To people asking for a debugging guide, @josh_tobin_ wrote a pretty comprehensive one: http://josh-tobin.com/assets/pdf/troubleshooting-deep-neural-networks-01-19.pdf …
-
-
-
Also, do use `python -m ipdb http://script.py ` for post-mortem debugging. Those things should be taught in all introductory ML courses in Python. I believe I did my due diligence my including it in
@MITxonedX's ML course 6.86xhttps://www.edx.org/course/machine-learning-with-python-from-linear-models-to-deep-learning-3 …
Kraj razgovora
Novi razgovor -
-
-
No print(“\n\nHEREEEE——”) ?
-
This is the only way
Kraj razgovora
Novi razgovor -
-
-
Debuging codes by statring at them is so satisfying
But ideas in this thread can come handy at times.https://twitter.com/radekosmulski/status/945739571735748609 …
-
Open debug in the console (right click on the tab I think) and you will jump for joy.
Kraj razgovora
Novi razgovor -
-
-
Was he coding on Jupyter notebook?
-
Hey now, Copy pasta in to colab is a perfectly valid workflow
- Još 2 druga odgovora
Novi razgovor -
-
-
I use pdb, unittests, build visualizations of data and predictions, monitor training and validation metrics, and execute ML-specific debugging strategies (overfitting to a small dataset, or a synthetic one). At the end of the day, I still stare at my screen, cursing my model.
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
I google for idpb and ohh my.
-
Inveraray & District Pipe Band ?
- Još 2 druga odgovora
Novi razgovor -
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.