Overview
Huawei supports the migration of the original network for training on the Ascend 910 AI Processor. As a result, the computation result of Huawei proprietary operators may be different from that of third-party operators (for example, from TensorFlow). The Model Accuracy Analyzer is designed to compare the computation results for developers to quickly resolve the operator accuracy issues.
The Model Accuracy Analyzer compares the computation result of the model running on the Ascend 910 AI Processor with that of the original model running on the GPU or CPU to locate the error operators.
This document describes how to use Model Accuracy Analyzer with the assumption that the tool is installed by HwHiAiUser in the default installation path /home/HwHiAiUser/Ascend. You can replace them as required. Ensure that HwHiAiUser has the read, or read and write permissions on the paths described in this document.
For an optimal experience, the following hardware configuration is recommended: 8-core CPU at 2.6 GHz with 16 GB memory.
The following data types are supported:
- FLOAT
- FLOAT16
- DT_INT8
- DT_UINT8
- DT_INT16
- DT_UINT16
- DT_INT32
- DT_INT64
- DT_UINT32
- DT_UINT64
- DT_BOOL
- DT_DOUBLE
