diff --git a/content/docs/user-guide/experiment-management/experiments-overview.md b/content/docs/user-guide/experiment-management/experiments-overview.md index 25cefcd2537..584504769cd 100644 --- a/content/docs/user-guide/experiment-management/experiments-overview.md +++ b/content/docs/user-guide/experiment-management/experiments-overview.md @@ -1,26 +1,30 @@ ## DVC Experiments Overview -`dvc exp` commands let you automatically track a variation to an established -[data pipeline](/doc/command-reference/dag) baseline. You can create multiple -isolated experiments this way, as well as review, compare, and restore them -later, or roll back to the baseline. The basic workflow goes like this: +`dvc exp` commands let you automatically track a variation to a committed +project version (baseline). You can create independent groups of experiments +this way, as well as review, compare, and restore them later. The basic workflow +goes like this: -- Modify stage parameters or other dependencies (e.g. input data, - source code) of committed stages. +- Modify parameters or other dependencies (input data, source code, + stage definitions, etc.) of committed stages. - [Run experiments] with `dvc exp run` (instead of `repro`) to execute the pipeline. The results are reflected in your workspace, and tracked automatically. -- Use `dvc metrics` to identify the best experiment(s). -- Visualize, compare experiments with `dvc exp show` or `dvc exp diff`. Repeat - 🔄 -- Use `dvc exp apply` to roll back to the best one. -- Make the selected experiment persistent by committing its results to Git. This - cleans the slate so you can repeat the process. +- Use [metrics](/doc/command-reference/metrics) to identify the best + experiment(s). +- Visualize and compare experiments with `dvc exp show` or `dvc exp diff`. + Repeat 🔄 +- Make certain experiments [persistent](#persistent-experiments) by committing + their results to Git. This cleans the slate so you can repeat the process + later. [run experiments]: /doc/user-guide/experiment-management/running-experiments +[persistent]: /doc/user-guide/experiment-management/persisting-experiments ## Persistent Experiments +📖 See [full guide][persistent]. + When your experiments are good enough to save or share, you may want to store them persistently as Git commits in your repository. diff --git a/content/docs/user-guide/experiment-management/index.md b/content/docs/user-guide/experiment-management/index.md index a9e198b9586..6659575c7bf 100644 --- a/content/docs/user-guide/experiment-management/index.md +++ b/content/docs/user-guide/experiment-management/index.md @@ -1,8 +1,8 @@ # Experiment Management -Data science and ML are iterative processes that require a large number of -attempts to reach a certain level of a metric. Experimentation is part of the -development of data features, hyperspace exploration, deep learning +Data science and machine learning are iterative processes that require a large +number of attempts to reach a certain level of a metric. Experimentation is part +of the development of data features, hyperspace exploration, deep learning optimization, etc. Some of DVC's base features already help you codify and analyze experiments. @@ -12,35 +12,30 @@ the other end, [metrics](/doc/command-reference/metrics) (and [plots](/doc/command-reference/plots)) let you define, visualize, and compare quantitative measures of your results. -## DVC Experiments +## Experimentation methods in DVC _New in DVC 2.0_ -DVC experiment management features are designed to support these main -approaches: - -1. [Run] and capture [experiments] that derive from your latest project version - without polluting your Git history. DVC tracks them for you, letting you list - and compare them. The best ones can be made persistent, and the rest left as - history or cleared. -1. [Queue] and process series of experiments based on a parameter search or - other modifications to your baseline. -1. Generate [checkpoints] during your code execution to analyze the internal - progress of deep experiments. DVC captures them at runtime, and can manage - them in batches. -1. Make experiments [persistent] by committing them to your - repository history. - -[run]: /doc/user-guide/experiment-management/running-experiments +DVC experiment management features build on top of base DVC features to form a +comprehensive framework to organize, execute, manage, and share ML experiments. +They support support these main approaches: + +- Compare params and metrics of existing project versions (for example different + Git branches) against each other or against new results in your workspace + (without committing them). + +- [Run and capture] multiple experiments (derived from any project version as + baseline) without polluting your Git history. DVC tracks them for you, letting + you compare and share them. 📖 More info in the [Experiments + Overview][experiments]. + +- Generate [checkpoints] during your code execution to analyze the internal + progress of deep experiments. DVC captures [live metrics](/doc/dvclive) at + runtime, and lets you manage them in batches. + +[run and capture]: /doc/user-guide/experiment-management/running-experiments [experiments]: /doc/user-guide/experiment-management/experiments-overview -[queue]: - /doc/user-guide/experiment-management/running-experiments#the-experiments-queue [checkpoints]: /doc/user-guide/experiment-management/checkpoints -[persistent]: - /doc/user-guide/experiment-management/experiments-overview#persistent-experiments - -📖 More information in the -[full guide](/doc/user-guide/experiment-management/experiments-overview). > 👨‍💻 See [Get Started: Experiments](/doc/start/experiments) for a hands-on > introduction to DVC experiments. @@ -55,10 +50,12 @@ main alternatives: other. Helpful if the Git [revisions](https://git-scm.com/docs/revisions) can be easily visualized, for example with tools [like GitHub](https://docs.github.com/en/github/visualizing-repository-data-with-graphs/viewing-a-repositorys-network). + - **Directories** - the project's "space dimension" can be structured with directories (folders) to organize experiments. Useful when you want to see all your experiments at the same time (without switching versions) by just exploring the file system. + - **Hybrid** - combining an intuitive directory structure with a good repo branching strategy tends to be the best option for complex projects. Completely independent experiments live in separate directories, while their