AlexanderMath

Sinclair suggests aging is caused by loss of epigenetic information [1]. The MOSES dataset contains chemical information for small molecules, but no information on how small molecules alter the epigenetics...

Initial Colab Notebook Demonstrating Inference

2

Great work. It takes some time (1-2 hours) to get all dependencies right and use your model for inference. I made a notebook so anyone can use it for inference...

Error in Likelihood Computations?

1

It seems there is an error in the NLL computations on MNIST. The article reports NLL in "bits per pixels". The "per pixel" part is computed by dividing by the...

Squeeze implementations differs to that reported in article.

1

The article briefly describe the architecture, and refer to Real NVP for more details.  This lead me to believe Glow implemented the Squeeze operation as done by Real NVP....

In ```README.md``` there is a section with a link to pre-trained models: > Clicking the link leads me to > The page https://alpha.openai.com/ also turns up dead, but https://openai.com/ works...

Keras version and ?tensorflow? backend version?

2

Which version of keras was used for this implementation? Also, which backend was used, and which version of the backend?

"too many indices for tensor of dimension 4"

2

After installation (across three different setups) the following code does not work: ``` from butterfly import Butterfly Butterfly(10,10) ``` **Error:** "too many indices for tensor of dimension 4". **Question.** Can...

Autograd error: Inplace operations modify variable needed for gradient computations.

3

I followed instructions from ```README.md``` and ran the following ```python train.py --batch_size 1 --gpu_ids [0]``` This yields the following error: ```RuntimeError: one of the variables needed for gradient computation has...

Classifier OOM when computing on test set.

1

Thanks for a great repository, the code works very well, is nicely documented and the overall structure is intuitive. I found a minor issue which can easily be solved. The...

BUG: jnp.cumsum(np.arange(2**14)) gives segment fault.

2

### Description Reproducer ``` import numpy as np import jax import jax.numpy as jnp # works #range = np.arange(2**13) #print(np.cumsum(range)) #print(jax.jit(jnp.cumsum, backend="ipu")(range)) # gives segment fault range = np.arange(2**14) print(np.cumsum(range))...