program(1.0)
[buildInfo = dict<tensor<string, []>, tensor<string, []>>({{"coremlc-component-MIL", "3520.4.1"}, {"coremlc-version", "3520.5.1"}, {"coremltools-component-torch", "2.5.0"}, {"coremltools-source-dialect", "TorchScript"}, {"coremltools-version", "8.3.0"}})]
{
    func main<ios17>(tensor<fp32, [1, 1, 160000]> waveform) {
            tensor<fp32, [1]> wav_norm_bias = const()[name = tensor<string, []>("wav_norm_bias"), val = tensor<fp32, [1]>([0x1.73505ep-5])];
            tensor<fp32, [1]> wav_norm_weight = const()[name = tensor<string, []>("wav_norm_weight"), val = tensor<fp32, [1]>([0x1.43f862p-7])];
            tensor<fp32, [80, 1, 251]> conv0_weight = const()[name = tensor<string, []>("conv0_weight"), val = tensor<fp32, [80, 1, 251]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(64)))];
            tensor<fp32, [80]> norm0_bias = const()[name = tensor<string, []>("norm0_bias"), val = tensor<fp32, [80]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(80448)))];
            tensor<fp32, [80]> norm0_weight = const()[name = tensor<string, []>("norm0_weight"), val = tensor<fp32, [80]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(80832)))];
            tensor<fp32, [60]> conv1_bias = const()[name = tensor<string, []>("conv1_bias"), val = tensor<fp32, [60]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(81216)))];
            tensor<fp32, [60, 80, 5]> conv1_weight = const()[name = tensor<string, []>("conv1_weight"), val = tensor<fp32, [60, 80, 5]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(81536)))];
            tensor<fp32, [60]> norm1_bias = const()[name = tensor<string, []>("norm1_bias"), val = tensor<fp32, [60]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(177600)))];
            tensor<fp32, [60]> norm1_weight = const()[name = tensor<string, []>("norm1_weight"), val = tensor<fp32, [60]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(177920)))];
            tensor<fp32, [60]> conv2_bias = const()[name = tensor<string, []>("conv2_bias"), val = tensor<fp32, [60]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(178240)))];
            tensor<fp32, [60, 60, 5]> conv2_weight = const()[name = tensor<string, []>("conv2_weight"), val = tensor<fp32, [60, 60, 5]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(178560)))];
            tensor<fp32, [60]> norm2_bias = const()[name = tensor<string, []>("norm2_bias"), val = tensor<fp32, [60]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(250624)))];
            tensor<fp32, [60]> norm2_weight = const()[name = tensor<string, []>("norm2_weight"), val = tensor<fp32, [60]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(250944)))];
            tensor<fp32, [128]> linear0_bias = const()[name = tensor<string, []>("linear0_bias"), val = tensor<fp32, [128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(251264)))];
            tensor<fp32, [128, 256]> linear0_weight = const()[name = tensor<string, []>("linear0_weight"), val = tensor<fp32, [128, 256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(251840)))];
            tensor<fp32, [128]> linear1_bias = const()[name = tensor<string, []>("linear1_bias"), val = tensor<fp32, [128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(382976)))];
            tensor<fp32, [128, 128]> linear1_weight = const()[name = tensor<string, []>("linear1_weight"), val = tensor<fp32, [128, 128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(383552)))];
            tensor<fp32, [7]> classifier_bias = const()[name = tensor<string, []>("classifier_bias"), val = tensor<fp32, [7]>([-0x1.00e888p+0, 0x1.67cb52p-2, 0x1.3d87fp-1, 0x1.c8aa8p-2, -0x1.445f5ep-2, -0x1.591274p-1, -0x1.8fb70ep-2])];
            tensor<fp32, [7, 128]> classifier_weight = const()[name = tensor<string, []>("classifier_weight"), val = tensor<fp32, [7, 128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(449152)))];
            tensor<fp32, []> var_13 = const()[name = tensor<string, []>("op_13"), val = tensor<fp32, []>(0x1.4f8b58p-17)];
            tensor<fp32, [1, 1, 160000]> input_1 = instance_norm(beta = wav_norm_bias, epsilon = var_13, gamma = wav_norm_weight, x = waveform)[name = tensor<string, []>("input_1")];
            tensor<string, []> var_31_pad_type_0 = const()[name = tensor<string, []>("op_31_pad_type_0"), val = tensor<string, []>("valid")];
            tensor<int32, [1]> var_31_strides_0 = const()[name = tensor<string, []>("op_31_strides_0"), val = tensor<int32, [1]>([10])];
            tensor<int32, [2]> var_31_pad_0 = const()[name = tensor<string, []>("op_31_pad_0"), val = tensor<int32, [2]>([0, 0])];
            tensor<int32, [1]> var_31_dilations_0 = const()[name = tensor<string, []>("op_31_dilations_0"), val = tensor<int32, [1]>([1])];
            tensor<int32, []> var_31_groups_0 = const()[name = tensor<string, []>("op_31_groups_0"), val = tensor<int32, []>(1)];
            tensor<fp32, [1, 80, 15975]> var_31 = conv(dilations = var_31_dilations_0, groups = var_31_groups_0, pad = var_31_pad_0, pad_type = var_31_pad_type_0, strides = var_31_strides_0, weight = conv0_weight, x = input_1)[name = tensor<string, []>("op_31")];
            tensor<fp32, [1, 80, 15975]> input_3 = abs(x = var_31)[name = tensor<string, []>("input_3")];
            tensor<int32, [1]> var_34 = const()[name = tensor<string, []>("op_34"), val = tensor<int32, [1]>([3])];
            tensor<int32, [1]> var_36 = const()[name = tensor<string, []>("op_36"), val = tensor<int32, [1]>([3])];
            tensor<string, []> input_5_pad_type_0 = const()[name = tensor<string, []>("input_5_pad_type_0"), val = tensor<string, []>("custom")];
            tensor<int32, [2]> input_5_pad_0 = const()[name = tensor<string, []>("input_5_pad_0"), val = tensor<int32, [2]>([0, 0])];
            tensor<bool, []> input_5_ceil_mode_0 = const()[name = tensor<string, []>("input_5_ceil_mode_0"), val = tensor<bool, []>(false)];
            tensor<fp32, [1, 80, 5325]> input_5 = max_pool(ceil_mode = input_5_ceil_mode_0, kernel_sizes = var_34, pad = input_5_pad_0, pad_type = input_5_pad_type_0, strides = var_36, x = input_3)[name = tensor<string, []>("input_5")];
            tensor<fp32, []> var_43 = const()[name = tensor<string, []>("op_43"), val = tensor<fp32, []>(0x1.4f8b58p-17)];
            tensor<fp32, [1, 80, 5325]> input_7 = instance_norm(beta = norm0_bias, epsilon = var_43, gamma = norm0_weight, x = input_5)[name = tensor<string, []>("input_7")];
            tensor<fp32, []> var_50 = const()[name = tensor<string, []>("op_50"), val = tensor<fp32, []>(0x1.47ae14p-7)];
            tensor<fp32, [1, 80, 5325]> input_9 = leaky_relu(alpha = var_50, x = input_7)[name = tensor<string, []>("input_9")];
            tensor<string, []> input_11_pad_type_0 = const()[name = tensor<string, []>("input_11_pad_type_0"), val = tensor<string, []>("valid")];
            tensor<int32, [1]> input_11_strides_0 = const()[name = tensor<string, []>("input_11_strides_0"), val = tensor<int32, [1]>([1])];
            tensor<int32, [2]> input_11_pad_0 = const()[name = tensor<string, []>("input_11_pad_0"), val = tensor<int32, [2]>([0, 0])];
            tensor<int32, [1]> input_11_dilations_0 = const()[name = tensor<string, []>("input_11_dilations_0"), val = tensor<int32, [1]>([1])];
            tensor<int32, []> input_11_groups_0 = const()[name = tensor<string, []>("input_11_groups_0"), val = tensor<int32, []>(1)];
            tensor<fp32, [1, 60, 5321]> input_11 = conv(bias = conv1_bias, dilations = input_11_dilations_0, groups = input_11_groups_0, pad = input_11_pad_0, pad_type = input_11_pad_type_0, strides = input_11_strides_0, weight = conv1_weight, x = input_9)[name = tensor<string, []>("input_11")];
            tensor<int32, [1]> var_64 = const()[name = tensor<string, []>("op_64"), val = tensor<int32, [1]>([3])];
            tensor<int32, [1]> var_66 = const()[name = tensor<string, []>("op_66"), val = tensor<int32, [1]>([3])];
            tensor<string, []> input_13_pad_type_0 = const()[name = tensor<string, []>("input_13_pad_type_0"), val = tensor<string, []>("custom")];
            tensor<int32, [2]> input_13_pad_0 = const()[name = tensor<string, []>("input_13_pad_0"), val = tensor<int32, [2]>([0, 0])];
            tensor<bool, []> input_13_ceil_mode_0 = const()[name = tensor<string, []>("input_13_ceil_mode_0"), val = tensor<bool, []>(false)];
            tensor<fp32, [1, 60, 1773]> input_13 = max_pool(ceil_mode = input_13_ceil_mode_0, kernel_sizes = var_64, pad = input_13_pad_0, pad_type = input_13_pad_type_0, strides = var_66, x = input_11)[name = tensor<string, []>("input_13")];
            tensor<fp32, []> var_73 = const()[name = tensor<string, []>("op_73"), val = tensor<fp32, []>(0x1.4f8b58p-17)];
            tensor<fp32, [1, 60, 1773]> input_15 = instance_norm(beta = norm1_bias, epsilon = var_73, gamma = norm1_weight, x = input_13)[name = tensor<string, []>("input_15")];
            tensor<fp32, []> var_80 = const()[name = tensor<string, []>("op_80"), val = tensor<fp32, []>(0x1.47ae14p-7)];
            tensor<fp32, [1, 60, 1773]> input_17 = leaky_relu(alpha = var_80, x = input_15)[name = tensor<string, []>("input_17")];
            tensor<string, []> input_19_pad_type_0 = const()[name = tensor<string, []>("input_19_pad_type_0"), val = tensor<string, []>("valid")];
            tensor<int32, [1]> input_19_strides_0 = const()[name = tensor<string, []>("input_19_strides_0"), val = tensor<int32, [1]>([1])];
            tensor<int32, [2]> input_19_pad_0 = const()[name = tensor<string, []>("input_19_pad_0"), val = tensor<int32, [2]>([0, 0])];
            tensor<int32, [1]> input_19_dilations_0 = const()[name = tensor<string, []>("input_19_dilations_0"), val = tensor<int32, [1]>([1])];
            tensor<int32, []> input_19_groups_0 = const()[name = tensor<string, []>("input_19_groups_0"), val = tensor<int32, []>(1)];
            tensor<fp32, [1, 60, 1769]> input_19 = conv(bias = conv2_bias, dilations = input_19_dilations_0, groups = input_19_groups_0, pad = input_19_pad_0, pad_type = input_19_pad_type_0, strides = input_19_strides_0, weight = conv2_weight, x = input_17)[name = tensor<string, []>("input_19")];
            tensor<int32, [1]> var_94 = const()[name = tensor<string, []>("op_94"), val = tensor<int32, [1]>([3])];
            tensor<int32, [1]> var_96 = const()[name = tensor<string, []>("op_96"), val = tensor<int32, [1]>([3])];
            tensor<string, []> input_21_pad_type_0 = const()[name = tensor<string, []>("input_21_pad_type_0"), val = tensor<string, []>("custom")];
            tensor<int32, [2]> input_21_pad_0 = const()[name = tensor<string, []>("input_21_pad_0"), val = tensor<int32, [2]>([0, 0])];
            tensor<bool, []> input_21_ceil_mode_0 = const()[name = tensor<string, []>("input_21_ceil_mode_0"), val = tensor<bool, []>(false)];
            tensor<fp32, [1, 60, 589]> input_21 = max_pool(ceil_mode = input_21_ceil_mode_0, kernel_sizes = var_94, pad = input_21_pad_0, pad_type = input_21_pad_type_0, strides = var_96, x = input_19)[name = tensor<string, []>("input_21")];
            tensor<fp32, []> var_103 = const()[name = tensor<string, []>("op_103"), val = tensor<fp32, []>(0x1.4f8b58p-17)];
            tensor<fp32, [1, 60, 589]> input_23 = instance_norm(beta = norm2_bias, epsilon = var_103, gamma = norm2_weight, x = input_21)[name = tensor<string, []>("input_23")];
            tensor<fp32, []> var_110 = const()[name = tensor<string, []>("op_110"), val = tensor<fp32, []>(0x1.47ae14p-7)];
            tensor<fp32, [1, 60, 589]> x = leaky_relu(alpha = var_110, x = input_23)[name = tensor<string, []>("x")];
            tensor<int32, [3]> transpose_0_perm_0 = const()[name = tensor<string, []>("transpose_0_perm_0"), val = tensor<int32, [3]>([2, 0, 1])];
            tensor<fp32, [512]> add_0 = const()[name = tensor<string, []>("add_0"), val = tensor<fp32, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(452800)))];
            tensor<fp32, [512]> add_1 = const()[name = tensor<string, []>("add_1"), val = tensor<fp32, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(454912)))];
            tensor<fp32, [512, 60]> concat_4 = const()[name = tensor<string, []>("concat_4"), val = tensor<fp32, [512, 60]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(457024)))];
            tensor<fp32, [512, 128]> concat_5 = const()[name = tensor<string, []>("concat_5"), val = tensor<fp32, [512, 128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(579968)))];
            tensor<fp32, [512, 60]> concat_6 = const()[name = tensor<string, []>("concat_6"), val = tensor<fp32, [512, 60]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(842176)))];
            tensor<fp32, [512, 128]> concat_7 = const()[name = tensor<string, []>("concat_7"), val = tensor<fp32, [512, 128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(965120)))];
            tensor<fp32, [1, 256]> input_27_lstm_layer_0_lstm_h0_reshaped = const()[name = tensor<string, []>("input_27_lstm_layer_0_lstm_h0_reshaped"), val = tensor<fp32, [1, 256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(1227328)))];
            tensor<string, []> input_27_lstm_layer_0_direction_0 = const()[name = tensor<string, []>("input_27_lstm_layer_0_direction_0"), val = tensor<string, []>("bidirectional")];
            tensor<bool, []> input_27_lstm_layer_0_output_sequence_0 = const()[name = tensor<string, []>("input_27_lstm_layer_0_output_sequence_0"), val = tensor<bool, []>(true)];
            tensor<string, []> input_27_lstm_layer_0_recurrent_activation_0 = const()[name = tensor<string, []>("input_27_lstm_layer_0_recurrent_activation_0"), val = tensor<string, []>("sigmoid")];
            tensor<string, []> input_27_lstm_layer_0_cell_activation_0 = const()[name = tensor<string, []>("input_27_lstm_layer_0_cell_activation_0"), val = tensor<string, []>("tanh")];
            tensor<string, []> input_27_lstm_layer_0_activation_0 = const()[name = tensor<string, []>("input_27_lstm_layer_0_activation_0"), val = tensor<string, []>("tanh")];
            tensor<fp32, [589, 1, 60]> transpose_0 = transpose(perm = transpose_0_perm_0, x = x)[name = tensor<string, []>("transpose_2")];
            tensor<fp32, [589, 1, 256]> input_27_lstm_layer_0_0, tensor<fp32, [1, 256]> input_27_lstm_layer_0_1, tensor<fp32, [1, 256]> input_27_lstm_layer_0_2 = lstm(activation = input_27_lstm_layer_0_activation_0, bias = add_0, bias_back = add_1, cell_activation = input_27_lstm_layer_0_cell_activation_0, direction = input_27_lstm_layer_0_direction_0, initial_c = input_27_lstm_layer_0_lstm_h0_reshaped, initial_h = input_27_lstm_layer_0_lstm_h0_reshaped, output_sequence = input_27_lstm_layer_0_output_sequence_0, recurrent_activation = input_27_lstm_layer_0_recurrent_activation_0, weight_hh = concat_5, weight_hh_back = concat_7, weight_ih = concat_4, weight_ih_back = concat_6, x = transpose_0)[name = tensor<string, []>("input_27_lstm_layer_0")];
            tensor<fp32, [512]> add_2 = const()[name = tensor<string, []>("add_2"), val = tensor<fp32, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(1228416)))];
            tensor<fp32, [512]> add_3 = const()[name = tensor<string, []>("add_3"), val = tensor<fp32, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(1230528)))];
            tensor<fp32, [512, 256]> concat_14 = const()[name = tensor<string, []>("concat_14"), val = tensor<fp32, [512, 256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(1232640)))];
            tensor<fp32, [512, 128]> concat_15 = const()[name = tensor<string, []>("concat_15"), val = tensor<fp32, [512, 128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(1756992)))];
            tensor<fp32, [512, 256]> concat_16 = const()[name = tensor<string, []>("concat_16"), val = tensor<fp32, [512, 256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(2019200)))];
            tensor<fp32, [512, 128]> concat_17 = const()[name = tensor<string, []>("concat_17"), val = tensor<fp32, [512, 128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(2543552)))];
            tensor<string, []> input_27_lstm_layer_1_direction_0 = const()[name = tensor<string, []>("input_27_lstm_layer_1_direction_0"), val = tensor<string, []>("bidirectional")];
            tensor<bool, []> input_27_lstm_layer_1_output_sequence_0 = const()[name = tensor<string, []>("input_27_lstm_layer_1_output_sequence_0"), val = tensor<bool, []>(true)];
            tensor<string, []> input_27_lstm_layer_1_recurrent_activation_0 = const()[name = tensor<string, []>("input_27_lstm_layer_1_recurrent_activation_0"), val = tensor<string, []>("sigmoid")];
            tensor<string, []> input_27_lstm_layer_1_cell_activation_0 = const()[name = tensor<string, []>("input_27_lstm_layer_1_cell_activation_0"), val = tensor<string, []>("tanh")];
            tensor<string, []> input_27_lstm_layer_1_activation_0 = const()[name = tensor<string, []>("input_27_lstm_layer_1_activation_0"), val = tensor<string, []>("tanh")];
            tensor<fp32, [589, 1, 256]> input_27_lstm_layer_1_0, tensor<fp32, [1, 256]> input_27_lstm_layer_1_1, tensor<fp32, [1, 256]> input_27_lstm_layer_1_2 = lstm(activation = input_27_lstm_layer_1_activation_0, bias = add_2, bias_back = add_3, cell_activation = input_27_lstm_layer_1_cell_activation_0, direction = input_27_lstm_layer_1_direction_0, initial_c = input_27_lstm_layer_0_lstm_h0_reshaped, initial_h = input_27_lstm_layer_0_lstm_h0_reshaped, output_sequence = input_27_lstm_layer_1_output_sequence_0, recurrent_activation = input_27_lstm_layer_1_recurrent_activation_0, weight_hh = concat_15, weight_hh_back = concat_17, weight_ih = concat_14, weight_ih_back = concat_16, x = input_27_lstm_layer_0_0)[name = tensor<string, []>("input_27_lstm_layer_1")];
            tensor<fp32, [512]> add_4 = const()[name = tensor<string, []>("add_4"), val = tensor<fp32, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(2805760)))];
            tensor<fp32, [512]> add_5 = const()[name = tensor<string, []>("add_5"), val = tensor<fp32, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(2807872)))];
            tensor<fp32, [512, 256]> concat_24 = const()[name = tensor<string, []>("concat_24"), val = tensor<fp32, [512, 256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(2809984)))];
            tensor<fp32, [512, 128]> concat_25 = const()[name = tensor<string, []>("concat_25"), val = tensor<fp32, [512, 128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3334336)))];
            tensor<fp32, [512, 256]> concat_26 = const()[name = tensor<string, []>("concat_26"), val = tensor<fp32, [512, 256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3596544)))];
            tensor<fp32, [512, 128]> concat_27 = const()[name = tensor<string, []>("concat_27"), val = tensor<fp32, [512, 128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4120896)))];
            tensor<string, []> input_27_lstm_layer_2_direction_0 = const()[name = tensor<string, []>("input_27_lstm_layer_2_direction_0"), val = tensor<string, []>("bidirectional")];
            tensor<bool, []> input_27_lstm_layer_2_output_sequence_0 = const()[name = tensor<string, []>("input_27_lstm_layer_2_output_sequence_0"), val = tensor<bool, []>(true)];
            tensor<string, []> input_27_lstm_layer_2_recurrent_activation_0 = const()[name = tensor<string, []>("input_27_lstm_layer_2_recurrent_activation_0"), val = tensor<string, []>("sigmoid")];
            tensor<string, []> input_27_lstm_layer_2_cell_activation_0 = const()[name = tensor<string, []>("input_27_lstm_layer_2_cell_activation_0"), val = tensor<string, []>("tanh")];
            tensor<string, []> input_27_lstm_layer_2_activation_0 = const()[name = tensor<string, []>("input_27_lstm_layer_2_activation_0"), val = tensor<string, []>("tanh")];
            tensor<fp32, [589, 1, 256]> input_27_lstm_layer_2_0, tensor<fp32, [1, 256]> input_27_lstm_layer_2_1, tensor<fp32, [1, 256]> input_27_lstm_layer_2_2 = lstm(activation = input_27_lstm_layer_2_activation_0, bias = add_4, bias_back = add_5, cell_activation = input_27_lstm_layer_2_cell_activation_0, direction = input_27_lstm_layer_2_direction_0, initial_c = input_27_lstm_layer_0_lstm_h0_reshaped, initial_h = input_27_lstm_layer_0_lstm_h0_reshaped, output_sequence = input_27_lstm_layer_2_output_sequence_0, recurrent_activation = input_27_lstm_layer_2_recurrent_activation_0, weight_hh = concat_25, weight_hh_back = concat_27, weight_ih = concat_24, weight_ih_back = concat_26, x = input_27_lstm_layer_1_0)[name = tensor<string, []>("input_27_lstm_layer_2")];
            tensor<fp32, [512]> add_6 = const()[name = tensor<string, []>("add_6"), val = tensor<fp32, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4383104)))];
            tensor<fp32, [512]> add_7 = const()[name = tensor<string, []>("add_7"), val = tensor<fp32, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4385216)))];
            tensor<fp32, [512, 256]> concat_34 = const()[name = tensor<string, []>("concat_34"), val = tensor<fp32, [512, 256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4387328)))];
            tensor<fp32, [512, 128]> concat_35 = const()[name = tensor<string, []>("concat_35"), val = tensor<fp32, [512, 128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4911680)))];
            tensor<fp32, [512, 256]> concat_36 = const()[name = tensor<string, []>("concat_36"), val = tensor<fp32, [512, 256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(5173888)))];
            tensor<fp32, [512, 128]> concat_37 = const()[name = tensor<string, []>("concat_37"), val = tensor<fp32, [512, 128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(5698240)))];
            tensor<string, []> input_27_batch_first_direction_0 = const()[name = tensor<string, []>("input_27_batch_first_direction_0"), val = tensor<string, []>("bidirectional")];
            tensor<bool, []> input_27_batch_first_output_sequence_0 = const()[name = tensor<string, []>("input_27_batch_first_output_sequence_0"), val = tensor<bool, []>(true)];
            tensor<string, []> input_27_batch_first_recurrent_activation_0 = const()[name = tensor<string, []>("input_27_batch_first_recurrent_activation_0"), val = tensor<string, []>("sigmoid")];
            tensor<string, []> input_27_batch_first_cell_activation_0 = const()[name = tensor<string, []>("input_27_batch_first_cell_activation_0"), val = tensor<string, []>("tanh")];
            tensor<string, []> input_27_batch_first_activation_0 = const()[name = tensor<string, []>("input_27_batch_first_activation_0"), val = tensor<string, []>("tanh")];
            tensor<fp32, [589, 1, 256]> input_27_batch_first_0, tensor<fp32, [1, 256]> input_27_batch_first_1, tensor<fp32, [1, 256]> input_27_batch_first_2 = lstm(activation = input_27_batch_first_activation_0, bias = add_6, bias_back = add_7, cell_activation = input_27_batch_first_cell_activation_0, direction = input_27_batch_first_direction_0, initial_c = input_27_lstm_layer_0_lstm_h0_reshaped, initial_h = input_27_lstm_layer_0_lstm_h0_reshaped, output_sequence = input_27_batch_first_output_sequence_0, recurrent_activation = input_27_batch_first_recurrent_activation_0, weight_hh = concat_35, weight_hh_back = concat_37, weight_ih = concat_34, weight_ih_back = concat_36, x = input_27_lstm_layer_2_0)[name = tensor<string, []>("input_27_batch_first")];
            tensor<int32, [3]> input_27_perm_0 = const()[name = tensor<string, []>("input_27_perm_0"), val = tensor<int32, [3]>([1, 0, 2])];
            tensor<fp32, [1, 589, 256]> input_27 = transpose(perm = input_27_perm_0, x = input_27_batch_first_0)[name = tensor<string, []>("transpose_1")];
            tensor<fp32, [1, 589, 128]> input_29 = linear(bias = linear0_bias, weight = linear0_weight, x = input_27)[name = tensor<string, []>("linear_0")];
            tensor<fp32, []> var_170 = const()[name = tensor<string, []>("op_170"), val = tensor<fp32, []>(0x1.47ae14p-7)];
            tensor<fp32, [1, 589, 128]> input_31 = leaky_relu(alpha = var_170, x = input_29)[name = tensor<string, []>("input_31")];
            tensor<fp32, [1, 589, 128]> input_33 = linear(bias = linear1_bias, weight = linear1_weight, x = input_31)[name = tensor<string, []>("linear_1")];
            tensor<fp32, []> var_175 = const()[name = tensor<string, []>("op_175"), val = tensor<fp32, []>(0x1.47ae14p-7)];
            tensor<fp32, [1, 589, 128]> input_35 = leaky_relu(alpha = var_175, x = input_33)[name = tensor<string, []>("input_35")];
            tensor<fp32, [1, 589, 7]> input = linear(bias = classifier_bias, weight = classifier_weight, x = input_35)[name = tensor<string, []>("linear_2")];
            tensor<int32, []> var_180 = const()[name = tensor<string, []>("op_180"), val = tensor<int32, []>(-1)];
            tensor<fp32, [1, 589, 7]> var_182_softmax = softmax(axis = var_180, x = input)[name = tensor<string, []>("op_182_softmax")];
            tensor<fp32, []> var_182_epsilon_0 = const()[name = tensor<string, []>("op_182_epsilon_0"), val = tensor<fp32, []>(0x1p-149)];
            tensor<fp32, [1, 589, 7]> log_probabilities = log(epsilon = var_182_epsilon_0, x = var_182_softmax)[name = tensor<string, []>("op_182")];
        } -> (log_probabilities);
}