[Fix] Supplying the model for lyft trained by pointpillars

[wHao-Wu]

• Supplying the model for lyft trained by pointpillars with FPN architecture
• Supplying the model for lyft trained by pointpillars with SECFPN architecture

[codecov]

Codecov Report

Merging #578 (df8efc5) into master (b3a6346) will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master     #578   +/-   ##
=======================================
  Coverage   50.77%   50.77%           
=======================================
  Files         197      197           
  Lines       14923    14923           
  Branches     2426     2426           
=======================================
  Hits         7577     7577           
  Misses       6851     6851           
  Partials      495      495           

Flag	Coverage Δ
unittests	50.77% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update b3a6346...df8efc5. Read the comment docs.

[yezhen17]

Checked that the links are valid.

[Tai-Wang]

Please also modify the corresponding benchmarks in other config folders like regnet and ssn.

[wHao-Wu]

Please also modify the corresponding benchmarks in other config folders like regnet and ssn.

Thanks for your advice. Regnet and ssn model on Lyft is still in progress. I will release those models after completion.