picsellia-pipelines-cli 0.7.1

No project description provided

Picsellia Pipelines CLI

The Picsellia Pipelines CLI lets you quickly create, test, dockerize, deploy, and manage custom processing or training pipelines.

How it works

A pipeline is simply:

• a Python function (pipeline.py)

• calling steps (steps.py)

• configured via run_config.toml

• run locally or on Picsellia infrastructure

Workflow at a glance

1. Init → generate project template

2. Customize → implement steps & parameters

3. Test → run locally

4. Smoke Test → validate Docker image

5. Deploy — Publish Your Pipeline to Picsellia → push & register pipeline

1. Installation

Show installation instructions

With uv (recommended)

uv pip install picsellia-pipelines-cli

With Poetry:

poetry add picsellia-pipelines-cli

Check installation:

pxl-pipeline --help

2. Authentication

Objective: Use the same Picsellia user and environment across all commands

pxl-pipeline login

Show detailed explanation

This stores your:

• organization
• environment (PROD / BETA / DEV / CUSTOM)
• API token
• optional custom base URL

Other helpful commands:

pxl-pipeline whoami   # show active context
pxl-pipeline switch   # change organization/environment
pxl-pipeline logout   # clear active context

3. Init — Create a new pipeline

Objective: Generate a ready-to-use pipeline project folder with all required template files

pxl-pipeline init <pipeline_name> --type [training|processing] --template <template_name>

Show detailed explanation

Examples:

pxl-pipeline init yolov8 --type training --template yolov8
pxl-pipeline init resize-images --type processing --template dataset_version_creation

This generates:

• a single entrypoint: pipeline.py

• config.toml (metadata + execution parameters)

• Dockerfile and .dockerignore

• a consistent folder structure:

my-pipeline/
├── pipeline.py
├── steps.py
├── utils/
│   ├── parameters.py
├── config.toml
├── Dockerfile
├── .dockerignore
├── runs/
│   └── run_config.toml   # template for test/smoke-test/launch
└── pyproject.toml

You're now ready to implement your custom logic.

4. Customize your pipeline — Add steps & parameters

Objective: Adapt the pipeline template to your specific use case

After running init, your pipeline project is generated with a default structure:

• pipeline.py — your entrypoint

• steps.py (a default process step) — implement processing or training steps

• utils/parameters.py — define your pipeline parameters

👉 In most cases, if you chose the right template, you only need to modify the existing process step to implement your use case.

You do not need to redesign the whole pipeline unless your logic is more advanced.

Recommended approach

1. Start by editing the existing process step

   This step already exposes the correct inputs, outputs, and parameters expected by Picsellia.

2. Only add new steps if needed

   If your pipeline requires additional logic (pre-processing, post-processing, custom validation, chaining operations, etc.), you can:

   • modify the existing step
   • replace it entirely
   • or add new steps and compose them in pipeline.py

Working with Steps

Show detailed explanation

Steps are Python functions decorated with @step. You can define them in steps.py and call them freely inside pipeline.py.

Example (steps.py):

from picsellia_pipeline.core import step

@step
def process(dataset_input, dataset_output):
    # your processing logic here
    dataset_output["images"] = [img.upper() for img in dataset_input["images"]]
    return dataset_output

Example pipeline (pipeline.py):

from steps import process
from picsellia_pipeline.core import pipeline, step

@pipeline
def dataset_version_creation_pipeline():
    dataset_collection = load_coco_datasets()
    dataset_collection["output"] = process(
        dataset_collection["input"], dataset_collection["output"]
    )
    upload_full_dataset(dataset_collection["output"], use_id=False)
    return dataset_collection

Adding Parameters

Show detailed explanation

Define pipeline parameters in utils/parameters.py.

All parameters declared here are:

• automatically detected by the CLI

• injected at runtime

• uploaded to Picsellia during deploy or sync

Example:

from picsellia.types.schemas import LogDataType
from picsellia_cv_engine.core.parameters import Parameters

class ProcessingParameters(Parameters):
    def __init__(self, log_data: LogDataType):
        super().__init__(log_data=log_data)
        self.datalake = self.extract_parameter(["datalake"], expected_type=str, default="default")
        self.data_tag = self.extract_parameter(["data_tag"], expected_type=str, default="processed")

Each parameter requires:

• a key → used as the parameter name on the Picsellia platform

• an expected type → str, int, float, etc.

• a default value → mandatory for parameter registration

Once parameters are defined, you can reference them directly in your step logic and override their values in run_config.toml.

5. Test — Run your pipeline locally

Objective: Ensure your Python code works exactly as expected, using real Picsellia objects

pxl-pipeline test <pipeline_name> --run-config-file <path>

Show detailed explanation

⚠️ Important

Even though the pipeline runs locally, all datasets, experiments, and outputs are created and updated on the Picsellia platform.

This command:

• runs the pipeline locally in the virtual env (.venv/)
• loads the configuration from your run_config.toml
• interacts with real Picsellia objects
• uploads results to the platform exactly like a real run
• guarantees your step logic and parameters behave correctly

A template run config is generated automatically at:

<pipeline_name>/runs/run_config.toml

You simply need to fill it with your dataset/model IDs, parameters, or metadata.

Example (dataset version creation processing):

override_outputs = true

[job]
type = "DATASET_VERSION_CREATION"

[input.dataset_version]
id = ""

[output.dataset_version]
name = "test_my_pipeline"

[parameters]
datalake = "default"
data_tag = "processed"

Once the file is filled:

pxl-pipeline test my-pipeline \
  --run-config-file my-pipeline/runs/run_config.toml

6. Smoke Test — Validate the Docker runtime

Objective: Ensure your Dockerfile, dependencies, imports, paths, and runtime fully work before deployment

pxl-pipeline smoke-test <pipeline_name> --run-config-file <path>

Show detailed explanation

This command:

1. Builds the Docker image
2. Runs the pipeline inside the container (not Python locally)
3. Uses the same run_config.toml as the test command
4. Updates real objects/results on Picsellia

It is your final validation step before deployment.

A successful smoke test strongly indicates that the pipeline will run properly on Picsellia’s infrastructure.

7. Deploy — Publish your pipeline to Picsellia

Objective: Build, version, push the Docker image, and register/update the pipeline in your organization

pxl-pipeline deploy <pipeline_name>

Show detailed explanation

This command:

• builds the Docker image
• pushes it to your configured registry
• versions the image
• creates or updates the Picsellia processing or training asset
• ensures the pipeline is ready to be launched from the UI

After deployment, the pipeline becomes usable by your team in the Picsellia interface.

Launch — Run your pipeline on Picsellia’s infrastructure

(Optional)

Objective: Trigger a real Picsellia job (not local), using the same run_config.toml

pxl-pipeline launch <pipeline_name> --run-config-file <path>

Show detailed explanation

Launch behaves like:

• launching a processing job on a dataset
• or launching a training experiment
• without manually creating an experiment or job in the UI

The run_config.toml defines:

• the dataset/model input
• the output dataset or experiment name
• the pipeline parameters

This is equivalent to triggering an actual job from the Picsellia UI.

Sync — Synchronize local parameters with Picsellia

(Optional)

Objective: Update parameters stored on Picsellia to match your local

pxl-pipeline sync <pipeline_name>

Show detailed explanation

For processing pipelines, this syncs:

• default parameter values
• parameter schema / types

Sync is usually unnecessary if you run:

pxl-pipeline deploy

because deploy already updates the parameter definition on the platform.

Training sync is not yet implemented.

Tips

• Use --output-dir during init to generate the pipeline elsewhere
• Virtual environments are created in <pipeline>/.venv
• You can edit config.toml at any time (metadata, entrypoints, dependencies)
• Always run test → smoke-test → deploy for a clean workflow
• A successful smoke test almost guarantees a successful production run

---

Made with ❤️ by the Picsellia team.