157 - What is TensorBoard and how to launch it in a browser?

TensorBoard can be very useful to view training model and loss/accuracy curves in a dashboard. This video explains the process of setting up TensorBoard callback during training and also to open the recorded logs in a browser.
Dataset downloaded from: archive.ics.uc...)
Code generated in the video can be downloaded from here:
github.com/bns...

[@AldairHoyos-zd8kc]

Thanks a lot for providing this great tutorial, I didn't know how to run it but you explained clearly and I got exaclty what I wanted, to learn. Thanks a lot again and have a nice day my friend

[@khalidnajam3567]

Thank you so much for this, was wondering how to plot the Precision-Recall curve via tensorboard, couldn't find anything on the internet yet.

[@georgekokkinakis7288]

One question please . I've notice that you use fit and transform in X and then you do the test_split. Is this correct? Should't be fit and transform in X_train and then transform in the Xtest?

[@kerimmirzeyev4340]

Thank you so much for these videos, but I do have a one question, I'm currently doing super resolution with the ISR Library in python and it doesn't have built in 'callback' function for me to actually use it on the training and then plot the necessary graphs, my question is that how can I approach this problem and try to fix it?

[@thdestiny4907]

if i have stack model how to make tensor board for it note that my model made of first lstm and second liner regration

[@radiga925]

when i paste the link on chrome, im getting a blank white page with orange tensorboard bar...My events are not loading..any idea how to fix this?

[@ambus6818]

Thats how you teach stuff. Thanks was losing my mind over this

[@pedar]

thanks a lot man!
great video

[@Readwithaami]

To compare performance of 2 models like mobile net and efficient net can I use tensor board

[@kurniaanwarraif3688]

thank you for the explanation, clear and nice !

[@shashireddy7371]

Thanks Sreeni

[@ernestoyounes2946]

how do i make my tensorboard shows acc, loss, val_acc, val_loss instead of epoch_accuracy and epoch_loss?

[@DigitalSreeni]

Tensorboard is showing exactly that, but for each epoch. This is why the titles epoch_accuracy and epoch_loss.

[@MohammadFarhadBulbul]

Thank you very much. Really helpful

[@DigitalSreeni]

Glad it was helpful!

[@ahasanhabibrafy8092]

Thank you so much

[@miracleaniakor5393]

Please How do I delete previous log, to prevent overlapping, I am tired of using different names. I want to delete the history of event. How I do go about it?

[@DigitalSreeni]

Manually delete the logs folder and see if that answers your question. Unfortunately, TensorBoard can be painful if you are using it for day to day training. It is very helpful when you have loads of data for a new application and you are trying to understand affects of various parameters. Personally, I just plot the curves myself and rarely use TensorBoard.

[@miracleaniakor5393]

@@DigitalSreeni I tried deleting the logs directory but it refused to delete. Thank you for your Feedback 🙂

[@thdestiny4907]

dude you amazing thank you so much

[@rameshwarsingh5859]

Sreeni these are Precious _Tutorials ..

[@DigitalSreeni]

Thanks

[@rameshwarsingh5859]

@@DigitalSreeni 🤍🤍🤍

[@rameshwarsingh5859]

@@DigitalSreeni sreeni,may i know how much calculus is needed for the deep learning,

[@mahyasadatebrahimi3164]

خدا خیرت بده اسرینی:)

[@sanjeevdurge2876]

nice sir!

[@DigitalSreeni]

Thanks!

[@vasileioskosmidis1421]

very heavy accent

[@DigitalSreeni]

Sorry for that