Owner

pymc-labs

13 indexed repositories · View on GitHub

pymc-marketing
Bayesian marketing toolbox in PyMC. Media Mix (MMM), customer lifetime value (CLV), buy-till-you-die (BTYD) models and more.

Python · 1263 stars
CausalPy
A Python package for causal inference in quasi-experimental settings

Python · 1190 stars
decision-lab
Run tested, autonomous agent workflows on your data for meaningful decision-making

Python · 194 stars
pathmc
Structural causal models with Bayesian estimation and interventional simulation via a concise DSL.

Python · 128 stars
decision-hub

Python · 100 stars
pymc-modeling
LLM agent skill files for PyMC and PyTensor

Python · 85 stars
alchemize
LLM-based, self-correcting transpiler. Supports JAX, PyTorch, Rust, PyMC, Stan.

Rust · 52 stars
pymc-forecast
Port of numpyro_forecast

Python · 31 stars
daimon
The open source data science agent for Discord and Slack

Python · 15 stars
pymc-labs.github.io
PyMC Labs Website

HTML · 6 stars
nuts-rs-wasm
NUTS sampling in WebAssembly using nuts-rs, with a PyMC/Numba integration

JavaScript · 1 star
labs-sphinx-theme
Sphinx theme for OSS projectw within PyMC Labs

HTML · 0 stars
pymc-generator
A structural causal model (SCM) generator: it draws random Pearlian DAGs, turns each into one PyTensor / PyMC graph, and emits both the o...

Python · 0 stars

Beginner friendly Everything

Open All

Make MuEffect variables available to causal adjustment-set identification Open

Difficulty 5/5 Over a week Newbie friendliness 35/100

pymc-labs/pymc-marketing#3034 · Opened Sep 19, 2026
Re-enable LFO-CV notebook after an arviz-stats release containing lfo-cv Open

Difficulty 3/5 1-2 days Newbie friendliness 72/100

pymc-labs/pymc-marketing#3031 · Opened Sep 18, 2026
Docs: double-slash canonical URLs and /latest-only pages canonicalising to missing /stable URLs Open

Difficulty 3/5 1-2 days Newbie friendliness 58/100

pymc-labs/pymc-marketing#3028 · Opened Sep 16, 2026
docs-build: unbounded notebook NUTS runtime + cancel-in-progress kills every main build Open

documentation

Difficulty 4/5 3-5 days Newbie friendliness 48/100

pymc-labs/pymc-generator#9 · Opened Sep 16, 2026
SEO follow-up to #2930: RTD redirects, latest-only canonicals, link hygiene, GSC property Open

Difficulty 5/5 Over a week Newbie friendliness 35/100

pymc-labs/pymc-marketing#3008 · Opened Sep 12, 2026
Conditioned model abstraction: one owner of posterior binding for optimizer, counterfactual, incrementality and sensitivity Open

Difficulty 5/5 Over a week Newbie friendliness 32/100

pymc-labs/pymc-marketing#3007 · 2 comments · Opened Sep 10, 2026
Following the BudgetOptimizerWrapper deprecation loses sample_response_distribution and the budget plot namespaces Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

pymc-labs/pymc-marketing#3004 · Opened Sep 10, 2026
Single to_internal/to_business seam for scaling, with a parametrised contract test Open

Difficulty 5/5 Over a week Newbie friendliness 35/100

pymc-labs/pathmc#472 · Opened Sep 10, 2026
Budget optimizer: support discounted cash flow / NPV as an objective Open

enhancement MMM optimizer

Difficulty 4/5 3-5 days Newbie friendliness 48/100

pymc-labs/pymc-marketing#2990 · Opened Sep 9, 2026
Documentation Re-Design Open

docs enhancement priority: low

Difficulty 5/5 Over a week Newbie friendliness 20/100

pymc-labs/pymc-marketing#2983 · 1 comment · Opened Sep 9, 2026
PIE model support for terms Open

enhancement

Difficulty 5/5 Over a week Newbie friendliness 35/100

pymc-labs/pymc-marketing#2982 · 2 comments · Opened Sep 9, 2026
serialization support for terms Open

API enhancement priority: medium terms

Difficulty 3/5 1-2 days Newbie friendliness 45/100

pymc-labs/pymc-marketing#2980 · 1 comment · Opened Sep 9, 2026
Enable GitHub Merge Queue on main Open

Difficulty 3/5 1-2 days Newbie friendliness 68/100

pymc-labs/CausalPy#1182 · Opened Sep 8, 2026
LogLinkSpec.validate_target rejects an all-negative target that the model can fit once scaled Open

enhancement MMM

Difficulty 5/5 Over a week Newbie friendliness 35/100

pymc-labs/pymc-marketing#2974 · 2 comments · Opened Sep 8, 2026
DataDerivedScaling reduces with a signed max/mean while its docstring says max-absolute, and a zero scale silently zeroes the target Open

bug docs MMM priority: medium

Difficulty 4/5 3-5 days Newbie friendliness 48/100

pymc-labs/pymc-marketing#2973 · 3 comments · Opened Sep 8, 2026
The Scaled serialization shim duplicates pymc-extras 0.14, but dropping it changes the on-disk format Open

API customer choice dependencies maintenance MMM priority: medium

Difficulty 5/5 Over a week Newbie friendliness 25/100

pymc-labs/pymc-marketing#2970 · 1 comment · Opened Sep 8, 2026
log_model_graph leaves the DOT source in the working directory when graphviz is not installed Open

bug priority: medium

Difficulty 2/5 1-3 hours Newbie friendliness 78/100

pymc-labs/pymc-marketing#2968 · 1 comment · Opened Sep 8, 2026
docs: BTYD logp is not a density over recency (missing t_x^(x-1)/Gamma(x) factor) Open

CLV docs priority: low

Difficulty 2/5 1-3 hours Newbie friendliness 78/100

pymc-labs/pymc-marketing#2966 · 1 comment · Opened Sep 8, 2026
BG/NBD use Terms Open

CLV enhancement priority: medium terms

pymc-labs/pymc-marketing#2964 · 1 comment · 1 assignee · Opened Sep 8, 2026
Experimental MMM class Open

enhancement MMM terms

pymc-labs/pymc-marketing#2963 · 1 comment · 1 assignee · Opened Sep 8, 2026
Port CLV models to Terms Open

CLV maintenance terms

Difficulty 4/5 3-5 days Newbie friendliness 35/100

pymc-labs/pymc-marketing#2962 · 1 comment · Opened Sep 8, 2026
Port the Bass model to Terms Open

Bass model customer choice maintenance terms

pymc-labs/pymc-marketing#2961 · 1 assignee · Opened Sep 8, 2026
Three PurgePreview rows are counted but never rendered, so the erasure preview undercounts Open

bug

Difficulty 3/5 1-2 days Newbie friendliness 72/100

pymc-labs/daimon#159 · Opened Sep 8, 2026
Default Gamma lift-test likelihood can create a near-zero local mode over model-estimated lift Open

bug enhancement lift tests MMM priority: medium

Difficulty 5/5 Over a week Newbie friendliness 45/100

pymc-labs/pymc-marketing#2958 · 1 comment · Opened Sep 8, 2026
Stale observed target: fit() after sample_prior_predictive() trains on the placeholder y Open

bug MMM priority: high tests

Difficulty 3/5 1-2 days Newbie friendliness 75/100

pymc-labs/pymc-marketing#2956 · 4 comments · 1 reaction · Opened Sep 8, 2026
Report host: serve a report as a branded dashboard with a chat sidebar grounded in the run's artifacts Open

enhancement

Difficulty 5/5 Over a week Newbie friendliness 30/100

pymc-labs/daimon#151 · 2 comments · Opened Sep 7, 2026
Investigate direct Rust-to-Numba WASM calls with jointly managed memory Open

Difficulty 5/5 Over a week Newbie friendliness 35/100

pymc-labs/nuts-rs-wasm#6 · 6 comments · Opened Sep 6, 2026
Claude Code as a first-class entry point: npx plugin that links a platform identity and mints per-agent MCP servers Open

Difficulty 5/5 Over a week Newbie friendliness 30/100

pymc-labs/daimon#149 · Opened Sep 5, 2026
MMM: carry DataVarMuEffect data variables through build, fit, predict, CV, and save/load Open

API enhancement MMM priority: medium tests

Difficulty 5/5 Over a week Newbie friendliness 35/100

pymc-labs/pymc-marketing#2945 · 4 comments · Opened Sep 3, 2026
Optional: split media contribution into current (realized) vs future value via a spend-period x realization-period carryover matrix Open

enhancement MMM priority: high

Difficulty 5/5 Over a week Newbie friendliness 35/100

pymc-labs/pymc-marketing#2941 · 16 comments · Opened Sep 2, 2026
OpenCode + NVIDIA provider rejects requests when agent declares skills: frontmatter (Unsupported parameter(s): 'skills') Open

Difficulty 4/5 3-5 days Newbie friendliness 42/100

pymc-labs/decision-lab#95 · Opened Sep 1, 2026
Funnel/mu-effect models do not survive save/load round-trips Open

API bug MMM priority: medium

Difficulty 4/5 3-5 days Newbie friendliness 57/100

pymc-labs/pymc-marketing#2938 · 5 comments · Opened Aug 31, 2026
Budget optimizer: public API for objective/gradient evaluation and decision-vector packing Open

API enhancement MMM priority: medium

Difficulty 4/5 3-5 days Newbie friendliness 54/100

pymc-labs/pymc-marketing#2937 · 1 comment · Opened Aug 31, 2026
Stale copy: Slack file-posting refusal and tool docstring still claim delivery is impossible post-#101 Open

Difficulty 2/5 1-3 hours Newbie friendliness 86/100

pymc-labs/daimon#142 · Opened Aug 30, 2026
Headless/routine turns never deliver session output files — the sweep is wired only into the interactive Slack path Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

pymc-labs/daimon#141 · Opened Aug 30, 2026
Slack `send_message` refuses all file_handles/attachments unconditionally, though file upload works post-#101 Open

Difficulty 4/5 3-5 days Newbie friendliness 68/100

pymc-labs/daimon#140 · Opened Aug 30, 2026
Discord: post-turn session output files are never delivered — the core sweep has no Discord consumer Open

Difficulty 4/5 3-5 days Newbie friendliness 68/100

pymc-labs/daimon#132 · Opened Aug 26, 2026
Decide whether daimon ships an eval harness, and what it grades Open

enhancement question

Difficulty 5/5 Over a week Newbie friendliness 35/100

pymc-labs/daimon#121 · 1 comment · Opened Aug 25, 2026
BUG: BetaGeoBetaBinom.logp is silently wrong for a length-1 value (collapses BetaGeoBetaBinomModel posterior draws) Open

bug CLV priority: high

Difficulty 3/5 1-2 days Newbie friendliness 78/100

pymc-labs/pymc-marketing#2919 · 2 comments · Opened Aug 24, 2026
test_concurrent_consume_of_one_token_succeeds_exactly_once fails locally, passes in CI Open

question

Difficulty 4/5 3-5 days Newbie friendliness 64/100

pymc-labs/daimon#105 · 1 comment · Opened Aug 24, 2026
spike_and_slab defaults lack an expected-model-size control Open

Difficulty 4/5 3-5 days Newbie friendliness 58/100

pymc-labs/CausalPy#1177 · Opened Aug 23, 2026
variable selection API differs between InstrumentalVariableRegression and StateSpaceTimeSeries Open

Difficulty 4/5 3-5 days Newbie friendliness 52/100

pymc-labs/CausalPy#1176 · Opened Aug 23, 2026
Slack cancel-refusal ephemeral is missing thread_ts and lands at channel root Open

bug good first issue

Difficulty 3/5 1-2 days Newbie friendliness 68/100

pymc-labs/daimon#95 · 2 comments · Opened Aug 22, 2026
Discord routines panel cannot create or delete routines — Slack's can Open

enhancement good first issue

Difficulty 4/5 3-5 days Newbie friendliness 58/100

pymc-labs/daimon#94 · 1 comment · Opened Aug 21, 2026
Adjustment models silently drop parent priors set via set_priors() and transform-parameter priors Open

Difficulty 4/5 3-5 days Newbie friendliness 55/100

pymc-labs/pathmc#465 · Opened Aug 21, 2026
Add a `gamma` family and user-selectable link functions; validate `families` instead of silently falling back to Gaussian Open

pymc-labs/pathmc#464 · 1 assignee · Opened Aug 21, 2026
`adjustment_model()` raises on any parent with a custom `beta_{outcome}` prior; the error's suggested `priors=` workaround is unreachable Open

bug

pymc-labs/pathmc#463 · 1 assignee · Opened Aug 21, 2026
Auto-load root CLAUDE.md from bound repo into agent context Open

enhancement

Difficulty 4/5 3-5 days Newbie friendliness 55/100

pymc-labs/daimon#91 · 1 comment · Opened Aug 20, 2026
Add Daimon admin delegation (decouple agent config from platform workspace admin) Open

enhancement

Difficulty 5/5 Over a week Newbie friendliness 35/100

pymc-labs/daimon#90 · 1 comment · Opened Aug 20, 2026
devops: adopt uv as default dev environment; switch CI to uv; keep conda alternative Open

devops enhancement

Difficulty 5/5 Over a week Newbie friendliness 45/100

pymc-labs/CausalPy#1174 · 2 comments · Opened Aug 19, 2026

Showing the newest 100

This page lists what was indexed most recently. Advanced filter has the whole inventory, narrowed by language, difficulty and how long a task takes.

Open advanced filter

pymc-labs

Showing the newest 100

Get new issues in your inbox