Offline example | Pareto-efficient choices | Robustness | Real providers
Package: AutoGenAI 0.1.0
AutoGenAI: Adaptive Optimization of Prompts, Models and Generation Strategies
Provides provider-agnostic tools for jointly comparing and optimizing prompts, language-model providers, and generation strategies for generative artificial intelligence workflows. Candidate configurations can be evaluated using user-supplied scoring functions, cost and latency measurements, robustness perturbations, Pareto-front screening, budget and latency constraints, prompt evolution, adaptive routing, self-consistency, and text-output ensembles. The core workflow is designed to run offline with deterministic mock providers, while external model application programming interfaces can be connected through user-defined provider functions. Evolutionary search concepts are described by Goldberg (1989, ISBN:0201157675), and multi-objective optimization concepts are related to Deb, Pratap, Agarwal and Meyarivan (2002) <doi:10.1109/4235.996017>.
Authors:
AutoGenAI_0.1.0.tar.gz
AutoGenAI_0.1.0.zip(r-4.7-any)AutoGenAI_0.1.0.zip(r-4.6-any)AutoGenAI_0.1.0.zip(r-4.5-any)
AutoGenAI_0.1.0.tgz(r-4.6-any)AutoGenAI_0.1.0.tgz(r-4.5-any)
AutoGenAI_0.1.0.tar.gz(r-4.7-any)AutoGenAI_0.1.0.tar.gz(r-4.6-any)
AutoGenAI_0.1.0.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
AutoGenAI/json (API)
| # Install 'AutoGenAI' in R: |
| install.packages('AutoGenAI', repos = c('https://leilamarvian.r-universe.dev', 'https://cloud.r-project.org')) |
This package does not link to any Github/Gitlab/R-forge repository. No issue tracker or development information is available.
Last updated from:081d6b3902. Checks:9 OK. Indexed: yes.
| Target | Result | Time | Files | Syslog |
|---|---|---|---|---|
| linux-devel | OK | 105 | ||
| source / vignettes | OK | 153 | ||
| linux-release | OK | 94 | ||
| macos-release | OK | 81 | ||
| macos-oldrel | OK | 102 | ||
| windows-devel | OK | 78 | ||
| windows-release | OK | 65 | ||
| windows-oldrel | OK | 67 | ||
| wasm-release | OK | 103 |
Exports:abstain_decisionai_ensembleai_providerai_taskautogenai_examplebenchmark_aibudget_optimizeconsensus_votecrossover_promptdefault_scorerevaluate_configevolve_promptlatency_optimizemock_providermodel_generalizationmutate_promptoptimize_aioptimize_generationpareto_aiperturb_textprompt_candidatesprompt_portabilityrobustness_scoreroute_modelrouter_regretselect_configurationself_consistencystress_testtrain_routerupdate_router
Dependencies:jsonlite
Readme and manuals
Help Manual
| Help page | Topics |
|---|---|
| Adaptive Optimization of Prompts, Models and Generation Strategies | AutoGenAI-package AutoGenAI |
| Combine repeated or multi-provider outputs | ai_ensemble consensus_vote self_consistency |
| Create generative AI providers | ai_provider mock_provider print.autogen_provider |
| Define an AutoGenAI task and scorer | ai_task default_scorer print.autogen_task |
| Create an offline AutoGenAI example | autogenai_example |
| Methods for AutoGenAI optimization objects | plot.autogenai predict.autogenai print.autogenai summary.autogenai |
| Benchmark generative AI configurations | benchmark_ai evaluate_config optimize_generation |
| Optimize and select generative AI configurations | abstain_decision budget_optimize latency_optimize optimize_ai pareto_ai select_configuration |
| Prompt search and cross-provider generalization | crossover_prompt evolve_prompt model_generalization mutate_prompt prompt_candidates prompt_portability |
| Stress-test generative AI inputs | perturb_text robustness_score stress_test |
| Train and evaluate an adaptive provider router | router_regret route_model train_router update_router |