Documentation improvements #133

capjamesg · 2023-10-27T10:48:01Z

Description

This PR includes a restructure of the Roboflow Inference documentation.

Type of change

This change only affects documentation and does not modify Inference code.

How has this change been tested, please provide a testcase or example of how you tested the change?

This change was tested by opening http://localhost:8000 and checking that each page was formatted correctly.

Any specific deployment considerations

The documentation will be deployed when this PR is merge to main.

Docs

See above.

paulguerrie · 2023-10-27T19:44:52Z

docs/foundation/doctr.md

+
+To use DocTR with Inference, you will need a Roboflow API key. If you don't already have a Roboflow account, [sign up for a free Roboflow account](https://app.roboflow.com). Then, retrieve your API key from the Roboflow dashboard. Run the following command to set your API key in your coding environment:
+
+```


At this point, have we talked about how to start the roboflow inference server docker container? If now, we should mention it here and link to that part of the docs.

paulguerrie · 2023-10-31T17:47:53Z

docs/quickstart/what_is_inference.md

@@ -0,0 +1,21 @@
+Roboflow Inference enables you to deploy computer vision models faster than ever.
+
+Here is an example of a model running on a video using Inference (See the code):


Is "See the code" supposed to be a link?

paulguerrie · 2023-10-31T17:59:08Z

docs/quickstart/explore_models.md

+
+Click the "Deploy" link in the sidebar to find the information you will need to use your model with Inference:
+
+![Model list](https://media.roboflow.com/docs-sidebar-list.png)


This image appears very large. Any way to make it smaller?

paulguerrie · 2023-10-31T18:03:47Z

docs/quickstart/run_model_on_image.md

+from roboflow import Roboflow
+rf = Roboflow(api_key="API_KEY")
+project = rf.workspace().project("MODEL_ENDPOINT")
+model = project.version(VERSION, local="http://localhost:9001").model


We use localhost here but we haven't taught the user how to start a server yet. Can we add that snippet? Or better yet, can we just show how to run inference on an image the most simple way?

Suggested change

model = project.version(VERSION, local="http://localhost:9001").model

from inference.models.utils import get_roboflow_model

model = get_roboflow_model(...)

result = model.infer(image) #image can be a url string, numpy array, pil image, etc.

paulguerrie · 2023-10-31T18:07:02Z

docs/quickstart/run_model_on_video.md

+
+Create a new Python file and add the following code:
+
+```python


I think this is a convoluted way to run inference on a video now that we have the stream interface? Can this guide look more like the Webcam stream guide and the rtsp stream guide?

Do you have a code snippet? I don't have any examples to hand.

You can just pass the path of a video file or the url of a hosted video file to the source parameter.

paulguerrie · 2023-10-31T18:19:24Z

docs/quickstart/run_model_on_rtsp.md

@@ -0,0 +1,63 @@
+You can run computer vision models on RTSP stream frames with Inference.


This guide is nearly identical to the webcam guide. Should we combine them into a single guide?

paulguerrie · 2023-10-31T18:35:05Z

docs/foundation/about.md

+
+Foundation models are being built for a range of vision tasks, from image segmentation to classification to zero-shot object detection.
+
+Autodistill supports the following foundation models:


Typo? Autostill --> Inference

paulguerrie · 2023-10-31T18:35:28Z

docs/foundation/about.md

+
+Autodistill supports the following foundation models:
+
+- LC2S-Net: Detect the direction in which someone is looking.


I think we call this Gaze.

docs improvements

7741586

capjamesg added the documentation Improvements or additions to documentation label Oct 27, 2023

capjamesg self-assigned this Oct 27, 2023

paulguerrie reviewed Oct 27, 2023

View reviewed changes

add cta to file a feature request

86e636b

paulguerrie reviewed Oct 31, 2023

View reviewed changes

paulguerrie and others added 4 commits October 31, 2023 18:40

A few minor updates to index

79489a7

respond to feedback

6ec2e3a

update table of contents

c140b8a

consolidate video, webcam, rtsp pages

8a27b2f

paulguerrie approved these changes Oct 31, 2023

View reviewed changes

capjamesg merged commit b75e0ae into main Oct 31, 2023
2 checks passed

capjamesg deleted the docs-improvements branch October 31, 2023 20:20

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Documentation improvements #133

Documentation improvements #133

capjamesg commented Oct 27, 2023

paulguerrie Oct 27, 2023

paulguerrie Oct 31, 2023

paulguerrie Oct 31, 2023

paulguerrie Oct 31, 2023

paulguerrie Oct 31, 2023

capjamesg Oct 31, 2023

paulguerrie Oct 31, 2023

paulguerrie Oct 31, 2023

paulguerrie Oct 31, 2023

paulguerrie Oct 31, 2023


		To use DocTR with Inference, you will need a Roboflow API key. If you don't already have a Roboflow account, [sign up for a free Roboflow account](https://app.roboflow.com). Then, retrieve your API key from the Roboflow dashboard. Run the following command to set your API key in your coding environment:

		```

		@@ -0,0 +1,21 @@
		Roboflow Inference enables you to deploy computer vision models faster than ever.

		Here is an example of a model running on a video using Inference (See the code):


		Click the "Deploy" link in the sidebar to find the information you will need to use your model with Inference:

		![Model list](https://media.roboflow.com/docs-sidebar-list.png)

-model = project.version(VERSION, local="http://localhost:9001").model
+from inference.models.utils import get_roboflow_model
+model = get_roboflow_model(...)
+result = model.infer(image) #image can be a url string, numpy array, pil image, etc.


		Create a new Python file and add the following code:

		```python

		@@ -0,0 +1,63 @@
		You can run computer vision models on RTSP stream frames with Inference.


		Foundation models are being built for a range of vision tasks, from image segmentation to classification to zero-shot object detection.

		Autodistill supports the following foundation models:


		Autodistill supports the following foundation models:

		- LC2S-Net: Detect the direction in which someone is looking.

Documentation improvements #133

Documentation improvements #133

Conversation

capjamesg commented Oct 27, 2023

Description

Type of change

How has this change been tested, please provide a testcase or example of how you tested the change?

Any specific deployment considerations

Docs

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment