# \#from\_github **URL:** https://discourse.pymc.io/tag/from-github/1.md [Latest](https://discourse.pymc.io/latest.md) · [Categories](https://discourse.pymc.io/categories.md) · [Tags](https://discourse.pymc.io/tags.md) --- ## [Theano TypeError](https://discourse.pymc.io/t/theano-typeerror/1041)
**Author:** [@marcio](https://discourse.pymc.io/u/marcio)\ **Replies:** 2\ **Last updated:** [December 5, 2021, 3:06am UTC](https://discourse.pymc.io/t/theano-typeerror/1041 "2021-12-05T03:06:36Z")
Hi all, after try to sample I’m receiving this (for every model): Auto-assigning NUTS sampler… Initializing NUTS using jitter+adapt\_diag… /usr/local/lib/python3.5/dist-packages/pymc3/model.py:384: FutureWarning: Conve… --- ## [Problem with pymc3.DensityDist() #2846](https://discourse.pymc.io/t/problem-with-pymc3-densitydist-2846/843)
**Author:** [@LuisP1](https://discourse.pymc.io/u/LuisP1)\ **Replies:** 3\ **Last updated:** [February 8, 2018, 9:59pm UTC](https://discourse.pymc.io/t/problem-with-pymc3-densitydist-2846/843 "2018-02-08T21:59:47Z")
Hello everybody, In the last days I have been working with PyMC3. I am new using it and I am interested in make my own model for parameter estimation. Basically, my problem is that I am interested in using a different L… --- ## [Prediction using sample\_ppc in Hierarchical model](https://discourse.pymc.io/t/prediction-using-sample-ppc-in-hierarchical-model/116)
**Author:** [@junpenglao](https://discourse.pymc.io/u/junpenglao)\ **Replies:** 6\ **Last updated:** [December 14, 2017, 2:35pm UTC](https://discourse.pymc.io/t/prediction-using-sample-ppc-in-hierarchical-model/116 "2017-12-14T14:35:21Z")
Github issue #2352 by nikzadb: I have the attached data and following Hierarchical model (as a toy example of another model) and trying to draw posterior samples from it (of course to predict new values). new\_data.txt … --- ## [Possible Performance Issue Using Sparse Matrices in Bipartite Noisy-Or Model](https://discourse.pymc.io/t/possible-performance-issue-using-sparse-matrices-in-bipartite-noisy-or-model/319)
**Author:** [@PriceHardman](https://discourse.pymc.io/u/PriceHardman)\ **Replies:** 10\ **Last updated:** [September 15, 2017, 5:38pm UTC](https://discourse.pymc.io/t/possible-performance-issue-using-sparse-matrices-in-bipartite-noisy-or-model/319 "2017-09-15T17:38:02Z")
Github issue #2558: Hello, I’m trying to implement a certain model using pymc3, and have been running into what I believe are some severe performance issues. My system: 2015 Macbook Pro, 3.1GHz i7, 16GB RAM Python 3… --- ## [Sample\_ppc fails on traces containing a subset of variables](https://discourse.pymc.io/t/sample-ppc-fails-on-traces-containing-a-subset-of-variables/323)
**Author:** [@junpenglao](https://discourse.pymc.io/u/junpenglao)\ **Replies:** 0\ **Last updated:** [September 13, 2017, 7:00am UTC](https://discourse.pymc.io/t/sample-ppc-fails-on-traces-containing-a-subset-of-variables/323 "2017-09-13T07:00:30Z")
Github issue #2561 Trying to use sample\_ppc using a trace that does not contain all of the model’s RVs fails with “theano.gof.fg.MissingInputError: Undeclared input”. Basically, I want to sample the posterior predictiv… --- ## [Scipy.optimize.nnls usage problem in pymc3](https://discourse.pymc.io/t/scipy-optimize-nnls-usage-problem-in-pymc3/220)
**Author:** [@sezenyg](https://discourse.pymc.io/u/sezenyg)\ **Replies:** 5\ **Last updated:** [August 4, 2017, 8:30pm UTC](https://discourse.pymc.io/t/scipy-optimize-nnls-usage-problem-in-pymc3/220 "2017-08-04T20:30:38Z")
Hello everyone, I am trying to implement a graphical modelling in pymc3 and I probabilistically generate a matrix and a vector. I need to find a third vector which provides A\*x=b equality where x has nonnegative element… --- ## [Restructure docs](https://discourse.pymc.io/t/restructure-docs/76)
**Author:** [@owlas](https://discourse.pymc.io/u/owlas)\ **Replies:** 5\ **Last updated:** [June 30, 2017, 7:57pm UTC](https://discourse.pymc.io/t/restructure-docs/76 "2017-06-30T19:57:41Z")
Moved discussion from Github issue #2132 Colin suggested splitting out the PyMC3 API docs / tutorials from the more academic (“What is Bayes theorem?”) documentation. Readthedocs provides versioning and hosting for sph… --- ## [Perplexity of a model](https://discourse.pymc.io/t/perplexity-of-a-model/25)
**Author:** [@junpenglao](https://discourse.pymc.io/u/junpenglao)\ **Replies:** 21\ **Last updated:** [June 29, 2017, 8:54am UTC](https://discourse.pymc.io/t/perplexity-of-a-model/25 "2017-06-29T08:54:07Z")
Github issue #2297 by pwl: I noticed that in some papers people determine whether the model converged based on perplexity on some data set aside earlier for testing. Is there a way to compute the perplexity of a mod… --- ## [Geometric variable not being properly Sampled](https://discourse.pymc.io/t/geometric-variable-not-being-properly-sampled/28)
**Author:** [@junpenglao](https://discourse.pymc.io/u/junpenglao)\ **Replies:** 22\ **Last updated:** [June 22, 2017, 4:39am UTC](https://discourse.pymc.io/t/geometric-variable-not-being-properly-sampled/28 "2017-06-22T04:39:47Z")
Github issue #2303 by kpei: I am new to pymc3 and I have scanned through the list of outstanding/closed issues and couldn't find an exact solution so I am posting it here. If there is any existing soln or written pi… --- ## [Metropolis taking too long to initialize](https://discourse.pymc.io/t/metropolis-taking-too-long-to-initialize/86)
**Author:** [@san7988](https://discourse.pymc.io/u/san7988)\ **Replies:** 6\ **Last updated:** [June 16, 2017, 9:00pm UTC](https://discourse.pymc.io/t/metropolis-taking-too-long-to-initialize/86 "2017-06-16T21:00:51Z")
Hi, I'm trying to use metropolis method for sampling. It takes too long to initialize the metropolis method. The code : step = pm.Metropolis() This single line takes around 3 minutes to execute. The sampling itself… --- ## [Leave One Out Cross-validation and RMSE](https://discourse.pymc.io/t/leave-one-out-cross-validation-and-rmse/88)
**Author:** [@marcodena](https://discourse.pymc.io/u/marcodena)\ **Replies:** 11\ **Last updated:** [June 16, 2017, 2:20pm UTC](https://discourse.pymc.io/t/leave-one-out-cross-validation-and-rmse/88 "2017-06-16T14:20:54Z")
Hello, I'm trying to compare the results of my model with a model that uses a frequentist approach. This frequentist model uses a LOO Cross-validation and gives the RMSE as result. Thus, my naive solution is to fit the… --- ## [Getting observations of distribution variables whilst sampling](https://discourse.pymc.io/t/getting-observations-of-distribution-variables-whilst-sampling/81)
**Author:** [@junpenglao](https://discourse.pymc.io/u/junpenglao)\ **Replies:** 2\ **Last updated:** [June 15, 2017, 2:40pm UTC](https://discourse.pymc.io/t/getting-observations-of-distribution-variables-whilst-sampling/81 "2017-06-15T14:40:38Z")
Github issue #2020 by 23pointsNorth: Hi, I have a model with a set of latent variables, which I am trying to evaluate. So far I have worked in a custom MCMC world and want to switch to HMC/NUTS. However, I was unabl… --- ## [Possible extensions of total\_size](https://discourse.pymc.io/t/possible-extensions-of-total-size/36)
**Author:** [@junpenglao](https://discourse.pymc.io/u/junpenglao)\ **Replies:** 2\ **Last updated:** [June 13, 2017, 2:19pm UTC](https://discourse.pymc.io/t/possible-extensions-of-total-size/36 "2017-06-13T14:19:56Z")
Github issue #2125 by pwl: I have some trouble understanding the total\_size parameter and there is no mention of it in the docs. From the examples I have seen so far the parameter should be set to the total size of …