Manage your iteration

Once you start logging assets, you can manage them seamlessly through the API or the app to keep your data science projects organized.

Update iteration status

Vectice API

Once you start an iteration in the API, its status automatically changes to "In progress." Once you are complete, you can update your iteration status by marking it complete.

Once an iteration is complete, you cannot reorganize the assets in the App. To reorganize, change the iteration status back to In Progress in the app.

iteration.complete()

Full example:

import vectice
from vectice import Model

connect = vectice.connect(api_token="your-api-key") #Paste your API key

# Connect to your project phase using your phase ID
phase = connect.phase("PHA-XXX") #You can fetch the relevant phase ID from your chosen Vectice project in the app.

# Create or get an iteration
iteration = phase.create_or_get_current_iteration()

# Define your model for logging to Vectice
model = Model(metrics, properties, attachments, predictor)

# Log your model to Vectice under the 'Model Used' section.
iteration.log(my_model, section="Model Used")

# Iteration is completed
iteration.complete()

Vectice Web App

To update an iteration in the Vectice Web App

1. Go to your phase iteration

2. Select the dropdown under Iteration status

3. Choose your updated status

Create sections

Sections in Vectice function like headers, helping you identify and organize assets. This improves how you access and manage datasets, models, notes, and graphs within your projects.

Vectice API

To define a section to organize your assets using the Vectice API...

# import and connect to Vectice
import vectice
from vectice import Model

connect = vectice.connect(api_token="your-api-key") #Paste your API key

# Connect to your project phase using your phase ID
phase = connect.phase("PHA-XXX") #You can fetch the relevant phase ID from your chosen Vectice project in the app.

# Create or get an iteration
iteration = phase.create_or_get_current_iteration()

# Define your model for logging to Vectice
model = Model(metrics, properties, attachments, predictor)

# Define a section by using the 'section' param and assigning a name
# to log your model to Vectice under the 'Model Used' section.
iteration.log(my_model, section="Model Used")

Vectice Web App

To define a section to organize your assets inside the Vectice Web App...

1. Go to your selected Project and Phase

2. Select the Iteration that has the assets you want to organize

3. Click Create section

4. Once the section is created, drag and drop your assets to the correct section

Delete Sections

You can delete sections in the Web App by selecting the menu button next to the section name and selecting Delete Section. If this section has assets, they will be moved to the default section for unassigned assets.

Move assets across sections in App

You can customize sections to fit your unique needs (e.g., organize assets, outline your workflow, etc.).

Drag and drop assets between sections to rearrange them as you like.

Delete assets in a section

Vectice API

You can delete assets within a specified section of the iteration using the API:

iteration.delete_assets_in_section(section="Collect Initial Data")

Vectice Web App

You can delete assets within a specified section of the iteration in the Web App by

1. Go to your phase iteration and identify the section

2. Select one or more of the assets you want to delete

3. Click the Delete assets

Create Notes

Add notes directly to your iteration for more context.

Vectice API

To add a note via Vectice API, use the following log() method and insert your note.

iteration.log("This is my note.")

Full example:

from vectice import Model
my_model = Model(name="my_model")
iteration.log(my_model)
iteration.log("Testing this new linear regression model.")

Vectice Web App

Inside the Web App, to add a note to your iteration:

1. Go to your phase iteration

2. Select Create note

Star your favorite iterations

Star the iterations you want to highlight to make them stand out, making it easier for stakeholders to identify the top iterations.