search

blakechen97 / SASA

SASA: Semantics-Augmented Set Abstraction for Point-based 3D Object Detection
Apache License 2.0
88 stars 14 forks source link

experimental results #7

Open pummi823 opened 2 years ago

pummi823 commented 2 years ago

Hi, Dear author! I would want to know, your public results of 3DSSD+SASA whose moderate's is 85.91, is trained by"80% train+20%val" or 70&30? Waiting for your answer~~~~

blakechen97 commented 2 years ago

Hi, the pre-trained models provided in this repo are all trained under the commonly used train-val setting (3,712 samples for training and 3,769 samples for validation). We use the 80%-20% division only when submitting results to the official test server.

pummi823 commented 2 years ago

OK！！！！！thank you very much kind author!!!!!

twn29004 commented 2 years ago

How can I get the Imagesets of 80%-20%?

pummi823 commented 2 years ago

How can I get the Imagesets of 80%-20%?

you can split by yourself, just modify the .txt files in ImageSets

twn29004 commented 2 years ago

How can I get the Imagesets of 80%-20%?

you can split by yourself, just modify the .txt files in ImageSets

I have the random split, but the result of val is too high. Is there any tricks?

pummi823 commented 2 years ago

How can I get the Imagesets of 80%-20%?

you can split by yourself, just modify the .txt files in ImageSets

I have the random split, but the result of val is too high. Is there any tricks?

I also do the random split without any trick

twn29004 commented 2 years ago

How can I get the Imagesets of 80%-20%?

you can split by yourself, just modify the .txt files in ImageSets

I have the random split, but the result of val is too high. Is there any tricks?

I also do the random split without any trick Thank you very much for your reply.