prism

Plugin prism from marketplace prismplp in prismplp/prism. Commands read from the repo's manifest — 1 month ago.

Skills (6)

• install-linux

  Build PRISM and T-PRISM from source and set up their environment on Linux (Ubuntu 22.04, 24.04, 26.04). Covers the apt packages, the C and Prolog build with the USE_NPY, USE_H5, USE_PB and PROCTYPE=mp options, PATH, regression tests, the T-PRISM Python environment (venv, PyTorch, protobuf), the runt

• install-mingw

  Build and verify PRISM and bp4prism on Windows with native MinGW-w64, including optional 32bit and 64bit MS-MPI builds, without tools/mingw helper scripts.

• prism-programming

  How to write, run, debug and learn parameters for PRISM programs (probabilistic Prolog on B-Prolog, .psm files, prismplp/prism), with the differences from ProbLog and plain Prolog spelled out, plus two recipes - turning a deterministic Prolog rule base into a PRISM model whose default reproduces the

• programming-prism-ds

  Data science with PRISM from Python (pyprism). Turn tabular data (pandas DataFrames, CSV, scikit-learn datasets) into discretized PRISM facts with pyprism.dataset. Write generative models for it (independent features, naive Bayes classifiers, latent-class clustering). Learn parameters by EM, MAP or

• programming-pyprism

  Install PyPRISM (the pyprism package, the Python interface of PRISM, prismplp/pyprism) so that it matches the Python environment (Google Colab, venv or conda on Linux, Jupyter), and use its PrismEngine to run Prolog and PRISM programs and queries from Python (set_db, query with out/findall/find_n, p

• tprism-programming

  How to write, build and train T-PRISM (Tensorized PRISM, prismplp/prism) programs, where Prolog atoms denote tensors combined by einsum and trained with PyTorch. Covers .psm programs (tensor/2, tensor_atom/2, operator/1, subgoal/2, get/2, placeholders, embeddings, save_expl_graph), `tprism train/tes

Similar plugins