Matt Whiteway

It appears you are still using the 3 csv files (I think). You'll need to concatenate them into a single csv file and then use that name in the `data.csv_file`...

yes, I think the nomenclature here is a bit confusing. The "multi-view" training referred to in the LP documentation specifically means training on labels present from all views at a...

There are two options: 1. use your current labeled frames to train a "single view" model, which, because it is trained on frames from multiple cameras, will actually be a...

@wyclearnpy just wanted to check in to see if you're all set here - if so I'll close the issue

@wyclearnpy there shouldn't be any issue with you labeled data. If you are using frame 0, for example, the code will automatically load frames [0, 0, 0, 1, 2]. If...

@wyclearnpy just wanted to check in on this and see if you were able to train with an increased unlabeled batch size?

@wyclearnpy glad you were able to get the unsupervised model training. One comment is that to really compare the semi-supervised and supervised models you should set the resizing to be...

ok, so I would suggest the following type of experiment to compare supervised and semi-supervised (both context): * set resize dims to 256x256 (can change this later) * set `training.train_batch_size`...

@wyclearnpy I would recommend comparing two models first: fully supervised (already done), and supervised context (i.e. `model_type: heatmap_mhcrnn` and `losses: []`). Importantly, you should make every other parameter the exact...

very cool, thanks for sharing. I would suggest for now only turning on the "temporal" unsupervised loss. The other losses will not work well for this setup for the following...