Analysis Results

Float16 Overflow/Underflow Detection

Detects overflow/underflow errors in Float16 data of the comparison data. If overflow/underflow data exists, an expert suggestion is output.

Analysis result of Advisor:

Detection Type: Float16 overflow/underflow detection

Operator Index: 228

Expert Suggestion: Float16 data overflow/underflow occurs. Rectify the fault and perform comparison again.

Input Inconsistency Check

Checks whether the two batches of input data to be compared on the entire network are consistent. If inconsistency exists (the cosine similarity is less than 0.99), an expert suggestion is output.

Analysis result of Advisor:

Detection Type: Input inconsistency check

Operator Index: 0

Expert Suggestion: The input data of NPUDump is inconsistent with that of GroundTruth. Use the same data or check the data preprocessing process.

Network-wide Consistency Check (Faulty Node Detection)

In the network-wide comparison result, checks whether the value of a layer is less than the threshold, whether the values of subsequent layers are less than the threshold, or whether the value of the last layer is less than the threshold (the cosine similarity is less than 0.99), and outputs quantized error correction suggestions.

Analysis result of Advisor:

Detection Type: Network-wide consistency check

Operator Index: 1174

Expert Suggestion: The accuracy of some tensors is low, resulting in an unqualified final accuracy. This may be caused by quantization. Calibrate the data or contact Huawei for further location.

Network-wide Consistency Check (Single-Point Error Detection)

In the network-wide comparison result, checks that the value of a layer is less than the threshold (the cosine similarity is less than 0.99), but the final result meets the accuracy requirement, and outputs the expert suggestion.

Analysis result of Advisor:

Detection Type: Network-wide consistency check

Operator Index: 195

Expert Suggestion: The accuracy of some tensors is low, but the final accuracy is qualified. This may be caused by internal optimization. Ignore this problem or contact Huawei for further location.

Network-wide Consistency Check (Consistency Check)

Checks that all data in the comparison result meets the accuracy requirements, and outputs the expert suggestion.

Analysis result of Advisor:

Detection Type: Network-wide consistency check

Operator Index: N/A

Expert Suggestion: All data in the comparison result meets the accuracy requirements.

If data accuracy problems occur in the use of actual models, check postprocessing of model outputs.