Geometric variable not being properly Sampled

Questions
21

Yes however the follow code would not work:

    for i in range(x.shape[0]):
        v = x[i,:]
        v = v[~np.isnan(v).any(axis=1)]

as they are numpy procedures and not theano. Do you know anyway to be able to efficiently get the likelihood for the values that are there but return a 0 for nans? This would deal with any different vector lengths and also get around list implementation

Edit: Getting the same results using Potential: ll = pm.Potential('likelihood', logp(px, drift, cov, inital_pos["CT"].mu, inital_pos["CT"].cov)). The logp function no longer has the as_op decorator but still getting the same error.

Edit: cleared the cache, the potential method seems to be sampling albeit I get a type error sometimes (this has happened throughout a lot of my computations): TypeError: expected type_num 11 (NPY_FLOAT32) got 12. I am using the cuda gpu to train it

Edit v4: Seems like removing floatx=float from theanorc fixed it. It is now sampling at around ~16 it/s for metropolis with the potential method, will try the pm.switch workaround + using NUTS/ADVI and see how they work.

22

Thanks for the update.

Just a quick heads up: training under GPU has not been tested extensively, and there are quite a few known issue using GPU (e.g., here). For now, try to make sure your model works on CPU first.

Keep me posted!

23

Just wanted to update that my original code samples perfectly under cpu (and even faster under single core!!) :rage:. nevertheless I made a few switches based on your original suggestion so i definitely appreciate that.