swolchok

new code
def forward(self, b_conformer_layers_0_conv_module_sequential_3_num_batches_tracked, b_conformer_layers_1_conv_module_sequential_3_num_batches_tracked, input, lengths):                                                                                 input_1 = input
    aten_arange_start_step = executorch_exir_dialects_edge__ops_aten_arange_start_step(0, 97, dtype = torch.int32, layout = torch.strided, device = device(type='cpu'), pin_memory = False)
    scalar_tensor = torch.ops.aten.scalar_tensor.default(-inf, dtype = torch.float32, layout = torch.strided, device = device(type='cpu'))
    scalar_tensor_1 = torch.ops.aten.scalar_tensor.default(-inf, dtype = torch.float32, layout = torch.strided, device = device(type='cpu'))
    lowered_module_0 = self.lowered_module_0
    lowered_module_1 = self.lowered_module_1
    lowered_module_2 = self.lowered_module_2
    lowered_module_3 = self.lowered_module_3
    aten_max_default = executorch_exir_dialects_edge__ops_aten_max_default(lengths)

swolchok

Nesting level: 1
output: Dict[str, Tuple[str, str]]
<torch.utils.benchmark.utils.common.Measurement object at 0x7f8a3a31ff40>
x.annotation_str
  Median: 20.85 us
  IQR:    0.67 us (20.55 to 21.22)
  95 measurements, 100 runs per measurement, 1 thread
Nesting level: 3
output: Dict[str, Tuple[Dict[str, Tuple[Dict[str, Tuple[str, str]], Dict[str,
<torch.utils.benchmark.utils.common.Measurement object at 0x7f8a3a31fe80>

swolchok

Nesting level: 1
output: Dict[str, Tuple[str, str]]
<torch.utils.benchmark.utils.common.Measurement object at 0x7f6969bb1f40>
x.annotation_str
  Median: 18.42 us
  IQR:    1.04 us (17.86 to 18.90)
  108 measurements, 100 runs per measurement, 1 thread
Nesting level: 3
output: Dict[str, Tuple[Dict[str, Tuple[Dict[str, Tuple[str, str]], Dict[str,
<torch.utils.benchmark.utils.common.Measurement object at 0x7f6969bb1e80>

swolchok

Nesting level: 1
output: Dict[str, Tuple[str, str]]
<torch.utils.benchmark.utils.common.Measurement object at 0x7f7bcb854f40>
x.annotation_str
  Median: 19.34 us
  IQR:    0.81 us (19.02 to 19.83)
  101 measurements, 100 runs per measurement, 1 thread
Nesting level: 3
output: Dict[str, Tuple[Dict[str, Tuple[Dict[str, Tuple[str, str]], Dict[str,
<torch.utils.benchmark.utils.common.Measurement object at 0x7f7bcb854e80>

swolchok

Flat profile:

Each sample counts as 0.01 seconds.
  %   cumulative   self              self     total           
 time   seconds   seconds    calls   s/call   s/call  name    
 72.28     21.82    21.82 47800000     0.00     0.00  camlDefault_Tar__Layout__layoutNodeImpl_1039
 10.67     25.04     3.22 13800000     0.00     0.00  camlDefault_Tar__Layout__layoutNode_1161
  6.64     27.05     2.01 40900001     0.00     0.00  caml_obj_dup
  4.27     28.34     1.29 61600000     0.00     0.00  camlDefault_Tar__Layout__layoutNodeInternal_1038
  2.48     29.09     0.75   100000     0.00     0.00  camlDefault_Tar__LayoutTestFixedEncoding__fun_3905

swolchok

/*
 * robotMaze.js
 *
 * The blue key is inside a labyrinth, and extracting
 * it will not be easy.
 *
 * It's a good thing that you're a AI expert, or
 * we would have to leave empty-handed.
 */