Frederik Heber
Frederik Heber
Hey Marios, I stumbled upon your project which seems perfect for adding SQL querying to the internal state of my molecular dynamics software package molecuilder. I made some smaller enhancements...
CCAdL reproducibility regression test was working in branch *Averages_burnin* and was broken during the introduction of the ensemble of walkers in the huge branch *WalkerEnsemble*. See the description in the...
Usually, this occurs when trying to pass an array where a float is expected. This needs to be looked at on a Mac system.
Most of the examples are small code snippets inside the guide and *not* in external files as with the user guide. There are two approaches: 1. Extract snippets from a...
There are some more issues when running things in parallel that I just stumbled upon. These are partly related to the new input pipeline, see #22. - [ ] `cache()`...
@ZofiaTr wrote in #51: > I was also thinking, that it might be very useful if Ben @teebee2 could read some chosen parts of the userguide. I suggested some modifications...
`init_network()` is quite clumsy and the `setup` parameter is ont very intuitive. Maybe we could change this into an initial `init_network()` (which might be hidden in some general `init()` function...
TATi is so far limited to (fully) feed-forward networks. However, for image classification Convoluted NNs are better, for signal processing Long-Short-Term Memory (LSTM) NNs. Tensorflow 1.9 offers `tf.keras` as the...
Gabriel suggested to add an option to only evaluate the diagonal hessians instead of the full hessian matrix.
The sampler code in, e.g., src/TATi/samplers/baoabsampler.py adds nodes to the graph to perform a single sampling step for each of the parameter matrices. TATi creates two parameter matrices per layer,...