Can you explain the data format?

[ChrisWu1997]

Hi, I want to re-trained the network on my own data and I notice that each array entry in prim_gt folder is of shape (num_of_parts, 20). Can you explain what do these 20 parameters mean?

Also, what's the use for prim_rnn_batch_nz_all_tr.t7 and prim_rnn_batch_nz_all_val.t7?

Thanks in advance!

[zouchuhang]

@ChrisWu1997 You can refer to the format in the data loader file "getBatch.lua". In general every three params refers to an instance as described in Sec. 4.1 "Recurrent generator" of the paper.

[ChrisWu1997]

Thanks. But can you explain more about x_vals, y_vals and r_vals for each entry in prim_rnn_batch_nz_all_tr.t7? Are they direct values for scale, translation and rotation? Or have been normalized?

I am trying to understand those values because it's direct data source for training. I'm now having my own dataset of primitives(with OBB parameters), so I want to know whether there is a way to process the data into your format?