Is train dataset config is correct?

[frkake]

Below train config setting is correct?
It looks ignoring seen classes even during training.
I wonder why this data class is called during training.

Zero-shot-Instance-Segmentation/configs/zsi/65_15/train/zero-shot-mask-rcnn-BARPN-bbox_mask_sync_bg_65_15_decoder_notanh.py

Line 141 in 6b25b31

dataset_type = 'CocoDatasetUnseen15'

Zero-shot-Instance-Segmentation/configs/zsi/48_17/train/zero-shot-mask-rcnn-BARPN-bbox_mask_sync_bg_decoder.py

Line 142 in 6b25b31

dataset_type = 'CocoDatasetUnseen17'

I think that CocoDataset_48_17 and CocoDataset_65_15 data class looks appropriate.

Thank you a lot!

[zhengye1995]

Sorry for this confusing code. In fact, it only affects the category name during visualization.The data and categories used for training are completely determined by the label file

Zero-shot-Instance-Segmentation/configs/zsi/48_17/train/zero-shot-mask-rcnn-BARPN-bbox_mask_sync_bg_decoder.py

Line 176 in 6b25b31

ann_file=data_root + 'annotations/instances_train2014_seen_48_17.json',

. (note: 'dataset_type' affects training in the new version of mmdetection, but has no effect in this old version)

[frkake]

@zhengye1995
Thank you so much for fast response!
I'm convinced very well!

By the way, what version of mmdetection is based in this repository?
I would like to develop this in upper version of it.

[zhengye1995]

mmcv 0.2.14
mmdet 1.0rc0+45ee140

[frkake]

@taku-buntu Thank you so much!