import os
import unittest
from unittest.mock import patch
from msprobe.visualization.builder.msprobe_adapter import (
get_compare_mode,
run_real_data,
get_input_output,
format_node_data,
compare_node,
_format_decimal_string,
_format_data,
MatchedNodeCalculator
)
from msprobe.visualization.utils import GraphConst
from msprobe.visualization.graph.base_node import BaseNode
import torch
from msprobe.core.common.const import Const
from test.msprobe_test.msaccucmp_ut.test_mscmp_advisor import result
npu_data_in = {
"Functional.conv2d.0.forward.input.0": {
"type": "torch.Tensor",
"dtype": "torch.float32",
"shape": "[10, 3, 64, 64]",
"Max": "4.350435",
"Min": "-4.339223",
"Mean": "-0.001572",
"Norm": "350.247772",
"requires_grad": "False",
"data_name": "Functional.conv2d.0.forward.input.0.pt"
},
"Functional.conv2d.0.forward.input.1": {
"type": "torch.Tensor",
"dtype": "torch.float32",
"shape": "[64, 3, 7, 7]",
"Max": "0.099308",
"Min": "-0.108559",
"Mean": "0.000216",
"Norm": "2.48812",
"requires_grad": "True",
"data_name": "Functional.conv2d.0.forward.input.1.pt"
},
"Functional.conv2d.0.forward.input.2": {
"value": "null"
},
"Functional.conv2d.0.forward.input.3.0": {
"type": "int",
"value": "2",
"data_name": "-1",
"requires_grad": "null",
"Max": "2",
"Min": "2",
"Mean": "2",
"Norm": "2",
"dtype": "<class int>",
"shape": "[]"
},
"Functional.conv2d.0.forward.input.3.1": {
"type": "int",
"value": "2",
"data_name": "-1",
"requires_grad": "null",
"Max": "2",
"Min": "2",
"Mean": "2",
"Norm": "2",
"dtype": "<class int>",
"shape": "[]"
},
"Functional.conv2d.0.forward.input.4.0": {
"type": "int",
"value": "3",
"data_name": "-1",
"requires_grad": "null",
"Max": "3",
"Min": "3",
"Mean": "3",
"Norm": "3",
"dtype": "<class int>",
"shape": "[]"
},
"Functional.conv2d.0.forward.input.4.1": {
"type": "int",
"value": "3",
"data_name": "-1",
"requires_grad": "null",
"Max": "3",
"Min": "3",
"Mean": "3",
"Norm": "3",
"dtype": "<class int>",
"shape": "[]"
},
"Functional.conv2d.0.forward.input.5.0": {
"type": "int",
"value": "1",
"data_name": "-1",
"requires_grad": "null",
"Max": "1",
"Min": "1",
"Mean": "1",
"Norm": "1",
"dtype": "<class int>",
"shape": "[]"
},
"Functional.conv2d.0.forward.input.5.1": {
"type": "int",
"value": "1",
"data_name": "-1",
"requires_grad": "null",
"Max": "1",
"Min": "1",
"Mean": "1",
"Norm": "1",
"dtype": "<class int>",
"shape": "[]"
},
"Functional.conv2d.0.forward.input.6": {
"type": "int",
"value": "1",
"data_name": "-1",
"requires_grad": "null",
"Max": "1",
"Min": "1",
"Mean": "1",
"Norm": "1",
"dtype": "<class int>",
"shape": "[]"
}
}
npu_data_out = {"Functional.conv2d.0.forward.output.0": {"type": "torch.Tensor", "dtype": "torch.float32",
"shape": "[10, 64, 32, 32]",
"Max": "inf", "Min": "-1.559762", "Mean": "inf", "Norm": "inf",
"requires_grad": "True",
"data_name": "Functional.conv2d.0.forward.output.0.pt"}}
npu_data_dir = os.path.join(os.path.dirname(os.path.realpath(__file__)))
bench_data_in = {
"Functional.conv2d.0.forward.input.0": {
"type": "torch.Tensor",
"dtype": "torch.float32",
"shape": "[10, 3, 64, 64]",
"Max": "4.350435",
"Min": "-4.339223",
"Mean": "-0.001572",
"Norm": "350.247772",
"requires_grad": "False",
"data_name": "Functional.conv2d.0.forward.input.0.pt"
},
"Functional.conv2d.0.forward.input.1": {
"type": "torch.Tensor",
"dtype": "torch.float32",
"shape": "[64, 3, 7, 7]",
"Max": "0.099308",
"Min": "-0.108559",
"Mean": "0.000216",
"Norm": "2.48812",
"requires_grad": "True",
"data_name": "Functional.conv2d.0.forward.input.1.pt"
},
"Functional.conv2d.0.forward.input.2": {
"value": "null"
},
"Functional.conv2d.0.forward.input.3.0": {
"type": "int",
"value": "2",
"data_name": "-1",
"requires_grad": "null",
"Max": "2",
"Min": "2",
"Mean": "2",
"Norm": "2",
"dtype": "<class int>",
"shape": "[]"
},
"Functional.conv2d.0.forward.input.3.1": {
"type": "int",
"value": "2",
"data_name": "-1",
"requires_grad": "null",
"Max": "2",
"Min": "2",
"Mean": "2",
"Norm": "2",
"dtype": "<class int>",
"shape": "[]"
},
"Functional.conv2d.0.forward.input.4.0": {
"type": "int",
"value": "3",
"data_name": "-1",
"requires_grad": "null",
"Max": "3",
"Min": "3",
"Mean": "3",
"Norm": "3",
"dtype": "<class int>",
"shape": "[]"
},
"Functional.conv2d.0.forward.input.4.1": {
"type": "int",
"value": "3",
"data_name": "-1",
"requires_grad": "null",
"Max": "3",
"Min": "3",
"Mean": "3",
"Norm": "3",
"dtype": "<class int>",
"shape": "[]"
},
"Functional.conv2d.0.forward.input.5.0": {
"type": "int",
"value": "1",
"data_name": "-1",
"requires_grad": "null",
"Max": "1",
"Min": "1",
"Mean": "1",
"Norm": "1",
"dtype": "<class int>",
"shape": "[]"
},
"Functional.conv2d.0.forward.input.5.1": {
"type": "int",
"value": "1",
"data_name": "-1",
"requires_grad": "null",
"Max": "1",
"Min": "1",
"Mean": "1",
"Norm": "1",
"dtype": "<class int>",
"shape": "[]"
},
"Functional.conv2d.0.forward.input.6": {
"type": "int",
"value": "1",
"data_name": "-1",
"requires_grad": "null",
"Max": "1",
"Min": "1",
"Mean": "1",
"Norm": "1",
"dtype": "<class int>",
"shape": "[]"
}
}
bench_data_out = {"Functional.conv2d.0.forward.output.0": {"type": "torch.Tensor", "dtype": "torch.float32",
"shape": "[10, 64, 32, 32]",
"Max": "1.526799", "Min": "-1.559762", "Mean": "-0.000263",
"Norm": "244.860931",
"requires_grad": "True",
"data_name": "Functional.conv2d.0.forward.output.0.pt"}}
bench_data_dir = os.path.join(os.path.dirname(os.path.realpath(__file__)))
npu_data = {
'input_data': npu_data_in,
'output_data': npu_data_out,
'dump_data_dir': npu_data_dir
}
bench_data = {
'input_data': bench_data_in,
'output_data': bench_data_out,
'dump_data_dir': bench_data_dir
}
npu_data1 = {
'input_data': npu_data_in,
'output_data': npu_data_out,
'dump_data_dir': ''
}
bench_data1 = {
'input_data': bench_data_in,
'output_data': bench_data_out,
'dump_data_dir': ''
}
npu_statistics_data = {
'input_data': {"Module.bn1.BatchNorm2d.forward.0.input.0": {"type": "torch.Tensor", "dtype": "torch.float32",
"shape": "[10, 64, 32, 32]", "Max": "1.526799",
"Min": "-1.559762", "Mean": "-0.000263",
"Norm": "244.860931", "requires_grad": "True",
"data_name": "-1"},
"Module.bn1.BatchNorm2d.forward.0.parameters.weight": {"type": "torch.Tensor",
"dtype": "torch.float32", "shape": "[64]",
"Max": "1.0", "Min": "1.0", "Mean": "1.0",
"Norm": "8.0", "requires_grad": "True",
"data_name": "-1"},
"Module.bn1.BatchNorm2d.forward.0.parameters.bias": {"type": "torch.Tensor",
"dtype": "torch.float32", "shape": "[64]",
"Max": "0.0", "Min": "0.0", "Mean": "0.0",
"Norm": "0.0", "requires_grad": "True",
"data_name": "-1"}}, 'output_data': {
"Module.bn1.BatchNorm2d.forward.0.output.0": {"type": "torch.Tensor", "dtype": "torch.float32",
"shape": "[10, 64, 32, 32]", "Max": "4.673826",
"Min": "-5.146786",
"Mean": "4.656613e-11", "Norm": "80.949384",
"requires_grad": "True",
"data_name": "-1"}}}
bench_statistics_data = {
'input_data': {"Module.bn1.BatchNorm2d.forward.0.input.0": {"type": "torch.Tensor", "dtype": "torch.float32",
"shape": "[10, 64, 32, 32]", "Max": "1.526799",
"Min": "-1.559762", "Mean": "-0.000263",
"Norm": "244.860931", "requires_grad": "True",
"data_name": "-1"},
"Module.bn1.BatchNorm2d.forward.0.parameters.weight": {"type": "torch.Tensor",
"dtype": "torch.float32", "shape": "[64]",
"Max": "1.0", "Min": "1.0", "Mean": "1.0",
"Norm": "8.0", "requires_grad": "True",
"data_name": "-1"},
"Module.bn1.BatchNorm2d.forward.0.parameters.bias": {"type": "torch.Tensor",
"dtype": "torch.float32", "shape": "[64]",
"Max": "0.0", "Min": "0.0", "Mean": "0.0",
"Norm": "0.0", "requires_grad": "True",
"data_name": "-1"}},
'output_data': {"Module.bn1.BatchNorm2d.forward.0.output.0": {"type": "torch.Tensor", "dtype": "torch.float32",
"shape": "[10, 64, 32, 32]", "Max": "4.673826",
"Min": "-5.146786", "Mean": "4.656613e-11",
"Norm": "809.493835", "requires_grad": "True",
"data_name": "-1"}}}
npu_md5_data = {
'input_data': {"Module.conv1.Conv2d.forward.0.input.0": {"type": "torch.Tensor", "dtype": "torch.float32",
"shape": "[10, 3, 64, 64]", "Max": "4.350435",
"Min": "-4.339223", "Mean": "-0.001572",
"Norm": "350.247772", "requires_grad": "False",
"md5": "c20541291", "data_name": "-1"},
"Module.conv1.Conv2d.forward.0.parameters.weight": {"type": "torch.Tensor",
"dtype": "torch.float32",
"shape": "[64, 3, 7, 7]", "Max": "0.099308",
"Min": "-0.108559", "Mean": "0.000216",
"Norm": "2.48812", "requires_grad": "True",
"md5": "3321b66e", "data_name": "-1"}},
'output_data': {"Module.conv1.Conv2d.forward.0.output.0": {"type": "torch.Tensor", "dtype": "torch.float32",
"shape": "[10, 64, 32, 32]", "Max": "1.526799",
"Min": "-1.559762", "Mean": "-0.000263",
"Norm": "244.860931", "requires_grad": "True",
"md5": "1155e5a11", "data_name": "-1"}}}
bench_md5_data = {
'input_data': {"Module.conv1.Conv2d.forward.0.input.0": {"type": "torch.Tensor", "dtype": "torch.float32",
"shape": "[10, 3, 64, 64]", "Max": "4.350435",
"Min": "-4.339223", "Mean": "-0.001572",
"Norm": "350.247772", "requires_grad": "False",
"md5": "c2054129", "data_name": "-1"},
"Module.conv1.Conv2d.forward.0.parameters.weight": {"type": "torch.Tensor",
"dtype": "torch.float32",
"shape": "[64, 3, 7, 7]", "Max": "0.099308",
"Min": "-0.108559", "Mean": "0.000216",
"Norm": "2.48812", "requires_grad": "True",
"md5": "3321b66e", "data_name": "-1"}},
'output_data': {"Module.conv1.Conv2d.forward.0.output.0": {"type": "torch.Tensor", "dtype": "torch.float32",
"shape": "[10, 64, 32, 32]", "Max": "1.526799",
"Min": "-1.559762", "Mean": "-0.000263",
"Norm": "244.860931", "requires_grad": "True",
"md5": "1155e5a1", "data_name": "-1"}}}
class TestMsprobeAdapter(unittest.TestCase):
@patch('msprobe.visualization.builder.msprobe_adapter.set_dump_path')
@patch('msprobe.visualization.builder.msprobe_adapter.get_dump_mode', return_value=Const.SUMMARY)
def test_get_compare_mode_summary(self, mock_get_dump_mode, mock_set_dump_path):
mode = get_compare_mode("dummy_param")
self.assertEqual(mode, GraphConst.SUMMARY_COMPARE)
def test_get_input_output(self):
node_data = {
'input_args': [{'type': 'torch.Tensor', 'dtype': 'torch.int64', 'shape': [5],
'Max': 2049.0, 'Min': 0.0, 'Mean': 410.20001220703125, 'Norm': 2049.0009765625,
'requires_grad': False, 'full_op_name': 'Distributed.broadcast.0.forward_input.0'},
{'type': 'int', 'value': 0}],
'input_kwargs': {'group': None},
'output': [{'type': 'torch.Tensor', 'dtype': 'torch.int64', 'shape': [5],
'Max': 2049.0, 'Min': 0.0, 'Mean': 410.20001220703125, 'Norm': 2049.0009765625,
'requires_grad': False, 'full_op_name': 'Distributed.broadcast.0.forward_output.0'},
{'type': 'int', 'value': 0}, None]
}
node_id = "Distributed.broadcast.0.forward"
input_data, output_data = get_input_output(node_data, node_id)
self.assertIn("Distributed.broadcast.0.forward.output.0", output_data)
self.assertIn("Distributed.broadcast.0.forward.input.0", input_data)
def test_format_node_data(self):
data_dict = {'node1': {'data_name': 'data1', 'full_op_name': 'op1'}}
result = format_node_data(data_dict)
self.assertNotIn('requires_grad', result['node1'])
@patch('msprobe.visualization.builder.msprobe_adapter.get_accuracy')
def test_compare_node(self, mock_get_accuracy):
node_n = BaseNode('', 'node1')
node_b = BaseNode('', 'node2')
result = compare_node(node_n, node_b, GraphConst.REAL_DATA_COMPARE)
mock_get_accuracy.assert_called_once()
self.assertIsInstance(result, list)
def test__format_decimal_string(self):
s = "0.123456789%"
formatted_s = _format_decimal_string(s)
self.assertIn("0.123457%", formatted_s)
self.assertEqual('0.123457', _format_decimal_string('0.12345678'))
self.assertEqual('-1', _format_decimal_string('-1'))
self.assertEqual('0.0.25698548%', _format_decimal_string('0.0.25698548%'))
def test__format_data(self):
data_dict = {'value': 0.123456789, 'value1': None, 'value2': "<class 'str'>", 'value3': 1.123123123123e-11,
'value4': torch.inf, 'value5': -1}
_format_data(data_dict)
self.assertEqual(data_dict['value'], '0.123456789')
self.assertEqual(data_dict['value1'], 'null')
self.assertEqual(data_dict['value2'], '<class str>')
self.assertEqual(data_dict['value3'], '1.123123123123e-11')
self.assertEqual(data_dict['value4'], 'inf')
self.assertEqual(data_dict['value5'], '-1')
all_none_dict = {'Max': None, 'Min': None, 'Mean': None, 'Norm': None, 'type': None}
_format_data(all_none_dict)
self.assertEqual({'value': 'null'}, all_none_dict)
def test_matched_node_calculator(self):
template = {'success': False, 'error': [
'The file path /home/louyujing/gitcode/MindStudio-Probe_3/test/msprobe_test/visualization_ut/builder/'
'Functional.conv2d.0.forward.input.0.pt does not exist.',
'The file path /home/louyujing/gitcode/MindStudio-Probe_3/test/msprobe_test/visualization_ut/builder/'
'Functional.conv2d.0.forward.input.1.pt does not exist.',
'The file path /home/louyujing/gitcode/MindStudio-Probe_3/test/msprobe_test/visualization_ut/builder/'
'Functional.conv2d.0.forward.output.0.pt does not exist.'],
'data': {'precision_index': 1, 'input_info': {
'Functional.conv2d.0.forward.input.0': {'Cosine': 'unsupported', 'EucDist': 'unsupported',
'MaxAbsErr': 'unsupported',
'MaxRelativeErr': 'unsupported',
'One Thousandth Err Ratio': 'unsupported',
'Five Thousandths Err Ratio': 'unsupported',
'Requires_grad Consistent': 'True', 'Result': 'pass',
'Err_message': '[]'},
'Functional.conv2d.0.forward.input.1': {'Cosine': 'unsupported', 'EucDist': 'unsupported',
'MaxAbsErr': 'unsupported',
'MaxRelativeErr': 'unsupported',
'One Thousandth Err Ratio': 'unsupported',
'Five Thousandths Err Ratio': 'unsupported',
'Requires_grad Consistent': 'True', 'Result': 'pass',
'Err_message': '[]'},
'Functional.conv2d.0.forward.input.2': {'Cosine': '', 'EucDist': '', 'MaxAbsErr': '',
'MaxRelativeErr': '', 'One Thousandth Err Ratio': '',
'Five Thousandths Err Ratio': '',
'Requires_grad Consistent': '', 'Result': 'pass',
'Err_message': '[]'},
'Functional.conv2d.0.forward.input.3.0': {'Cosine': 'unsupported', 'EucDist': 'unsupported',
'MaxAbsErr': 'unsupported',
'MaxRelativeErr': 'unsupported',
'One Thousandth Err Ratio': 'unsupported',
'Five Thousandths Err Ratio': 'unsupported',
'Requires_grad Consistent': 'True', 'Result': 'pass',
'Err_message': '[]'},
'Functional.conv2d.0.forward.input.3.1': {'Cosine': 'unsupported', 'EucDist': 'unsupported',
'MaxAbsErr': 'unsupported',
'MaxRelativeErr': 'unsupported',
'One Thousandth Err Ratio': 'unsupported',
'Five Thousandths Err Ratio': 'unsupported',
'Requires_grad Consistent': 'True', 'Result': 'pass',
'Err_message': '[]'},
'Functional.conv2d.0.forward.input.4.0': {'Cosine': 'unsupported', 'EucDist': 'unsupported',
'MaxAbsErr': 'unsupported',
'MaxRelativeErr': 'unsupported',
'One Thousandth Err Ratio': 'unsupported',
'Five Thousandths Err Ratio': 'unsupported',
'Requires_grad Consistent': 'True', 'Result': 'pass',
'Err_message': '[]'},
'Functional.conv2d.0.forward.input.4.1': {'Cosine': 'unsupported', 'EucDist': 'unsupported',
'MaxAbsErr': 'unsupported',
'MaxRelativeErr': 'unsupported',
'One Thousandth Err Ratio': 'unsupported',
'Five Thousandths Err Ratio': 'unsupported',
'Requires_grad Consistent': 'True', 'Result': 'pass',
'Err_message': '[]'},
'Functional.conv2d.0.forward.input.5.0': {'Cosine': 'unsupported', 'EucDist': 'unsupported',
'MaxAbsErr': 'unsupported',
'MaxRelativeErr': 'unsupported',
'One Thousandth Err Ratio': 'unsupported',
'Five Thousandths Err Ratio': 'unsupported',
'Requires_grad Consistent': 'True', 'Result': 'pass',
'Err_message': '[]'},
'Functional.conv2d.0.forward.input.5.1': {'Cosine': 'unsupported', 'EucDist': 'unsupported',
'MaxAbsErr': 'unsupported',
'MaxRelativeErr': 'unsupported',
'One Thousandth Err Ratio': 'unsupported',
'Five Thousandths Err Ratio': 'unsupported',
'Requires_grad Consistent': 'True', 'Result': 'pass',
'Err_message': '[]'},
'Functional.conv2d.0.forward.input.6': {'Cosine': 'unsupported', 'EucDist': 'unsupported',
'MaxAbsErr': 'unsupported',
'MaxRelativeErr': 'unsupported',
'One Thousandth Err Ratio': 'unsupported',
'Five Thousandths Err Ratio': 'unsupported',
'Requires_grad Consistent': 'True', 'Result': 'pass',
'Err_message': '[]'}}, 'output_info': {
'Functional.conv2d.0.forward.output.0': {'Cosine': 'unsupported', 'EucDist': 'unsupported',
'MaxAbsErr': 'unsupported',
'MaxRelativeErr': 'unsupported',
'One Thousandth Err Ratio': 'unsupported',
'Five Thousandths Err Ratio': 'unsupported',
'Requires_grad Consistent': 'True', 'Result': 'error',
'Err_message': '["error: There is nan/inf/-inf in the '
'maximum or minimum value of NPU."]'}}}}
m = MatchedNodeCalculator(npu_data, bench_data)
result = m.get_db_tensor_compare_result()
self.assertEqual(result.get('data'), template.get('data'))
self.assertEqual(result.get('success'), template.get('success'))
self.assertEqual(len(result.get('error')), len(template.get('error')))
template1 = {'success': False, 'error': ['The file path does not exist.', '[msprobe] 非法文件路径： '],
'data': {}}
m1 = MatchedNodeCalculator(npu_data1, bench_data1)
result1 = m1.get_db_tensor_compare_result()
self.assertEqual(result1, template1)
def test_compare_index(self):
tensor_index = ['Cosine', 'EucDist', 'MaxAbsErr', 'MaxRelativeErr', 'One Thousandth Err Ratio',
'Five Thousandths Err Ratio', 'Requires_grad Consistent', 'Result', 'Err_message']
self.assertEqual(MatchedNodeCalculator.TENSOR_COMPARE_INDEX, tensor_index)
def test_get_db_statistics_compare_result(self):
template = {"success": True, "error": None, "data": {"precision_index": 1, "input_info": {
"Module.bn1.BatchNorm2d.forward.0.input.0": {"Max diff": 0.0, "Min diff": 0.0, "Mean diff": 0.0,
"L2norm diff": 0.0, "MaxRelativeErr": "0.0%",
"MinRelativeErr": "0.0%", "MeanRelativeErr": "0.0%",
"NormRelativeErr": "0.0%", "Requires_grad Consistent": "True",
"Result": "pass", "Err_message": "[]"},
"Module.bn1.BatchNorm2d.forward.0.parameters.weight": {"Max diff": 0.0, "Min diff": 0.0, "Mean diff": 0.0,
"L2norm diff": 0.0, "MaxRelativeErr": "0.0%",
"MinRelativeErr": "0.0%", "MeanRelativeErr": "0.0%",
"NormRelativeErr": "0.0%",
"Requires_grad Consistent": "True", "Result": "pass",
"Err_message": "[]"},
"Module.bn1.BatchNorm2d.forward.0.parameters.bias": {"Max diff": 0.0, "Min diff": 0.0, "Mean diff": 0.0,
"L2norm diff": 0.0, "MaxRelativeErr": "N/A",
"MinRelativeErr": "N/A", "MeanRelativeErr": "N/A",
"NormRelativeErr": "N/A",
"Requires_grad Consistent": "True", "Result": "pass",
"Err_message": "[]"}}, "output_info": {
"Module.bn1.BatchNorm2d.forward.0.output.0": {"Max diff": 0.0, "Min diff": 0.0, "Mean diff": 0.0,
"L2norm diff": -728.544451, "MaxRelativeErr": "0.0%",
"MinRelativeErr": "0.0%", "MeanRelativeErr": "0.0%",
"NormRelativeErr": "89.99999993823302%",
"Requires_grad Consistent": "True", "Result": "error",
"Err_message": "[\"error: The NormRelativeErr of output is "
"greater than 0.5.\", \"warning: The norm "
"relative error of output is 10 times that "
"of input.\"]"}}}}
m = MatchedNodeCalculator(npu_statistics_data, bench_statistics_data)
result = m.get_db_statistics_compare_result()
self.assertEqual(result, template)
def test_get_db_md5_compare_result(self):
template = {"success": True, "error": None, "data": {"precision_index": 1, "input_info": {
"Module.conv1.Conv2d.forward.0.input.0": {"Requires_grad Consistent": "True", "Result": "error",
"Err_message": "[\"error: The CRC-32 value of NPU differs "
"from that of the bench.\"]"},
"Module.conv1.Conv2d.forward.0.parameters.weight": {"Requires_grad Consistent": "True", "Result": "pass",
"Err_message": "[]"}}, "output_info": {
"Module.conv1.Conv2d.forward.0.output.0": {"Requires_grad Consistent": "True", "Result": "error",
"Err_message": "[\"error: The CRC-32 value of NPU differs "
"from that of the bench.\"]"}}}}
m = MatchedNodeCalculator(npu_md5_data, bench_md5_data)
result = m.get_db_md5_compare_result()
self.assertEqual(result, template)
