Cambricon / mlu-ops

Efficient operation implementation based on the Cambricon Machine Learning Unit (MLU) .
MIT License
103 stars 102 forks source link

[WIP] add new operator LU factorization #1019

Open Chuancysun opened 6 months ago

Chuancysun commented 6 months ago

Thanks for your contribution and we appreciate it a lot. :rocket::rocket:

1. Motivation

add floating point operator LU factorization

2. Modification

add implementation of floating point LU factorization

3. Test Report

3.1 Modification Details

3.1.1 Accuracy Acceptance Standard

For static threshold standard details, see: MLU-OPS™ Accuracy Acceptance Standard.

3.1.2 Operator Scheme checklist

3.2 Accuracy Test

3.2.1 Accuracy Test

If you have checked the following items, please tick the relevant box.

3.3 Performance Test

Platform:MLU370

----------- case0 ----------- case0

[Shape ]: input.shape=[256,256], output.shape=[256,256]

[Diffs]: [output] DIFF1: 1.798500e-04 DIFF2: 7.016698e-04 [^ OK ] ../../test/mlu_op_gtest/pb_gtest/src/zoo/sgetrf/test_case/case0.prototxt [ OK ] sgetrf/TestSuite.mluOp/0 (36 ms) [----------] 1 test from sgetrf/TestSuite (36 ms total)

[----------] Global test environment tear-down [ SUMMARY ] Total 1 cases of 1 op(s). ALL PASSED. [==========] 1 test case from 1 test suite ran. (3727 ms total) [ PASSED ] 1 test case.

3.4 Summary Analysis

Please give a brief overview here, if you want to note and summarize the content.