cyberbeat

BinaryEvaluation often results in NaN for best F1-Scores

1

I then printed out all F1-Scores of the ConfusionMatrices-List, which looks as expected, but has some NaN at the end. Something like Threshold: F1-Score 0.1: 0.1 0.2: 0.2 0.4: 0.3...

Logging frequence has no effect on DL4J logging intervals

3

No matter what I set as logging frequence in "NeuralNetwork" (for example TrainingAmount.minibatches(1) or TrainingAmount.minibatches(300)), every iteration is logged. I have to admit, that I got new hardware, and compiled...

Add CRF (Conditional Random Fields) Layer

1

They are quite popular for POS/NER tagging

This would be a nice feature. Often there is not enough time doing this, when adding tasks.

IMU Calibration

3

I wonder, why the IMU is not calibrated by factory configuration. Could you at least provide some "factory spec" one, which I could upload to eeprom? In eeprom we now...

What is the license of your code?

Wish for using other features for tracking

6

For now you offer HARRIS or SHI_THOMASI features. But for applications, where later registration is required, it would be nice to use something like SIFT, which could be followed with...

You may also pick up also this link: https://github.com/SpectacularAI/HybVIO

It would be nice, if I could enter not only the buy-price, but also the date, and then have a red(or other color) vertical-line in the history-chart view to mark...

Often, there are some graphics, which are splitted in many small lines, which I would like to remove. I cannot select all and delete them all at once, but only...