-
Notifications
You must be signed in to change notification settings - Fork 386
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
start: reader instructions (in index pages) #4000
Changes from 4 commits
9d7bc7e
021299a
63354de
036d55a
d0e90ce
08fa408
3bd8b55
1c508f4
e6bfa08
d3583e4
7756639
7cbfb46
433de85
6ea6d62
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,58 @@ | ||
--- | ||
title: 'Get Started: Experiment Management' | ||
description: | ||
'Get started with experiment management and visualization in DVC, without | ||
bloating Git history.' | ||
--- | ||
|
||
# Get Started: Experiment Management | ||
|
||
Let's look at DVC's features from the perspective of machine learning experiment | ||
management. This includes bootstrapping, running and queueing experiments | ||
(possibly in parallel), as well as visually comparing the results to commit the | ||
best ones to Git. | ||
jorgeorpinel marked this conversation as resolved.
Show resolved
Hide resolved
|
||
|
||
## Following the Get Started | ||
|
||
Although presented in a logical order, each page in this trail is an independent | ||
chapter you can go to directly. There may be references to other chapters or | ||
docs, but you can skip them and still get a good idea of the features in | ||
question, even by only reading page sections. | ||
|
||
For better learning, you can also try each step yourself and get to similar | ||
results. Some of the required steps for this may be inside collapsed sections | ||
you can click on to expand: | ||
|
||
<details> | ||
|
||
### Click for an example! | ||
|
||
You'll find useful notes and tips throughout our docs such as this one: | ||
|
||
<admon type="tip"> | ||
|
||
The steps and results of these chapters are captured in our | ||
[example-dvc-experiments] repo on GitHub. Feel free to clone it and | ||
`git checkout` any of it's [tags] to explore further. | ||
|
||
[example-dvc-experiments]: https://github.com/iterative/example-dvc-experiments | ||
[tags]: https://github.com/iterative/example-dvc-experiments/tags | ||
|
||
</admon> | ||
|
||
Click again to close these details again. | ||
|
||
</details> | ||
|
||
<cards> | ||
|
||
<card href="/doc/start/experiment-management/experiments" heading="Experiments"> | ||
Enable exploration, iteration, and comparison across many trials in ML | ||
projects. | ||
</card> | ||
|
||
<card href="/doc/start/experiment-management/visualization" heading="Visualization"> | ||
Compare experiment results visually. | ||
</card> | ||
|
||
</cards> |
Original file line number | Diff line number | Diff line change |
---|---|---|
|
@@ -11,20 +11,24 @@ running `dvc init` inside a Git project: | |
|
||
<details> | ||
|
||
### ⚙️ Expand to prepare the project. | ||
### ⚙️ Expand to prepare a project. | ||
|
||
We'll be building an NLP project from scratch together. The end result is | ||
published on [GitHub](https://github.com/iterative/example-get-started) -- feel | ||
free to clone the repo. | ||
|
||
Let's start with `git init`: | ||
Imagine we want to building an ML project from scratch. Let's start by creating | ||
a Git repository: | ||
|
||
```dvc | ||
$ mkdir example-get-started | ||
$ cd example-get-started | ||
$ git init | ||
``` | ||
|
||
<admon type="info"> | ||
|
||
This name is actually used in our | ||
jorgeorpinel marked this conversation as resolved.
Show resolved
Hide resolved
|
||
[example-get-started](https://github.com/iterative/example-get-started) repo. | ||
|
||
</admon> | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. I also realized this is only applicable to the Data Mgmt trail now, so I updated the note here (made it more general) |
||
|
||
</details> | ||
|
||
```dvc | ||
|
@@ -45,20 +49,22 @@ $ git commit -m "Initialize DVC" | |
|
||
Now you're ready to DVC! | ||
|
||
DVC's multiple feature sets are best understood from different angles. Pick a | ||
trail below to see an overview of all features from that perspective: | ||
The value of DVC's many feature sets is best understood from different angles. | ||
Pick a trail below to see an overview of all features from that perspective: | ||
|
||
### Data Management | ||
|
||
- [Introduction](/doc/start/data-management) | ||
jorgeorpinel marked this conversation as resolved.
Show resolved
Hide resolved
|
||
|
||
- **[Data and model versioning]** is the base layer of DVC for large files, | ||
datasets, and machine learning models. Use a regular Git workflow, but without | ||
storing large files in the repo (think "Git for data"). Data is stored | ||
separately, which allows for efficient sharing. | ||
|
||
- **[Data and model access]** shows how to use data artifacts from outside of | ||
the project and how to import data artifacts from another DVC project. This | ||
can help to download a specific version of an ML model to a deployment server | ||
or import a model to another project. | ||
- **[Data and model access]** goes over using data artifacts from outside of the | ||
project and importing them from another DVC project. This can help to download | ||
a specific version of an ML model to a deployment server or import a model to | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Updated some of these descriptions |
||
another project. | ||
|
||
- **[Data pipelines]** describe how models and other data artifacts are built, | ||
and provide an efficient way to reproduce them. Think "Makefiles for data and | ||
|
@@ -76,10 +82,12 @@ trail below to see an overview of all features from that perspective: | |
|
||
### Experimentation | ||
|
||
- **[Experiments]** enable exploration, iteration, and comparison across many ML | ||
experiments. Track your experiments with automatic versioning and checkpoint | ||
logging. Compare differences in parameters, metrics, code, and data. Apply, | ||
drop, roll back, resume, or share any experiment. | ||
- [Introduction](/doc/start/experiment-management) | ||
|
||
- **[Experiments]** enable exploration, iteration, and comparison across many | ||
trials in ML projects. Track your experiments with automatic versioning and | ||
checkpoint logging. Compare differences in parameters, metrics, code, and | ||
data. Apply, drop, roll back, resume, or share any experiment. | ||
|
||
- **[Visualization]** compare experiment results visually, track your plots and | ||
generate them with library integrations. | ||
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Should we maybe turn the tags that correspond to certain chapters and sections into Releases in the GH repo? I.e. from some https://github.com/iterative/example-get-started/tags to https://github.com/iterative/example-get-started/releases in this case. May be easier to mention releases than all the tags (not so easy to explore and match to the GS).