diff --git a/content/docs/user-guide/concepts/data-pipelines.md b/content/docs/user-guide/concepts/data-pipelines.md index a0f640a9dd..49a02e53d7 100644 --- a/content/docs/user-guide/concepts/data-pipelines.md +++ b/content/docs/user-guide/concepts/data-pipelines.md @@ -25,3 +25,9 @@ In DVC, pipeline stages and commands, their data I/O, interdependencies, and results (intermediate or final) are specified in `dvc.yaml`, which can be written manually or built using `dvc run`. This allows DVC to restore one or more pipelines later. (See `dvc repro`) + +## Further Reading + +- [Data Pipelines](/doc/start/data-pipelines.md) in Get Started +- `dvc run` command reference +- `dvc repro` command reference diff --git a/content/docs/user-guide/concepts/dependency.md b/content/docs/user-guide/concepts/dependency.md index c426223ee0..b91722b813 100644 --- a/content/docs/user-guide/concepts/dependency.md +++ b/content/docs/user-guide/concepts/dependency.md @@ -23,3 +23,7 @@ this case DVC runs the stage again when running the pipeline. A dependency can be internal or external. Internal dependencies are found within the workspace and external dependencies are in other locations like remotes. + +## Further Reading + +- [Experiments](/doc/start/experiments.md) diff --git a/content/docs/user-guide/concepts/dvc-cache.md b/content/docs/user-guide/concepts/dvc-cache.md index 6ece5e2300..ff2a49ffa8 100644 --- a/content/docs/user-guide/concepts/dvc-cache.md +++ b/content/docs/user-guide/concepts/dvc-cache.md @@ -40,3 +40,10 @@ DVC can cache entire directories as a single unit similar to files. A `.dir` file for each tracked directory is created that lists names and hash values of files in the directory. This eliminates the need to create corresponding `*.dvc` files for each file in a directory, which can contain thousands of files. + +## Further Reading + +- [Internal Files](/doc/user-guide/project-structure/internal-files.md) in + User's Guide +- [Shared Development Server](/doc/use-cases/shared-development-server.md) in + Use Cases diff --git a/content/docs/user-guide/concepts/stage.md b/content/docs/user-guide/concepts/stage.md index dbd6d46462..9a1a0a9ecc 100644 --- a/content/docs/user-guide/concepts/stage.md +++ b/content/docs/user-guide/concepts/stage.md @@ -9,12 +9,16 @@ tooltip: >- # Stage A stage is a single unit of execution in a pipeline that takes some -input and produces an output. A stage is defined by its +input and produces some output. A stage is defined by its dependencies, its command and its outputs. -A stage may have multiple dependencies, multiple commands and multiple outputs. -A stage can also have parameter dependencies to represent model -hyperparameters and similar changes during execution. See `dvc run` for defining -a stage. +Stages that depend on other stages can be invalidated by the change in their +dependencies. A stage may have multiple dependencies, multiple commands and +multiple outputs. A stage can also have parameter dependencies to +represent model hyperparameters and similar changes during execution. A stage may also represent an import from outside sources. + +## Further Reading + +- `dvc run` command reference