Skip to content

Commit

Permalink
fix: fix line numbers because of ruff format (#35)
Browse files Browse the repository at this point in the history
  • Loading branch information
@frederik-encord
frederik-encord authored Dec 17, 2024
1 parent 56597d4 commit cfa530b
Show file tree
Hide file tree
Showing 2 changed files with 23 additions and 23 deletions.
42 changes: 21 additions & 21 deletions docs/editor_agents/examples/index.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -139,13 +139,13 @@ First, we import dependencies and set up the Project:
Make sure to insert your Project's hash here.

<!--codeinclude-->
[agent.py](../../code_examples/gcp/frame_classification.py) lines:1-14
[agent.py](../../code_examples/gcp/frame_classification.py) lines:1-15
<!--/codeinclude-->

Next, we create a data model and a system prompt based on the Project Ontology that will tell Claude how to structure its response:

<!--codeinclude-->
[agent.py](../../code_examples/gcp/frame_classification.py) lines:17-28
[agent.py](../../code_examples/gcp/frame_classification.py) lines:18-29
<!--/codeinclude-->


Expand Down Expand Up @@ -316,14 +316,14 @@ Next, we create a data model and a system prompt based on the Project Ontology t
We also need an Anthropic API client to communicate with Claude:

<!--codeinclude-->
[agent.py](../../code_examples/gcp/frame_classification.py) lines:31-32
[agent.py](../../code_examples/gcp/frame_classification.py) lines:32-33
<!--/codeinclude-->


Finally, we define our editor agent:

<!--codeinclude-->
[agent.py](../../code_examples/gcp/frame_classification.py) lines:35-66
[agent.py](../../code_examples/gcp/frame_classification.py) lines:36-65
<!--/codeinclude-->

The agent:
Expand Down Expand Up @@ -597,15 +597,15 @@ For this, you will need to have your `<project_hash>` ready.

<!--codeinclude-->

[agent.py](../../code_examples/gcp/object_classification.py) lines:1-13
[agent.py](../../code_examples/gcp/object_classification.py) lines:1-14

<!--/codeinclude-->

Now that we have the project, we can extract the generic ontology object as well as that actual ontology objects that we care about.

<!--codeinclude-->

[agent.py](../../code_examples/gcp/object_classification.py) lines:14-18
[agent.py](../../code_examples/gcp/object_classification.py) lines:15-19

<!--/codeinclude-->

Expand All @@ -621,7 +621,7 @@ is only allowed to choose between the object types that are not of the generic o

<!--codeinclude-->

[agent.py](../../code_examples/gcp/object_classification.py) lines:20-29
[agent.py](../../code_examples/gcp/object_classification.py) lines:22-30

<!--/codeinclude-->

Expand Down Expand Up @@ -905,15 +905,15 @@ With the system prompt ready, we can instantiate an api client for Claude.

<!--codeinclude-->

[agent.py](../../code_examples/gcp/object_classification.py) lines:31-33
[agent.py](../../code_examples/gcp/object_classification.py) lines:33-34

<!--/codeinclude-->

Now, let's define the editor agent.

<!--codeinclude-->

[agent.py](../../code_examples/gcp/object_classification.py) lines:36-45
[agent.py](../../code_examples/gcp/object_classification.py) lines:38-46

<!--/codeinclude-->

Expand All @@ -927,7 +927,7 @@ Notice how the `crop` variable has a convenient `b64_encoding` method to produce

<!--codeinclude-->

[agent.py](../../code_examples/gcp/object_classification.py) lines:46-59
[agent.py](../../code_examples/gcp/object_classification.py) lines:47-60

<!--/codeinclude-->

Expand All @@ -938,15 +938,15 @@ If successful, the old generic object can be removed and the newly classified ob

<!--codeinclude-->

[agent.py](../../code_examples/gcp/object_classification.py) lines:61-81
[agent.py](../../code_examples/gcp/object_classification.py) lines:63-80

<!--/codeinclude-->

Finally, we'll save the labels with Encord.

<!--codeinclude-->

[agent.py](../../code_examples/gcp/object_classification.py) lines:83-85
[agent.py](../../code_examples/gcp/object_classification.py) lines:83-84

<!--/codeinclude-->

Expand Down Expand Up @@ -1047,27 +1047,27 @@ Let us go through the code section by section.
First, we import dependencies and setup the FastAPI app with CORS middleware:

<!--codeinclude-->
[main.py](../../code_examples/fastapi/frame_classification.py) lines:1-22
[main.py](../../code_examples/fastapi/frame_classification.py) lines:1-25
<!--/codeinclude-->

The CORS middleware is crucial as it allows the Encord platform to make requests to your API.

Next, we set up the Project and create a data model based on the Ontology:

<!--codeinclude-->
[main.py](../../code_examples/fastapi/frame_classification.py) lines:24-28
[main.py](../../code_examples/fastapi/frame_classification.py) lines:28-30
<!--/codeinclude-->

We create the system prompt that tells Claude how to structure its response:

<!--codeinclude-->
[main.py](../../code_examples/fastapi/frame_classification.py) lines:30-41
[main.py](../../code_examples/fastapi/frame_classification.py) lines:33-45
<!--/codeinclude-->

Finally, we define the endpoint to handle the classification:

<!--codeinclude-->
[main.py](../../code_examples/fastapi/frame_classification.py) lines:44-71
[main.py](../../code_examples/fastapi/frame_classification.py) lines:48-78
<!--/codeinclude-->

The endpoint:
Expand All @@ -1079,7 +1079,7 @@ The endpoint:
5. Parses Claude's response into classification instances
6. Adds the classifications to the label row and saves it

### Testing the Agent**
### Testing the Agent

**STEP 1: Run the FastAPI Server**
With the agent laid down, we can run it and test it.
Expand Down Expand Up @@ -1155,25 +1155,25 @@ Let's walk through the key components.
First, we setup the FastAPI app and CORS middleware:

<!--codeinclude-->
[main.py](../../code_examples/fastapi/object_classification.py) lines:1-22
[main.py](../../code_examples/fastapi/object_classification.py) lines:1-23
<!--/codeinclude-->

Then we setup the client, Project, and extract the generic Ontology object:

<!--codeinclude-->
[main.py](../../code_examples/fastapi/object_classification.py) lines:24-30
[main.py](../../code_examples/fastapi/object_classification.py) lines:26-32
<!--/codeinclude-->

We create the data model and system prompt for Claude:

<!--codeinclude-->
[main.py](../../code_examples/fastapi/object_classification.py) lines:32-43
[main.py](../../code_examples/fastapi/object_classification.py) lines:34-47
<!--/codeinclude-->

Finally, we define our object classification endpoint:

<!--codeinclude-->
[main.py](../../code_examples/fastapi/object_classification.py) lines:46-84
[main.py](../../code_examples/fastapi/object_classification.py) lines:50-97
<!--/codeinclude-->

The endpoint:
Expand Down
4 changes: 2 additions & 2 deletions docs/task_agents/examples/index.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -148,7 +148,7 @@ Suppose you have a _fake_ model like this one, which predicts labels, bounding b

<!--codeinclude-->

[Fake model predictions](../../code_examples/tasks/prelabel_videos.py) lines:28-51
[Fake model predictions](../../code_examples/tasks/prelabel_videos.py) lines:29-52

<!--/codeinclude-->

Expand Down Expand Up @@ -176,7 +176,7 @@ Create a pre-labeling agent using the following code as a template:

<!--codeinclude-->

[prelabel_video.py](../../code_examples/tasks/prelabel_videos.py) lines:10-78
[prelabel_video.py](../../code_examples/tasks/prelabel_videos.py) lines:10-77

<!--/codeinclude-->

Expand Down

0 comments on commit cfa530b

Please sign in to comment.