-
We have made certain extensions (see [patched torch distributions](https://github.com/uber/pyro/blob/dev/pyro/distributions/torch.py)) to `torch.distributions` classes within Pyro, which we thought ma…
-
I saw the model in compilation, `metrics = [vamp. metric_VAMP, vamp. metric_VAMP2]`, and then I saw the training visualization as follows:
 which Jupinx can't support them like :
``R`` needs to be a matrix where ``R[s, a]`` is the reward at state ``s`` under action ``a``
![…
-
e.g. in `dirichlet()`, `dirichlet_multinomial()`, `categorical()` etc., the parameter (`alpha` or `prob`) can only be a vector of length `k`, and all `n` rows of the response are assumed to be identic…
-
Hello, i'm trying to make work a Hidden Markov Model with continuous observations. I try to fit the model using a single sequence of 3 normalized histograms (each histogram is an observation) but it s…
-
This bug is easy to reproduce:
```
def printA(group):
print group.A.values[0]
return 0
pandas.DataFrame({'A': [0,0,1,1], 'B': [0,1,0,1]}).groupby('A').apply(printA)
```
This should print
``…
-
While working on its calculation.. we are using cluster.dtrajs. why are we not using dtrajs concatenated as they represent the complete list of trajectories using following code?
its = pyemma.msm.…
-
I have some difficulties in understanding some parts of the new code so I thought of writing things here. It provides a better overview then email or gitter.
So, we have a three state hidden markov…
-
Use physical units instead of things like "steps" whenever they are available. Please check all notebooks for that.
![image](https://user-images.githubusercontent.com/6495810/42116320-fe5371da-7bf6…