bf07
visual.layer2.5.conv2.weight FLOAT[160,160,3,3] c2edefe8a076
visual.layer2.5.conv2.weight_bias FLOAT[160] dd5b8845c661
visual.layer2.5.conv3.weight FLOAT[640,160,1,1] 14d8860297e9
visual.layer2.5.conv3.weight_bias FLOAT[640] c446651da8f0
visual.layer3.0.conv1.weight FLOAT[320,640,1,1] f5a990de992d
visual.layer3.0.conv1.weight_bias FLOAT[320] 0235023ba0ad
visual.layer3.0.conv2.weight FLOAT[320,320,3,3] 6e73d63f4b16
visual.layer3.0.conv2.weight_bias FLOAT[320] d7721389a5f0
visual.layer3.0.conv3.weight FLOAT[1280,320,1,1] 746bab2fd6cd
visual.layer3.0.conv3.weight_bias FLOAT[1280] e25ef9157441
visual.layer3.0.downsample.0.weight FLOAT[1280,640,1,1] 86030ba75d04
visual.layer3.0.downsample.0.weight_bias FLOAT[1280] d6a86911f0f6
visual.layer3.1.conv1.weight FLOAT[320,1280,1,1] 2f6f67c747c2
visual.layer3.1.conv1.weight_bias FLOAT[320] 824000bc75ff
visual.layer3.1.conv2.weight FLOAT[320,320,3,3] 9a10c085b1d1
visual.layer3.1.conv2.weight_bias FLOAT[320] 874f690d69d1
visual.layer3.1.conv3.weight FLOAT[1280,320,1,1] 03bdfa3ab5bb
visual.layer3.1.conv3.weight_bias FLOAT[1280] 86311c514a66
visual.layer3.2.conv1.weight FLOAT[320,1280,1,1] 951843e6d211
visual.layer3.2.conv1.weight_bias FLOAT[320] feb7ff50c0b7
visual.layer3.2.conv2.weight FLOAT[320,320,3,3] 0f44f61100a9
visual.layer3.2.conv2.weight_bias FLOAT[320] 9d9e06c43296
visual.layer3.2.conv3.weight FLOAT[1280,320,1,1] 5999d26f258e
visual.layer3.2.conv3.weight_bias FLOAT[1280] ed0aaf7203dd
visual.layer3.3.conv1.weight FLOAT[320,1280,1,1] dafafc657e88
visual.layer3.3.conv1.weight_bias FLOAT[320] 1176fe66caaa
visual.layer3.3.conv2.weight FLOAT[320,320,3,3] db1b02a33a4a
visual.layer3.3.conv2.weight_bias FLOAT[320] 53392a38af4a
visual.layer3.3.conv3.weight FLOAT[1280,320,1,1] e36cbcc31e2c
visual.layer3.3.conv3.weight_bias FLOAT[1280] 1541dfa1e41a
visual.layer3.4.conv1.weight FLOAT[320,1280,1,1] 8f8c81061c7c
visual.layer3.4.conv1.weight_bias FLOAT[320] 3854f1f443cf
visual.layer3.4.conv2.weight FLOAT[320,320,3,3] e4f849d0d567
visual.layer3.4.conv2.weight_bias FLOAT[320] fff84395cc79
visual.layer3.4.conv3.weight FLOAT[1280,320,1,1] 99b81459dd34
visual.layer3.4.conv3.weight_bias FLOAT[1280] 4807455e3030
visual.layer3.5.conv1.weight FLOAT[320,1280,1,1] 2563338ceed2
visual.layer3.5.conv1.weight_bias FLOAT[320] bae4b2eb7da0
visual.layer3.5.conv2.weight FLOAT[320,320,3,3] 99ffb589312a
visual.layer3.5.conv2.weight_bias FLOAT[320] 1555dd4e2d6c
visual.layer3.5.conv3.weight FLOAT[1280,320,1,1] d2ebc477974d
visual.layer3.5.conv3.weight_bias FLOAT[1280] 0368a82b8e54
visual.layer3.6.conv1.weight FLOAT[320,1280,1,1] 9fc75976ab26
visual.layer3.6.conv1.weight_bias FLOAT[320] ceeb7d39dd5e
visual.layer3.6.conv2.weight FLOAT[320,320,3,3] f78f55346989
visual.layer3.6.conv2.weight_bias FLOAT[320] 3e075e4d8b53
visual.layer3.6.conv3.weight FLOAT[1280,320,1,1] 48833c16cffd
visual.layer3.6.conv3.weight_bias FLOAT[1280] 02a6d55f39a8
visual.layer3.7.conv1.weight FLOAT[320,1280,1,1] ef6a96929d87
visual.layer3.7.conv1.weight_bias FLOAT[320] 3495a266ae82
visual.layer3.7.conv2.weight FLOAT[320,320,3,3] 0ac3ec729e91
visual.layer3.7.conv2.weight_bias FLOAT[320] 9cb6228ab5d1
visual.layer3.7.conv3.weight FLOAT[1280,320,1,1] 3d1ce4907c99
visual.layer3.7.conv3.weight_bias FLOAT[1280] a3ae7ebadbc7
visual.layer3.8.conv1.weight FLOAT[320,1280,1,1] f3811c60495e
visual.layer3.8.conv1.weight_bias FLOAT[320] e136d6d7d163
visual.layer3.8.conv2.weight FLOAT[320,320,3,3] ec205c5dc010
visual.layer3.8.conv2.weight_bias FLOAT[320] f4bd4db0da5a
visual.layer3.8.conv3.weight FLOAT[1280,320,1,1] 0593faa437ca
visual.layer3.8.conv3.weight_bias FLOAT[1280] 9c82ea3d7ccb
visual.layer3.9.conv1.weight FLOAT[320,1280,1,1] c977a533043d
visual.layer3.9.conv1.weight_bias FLOAT[320] 9136d344bc61
visual.layer3.9.conv2.weight FLOAT[320,320,3,3] 43ba83570b84
visual.layer3.9.conv2.weight_bias FLOAT[320] 6aa806f51ee3
visual.layer3.9.conv3.weight FLOAT[1280,320,1,1] 58e8f3c33d2c
visual.layer3.9.conv3.weight_bias FLOAT[1280] 6e4e0dd7cdac
visual.layer4.0.conv1.weight FLOAT[640,1280,1,1] 33a2c2acf487
visual.layer4.0.conv1.weight_bias FLOAT[640] 56be1d72f013
visual.layer4.0.conv2.weight FLOAT[640,640,3,3] 313d718da807
visual.layer4.0.conv2.weight_bias FLOAT[640] 95c74759dfa0
visual.layer4.0.conv3.weight FLOAT[2560,640,1,1] 2eddcb356cd8
visual.layer4.0.conv3.weight_bias FLOAT[2560] dfc87fff3ef3
visual.layer4.0.downsample.0.weight FLOAT[2560,1280,1,1] 09a5f300d26c
visual.layer4.0.downsample.0.weight_bias FLOAT[2560] 0e098493ee4e
visual.layer4.1.conv1.weight FLOAT[640,2560,1,1] 94c1aeafc762
visual.layer4.1.conv1.weight_bias FLOAT[640] 08094d255cce
visual.layer4.1.conv2.weight FLOAT[640,640,3,3] 3f72aceb695f
visual.layer4.1.conv2.weight_bias FLOAT[640] 6628fd335b62
visual.layer4.1.conv3.weight FLOAT[2560,640,1,1] 8ee2318d0de5
visual.layer4.1.conv3.weight_bias FLOAT[2560] 8b2f3c1e4c92
visual.layer4.2.conv1.weight FLOAT[640,2560,1,1] 526c6edce078
visual.layer4.2.conv1.weight_bias FLOAT[640] 0ee11ab4bea3
visual.layer4.2.conv2.weight FLOAT[640,640,3,3] 3494c6a22e6e
visual.layer4.2.conv2.weight_bias FLOAT[640] a84b968e63e7
visual.layer4.2.conv3.weight FLOAT[2560,640,1,1] efaabe1c4a46
visual.layer4.2.conv3.weight_bias FLOAT[2560] 81e2d50ebf23
visual.layer4.3.conv1.weight FLOAT[640,2560,1,1] 3daeaf9b14fb
visual.layer4.3.conv1.weight_bias FLOAT[640] de2343bfbf05
visual.layer4.3.conv2.weight FLOAT[640,640,3,3] 3071670ac1a7
visual.layer4.3.conv2.weight_bias FLOAT[640] e59dfae7d259
visual.layer4.3.conv3.weight FLOAT[2560,640,1,1] a4832de03edd
visual.layer4.3.conv3.weight_bias FLOAT[2560] c7d4f70002be
visual.layer4.4.conv1.weight FLOAT[640,2560,1,1] 1ad51144187d
visual.layer4.4.conv1.weight_bias FLOAT[640] 2198dc6ac1c9
visual.layer4.4.conv2.weight FLOAT[640,640,3,3] 8355b0ebc706
visual.layer4.4.conv2.weight_bias FLOAT[640] b61e3126993f
visual.layer4.4.conv3.weight FLOAT[2560,640,1,1] 90cea2f1097c
visual.layer4.4.conv3.weight_bias FLOAT[2560] 60ade73a4090
visual.layer4.5.conv1.weight FLOAT[640,2560,1,1] 858b46edcf37
visual.layer4.5.conv1.weight_bias FLOAT[640] 75532c664d13
visual.layer4.5.conv2.weight FLOAT[640,640,3,3] 6e6673b1b765
visual.layer4.5.conv2.weight_bias FLOAT[640] 89df48a1dfc3
visual.layer4.5.conv3.weight FLOAT[2560,640,1,1] 63058554a723
visual.layer4.5.conv3.weight_bias FLOAT[2560] 44e46e36b5ac
