OverLordGoldDragon

OverLordGoldDragon has asked 18 questions and find answers to 57 problems.

Stats

About

Highlights:

Wavelet Scattering explanation
Joint Time-Frequency Scattering explanation
Synchrosqueezing Wavelet Transform explanation
Why is TensorFlow 2 much slower than TensorFlow 1?
Cascade multiple RNN models for N-dimensional output

OverLordGoldDragon questions

How to save user data in electron.js app?
Why is [0] a different function but 0 isn't?
Different slices give different inequalities for same elements
Faster for-loops with arrays in Python
How to visualize RNN/LSTM gradients in Keras/TensorFlow?
How to edit Google Colaboratory libraries?
Efficient elementwise argmin of matrix-vector difference
How to get value of a Keras tensor in TensorFlow 2?
LambdaType vs FunctionType
What is the difference between tf.keras and tf.python.keras?

OverLordGoldDragon answers

What does "model.trainable = False" mean in Keras?
If Keras results are not reproducible, what's the best practice for comparing models and choosing hyper parameters?
Keras model fails to decrease loss
Why are gradients incorrect for categorical crossentropy?
How can I retrieve the current seed of NumPy's random number generator?
Anaconda installs TensorFlow 1.15 instead of 2.0
Conda - unable to completely delete environment
How does Keras load_data() know what part of the data is the train and test set?
Doing prediction with RNN on Sequential data in keras
The clear_session() method of keras.backend does not clean up the fitting data