JLC

Same I get 0 unsup_loss bbox even though I used `load_from` with pretrained model. Sometimes it starts at a small value but eventually goes to zero and during evaluation there...

Yes I trained a model using the a config in config/baseline/ and it seems to do well: ```Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.691...

I get the same problem but using mmdet 2.17 still doesn't work for me.

Looking into this more, there is also some discrepancy when compared with paper's results for semi-supervised case. In the relevant lines of code, I set Adam optimizer as follows: ```...

I ran more trials with Adam at LR of 0.02 and decay after 15 and default TF setting. Here are results for the three models:    The convnets...

Yeah. I am also having a lot of trouble. What versions do we use?