search

stared / livelossplot

Live training loss plot in Jupyter Notebook for Keras, PyTorch and others
https://p.migdal.pl/livelossplot
MIT License
1.29k stars 143 forks source link

two small features request? #54

Closed Jiraiyah closed 5 years ago

Jiraiyah commented 5 years ago

@stared Hi. I would like to have two small features, take a look here :

Mean squared error (cost function): training (min: 0.000, max: 0.000, cur: 0.000) validation (min: 0.000, max: 0.000, cur: 0.000)

Accuracy: training (min: 0.549, max: 0.549, cur: 0.549) validation (min: 0.355, max: 0.355, cur: 0.355)

this won't be enough for many purposes on precision, why? first 10 epochs of my neural starts right off of 0.00001 epoch loss !!!! the top table is pretty much useless with that precision.

Also, a nice thing to have before the tables would be a line that shows the finishing time on last epoch (how much time the last epoch took in total) side by side with the last epoch's number (when you run 3000 epochs, reading what epoch you are in from the chart is a headache)! I would appreciate if you add these two features. for the precision, maybe you can accept an integer into the function that would let us define how many digits in decimal place we need?

Jiraiyah commented 5 years ago

On a side note, how about having an alternative to use plotly instead of matplot? that library let us have offline charts in notebook but those charts support zooming and have nice features.

stared commented 5 years ago

@Jiraiyah There is an idea to create Bokeh backend, so one would be able to zoom plots: #19. Feel invited to implement it.

For printed logs - there will be a way to add custom log prints (and plots), but I need to think well about creating a good modularization (or ideally - someone else, with way more experience in software engineering in Python).

Jiraiyah commented 5 years ago

@Jiraiyah There is an idea to create Bokeh backend, so one would be able to zoom plots: #19. Feel invited to implement it.

For printed logs - there will be a way to add custom log prints (and plots), but I need to think well about creating a good modularization (or ideally - someone else, with way more experience in software engineering in Python).

that person with good python experience ... definitely not me ! totally noob in python :)

offchan42 commented 4 years ago

Is this feature implemented yet? Just add digits as another parameter to __init__() function of PlotLossesCallback(). It shouldn't be hard.