-
It doesn't seem clear if the algorithm supports weighted edges. If it does should the edge simply be repeated in the input file multiple times?
I apologize if it is somewhere in the documentation a…
-
Is there currently a way to have update-able buffers that aren't parameters in the stax framework (and so won't get updated by gradient descent but by some other user-specified method)?
This is ve…
-
#### Issue Description
I am trying to use a CNN to predict a move policy for an arcade video game. The framework for the game is written in Java and therefore I can directly save the input of the n…
-
**Describe the bug**
For the projected gradient descent attack, we need to perform two projections.
1) ensure that the perturbation do not exceed the maximum allowed perturbation
2) ensure that the…
-
The NNUE branch maintained by @nodchip has demonstrated strong results and offers great potential, and we will proceed to merge it into master. This will assure that Stockfish remains a reference engi…
-
When I train the 2 stages, I find that the loss changes in a small oscillation from beginning to end even if it falls in a general. I think it is unrelated to the fix of learning rate for I just use t…
-
Build ops for computing the loss using MonteCarlo returns and the update step of gradient descent over the Value function parameters given a TensorFlow optimizer and a batch of experiences.
-
line 103: Why change the direction of the gradient?I think this step is not required. I do not really understand your meaning.
-
## Original
```
means = X_train.mean(axis=0, keepdims=True)
stds = X_train.std(axis=0, keepdims=True) + 1e-10
X_val_scaled = (X_valid - means) / stds
with tf.Session() as sess:
init.run…
-
Hey psanch21!
I just run the code:
`python3 GMVAE_main.py --model_type=2 --dataset_name=MNIST --sigma=0.001 --z_dim=8 --w_dim=2 --K_clusters=8 --hidden_dim=64 --num_layers=2 --epochs=20 --batch_si…