Anders Ohrn

AEs have been used by themselves for image clustering. The idea is: 1. The AE discovers the most efficient way to create a compact vector representation of the image data...

Generally, the files with names ending on `_runs.py` are the ones I execute. See for example [ae_runs.py](https://github.com/anderzzz/monkey_caput/blob/master/ae_runs.py). However, you should rewrite these or implement it as a Jupyter Notebook. The...

Hello. Sorry for late reply. The dataset is a PyTorch Dataset. https://pytorch.org/docs/stable/data.html The image data set enter the code here: https://github.com/anderzzz/monkey_caput/blob/d4789446636953377d1c709207ce15421897eeb9/_learner.py#L104 The custom datasets I create are subclasses of the...

Not sure what's causing that. You are running Python 3? The abstract class is contained in [_learner.py](https://github.com/anderzzz/monkey_caput/blob/767f5d1c817fb6ff2c03bcdafa028f64930854d2/_learner.py). Have you made any modifications to `AELearner`, like removing methods? The abstract base...

Hello. That's the path to the raw data for training and testing. It is not included in the repo. You can replace that path with a folder to your own...

I presume [the running loss](https://github.com/anderzzz/monkey_caput/blob/767f5d1c817fb6ff2c03bcdafa028f64930854d2/ae_learner.py#L124) is increasing then during each epoch? That is the optimizer is diverging rather than converging? I noticed myself that training the AE was by no...

Did any of my comments help? The AE optimization is tricky since the starting parameters are random.