rtv-workmate 0.1.1

Python client library for Workmate APIs

Workmate Library

A Python client library for interacting with Workmate APIs.

Installation

pip install -e .

Modules

• risk_assessment - Model management functions
• sbcc - SBCC message feedback management functions

Risk Assessment Module

The risk_assessment module provides functions for managing risk assessment models.

Setup

First, configure the API connection:

from rtv_workmate.risk_assessment import set_config

set_config(
    base_url="http://localhost:8000",
    api_key="wm_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
)

# Or use JWT token
set_config(
    base_url="http://localhost:8000",
    jwt_token="your_jwt_token_here"
)

Usage Examples

List all models

from rtv_workmate.risk_assessment import list_models, set_config

set_config(base_url="http://localhost:8000", api_key="your_api_key")
models = list_models()
for model in models:
    print(f"{model['name']} - {model['category']} v{model['version']}")

Get a specific model

from rtv_workmate.risk_assessment import get_model, set_config

set_config(base_url="http://localhost:8000", api_key="your_api_key")
model = get_model({
    'name': 'UG_North_1_C',
    'category': 'classification',
    'version': '2025.0.1'
})
print(model['metrics'])

Create a model

from rtv_workmate.risk_assessment import create_model, set_config

set_config(base_url="http://localhost:8000", api_key="your_api_key")
create_model({
    'name': 'UG_North_1_C',
    'category': 'classification',
    'version': '2025.0.1',
    'description': 'Latest model for North region',
    'metrics': {'accuracy': 0.76, 'precision': 0.72, 'recall': 0.80},
    'other_training_results': {
            'roc_auc': ...,
            'best_cutoff': ...,
            'feature_importances': ...,
            'numeric_features': ...,
            'categorical_features': ...,
            'one_way_pds': make_json_serializable(model_pdps),
            'two_way_pds': make_json_serializable(model_two_way_pdps)
        },
    'model_file_path': '/path/to/model.pkl'
})

Update a model

from rtv_workmate.risk_assessment import update_model, set_config

set_config(base_url="http://localhost:8000", api_key="your_api_key")
updated_model = update_model(
    model_identifier={
        'name': 'UG_North_1_C',
        'category': 'classification',
        'version': '2025.0.1'
    },
    model_data={
        'description': 'Updated description',
        'metrics': {'accuracy': 0.78}
    }
)

Delete a model

from rtv_workmate.risk_assessment import delete_model, set_config

set_config(base_url="http://localhost:8000", api_key="your_api_key")
delete_model({
    'name': 'UG_North_1_C',
    'category': 'classification',
    'version': '2025.0.1'
})

Error Handling

from rtv_workmate.risk_assessment import get_model, APIError, set_config

set_config(base_url="http://localhost:8000", api_key="your_api_key")

try:
    model = get_model({'name': 'test', 'category': 'test', 'version': '1.0'})
except APIError as e:
    print(f"API Error: {e.message} (Status: {e.status_code})")

API Reference

Configuration

• set_config(base_url, api_key=None, jwt_token=None) - Set global API configuration

Model Functions

• list_models() - Get all models
• get_model(model_identifier) - Get a specific model. model_identifier dict requires: name, category, version
• create_model(model_data) - Create a new model. model_data dict requires: name, category, version. Optional: description, metrics, other_training_results, model_file, model_file_path
• update_model(model_identifier, model_data) - Update an existing model. model_identifier dict requires: name, category, version. model_data dict can contain any fields to update
• delete_model(model_identifier) - Delete a model. model_identifier dict requires: name, category, version

Exceptions

• APIError - Exception raised for API errors. Contains message and status_code attributes.

SBCC Module

The sbcc module provides functions for managing SBCC message feedback.

Setup

First, configure the API connection:

from rtv_workmate.sbcc import set_config

set_config(
    base_url="http://localhost:8000",
    api_key="wm_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
)

# Or use JWT token
set_config(
    base_url="http://localhost:8000",
    jwt_token="your_jwt_token_here"
)

Usage Examples

List all message feedbacks

from rtv_workmate.sbcc import list_message_feedbacks, set_config

set_config(base_url="http://localhost:8000", api_key="your_api_key")
feedbacks = list_message_feedbacks()
for feedback in feedbacks:
    print(f"{feedback['practice']} - {feedback['barrier']}")

List with filters

from rtv_workmate.sbcc import list_message_feedbacks, set_config

set_config(base_url="http://localhost:8000", api_key="your_api_key")
# Filter by liked status and practice
feedbacks = list_message_feedbacks(filters={
    'is_liked': True,
    'practice': 'Handwashing'
})

Get a specific feedback

from rtv_workmate.sbcc import get_message_feedback, set_config

set_config(base_url="http://localhost:8000", api_key="your_api_key")
feedback = get_message_feedback(pk=1)
print(feedback['generated_message'])

Create a feedback

from rtv_workmate.sbcc import create_message_feedback, set_config

set_config(base_url="http://localhost:8000", api_key="your_api_key")
feedback = create_message_feedback({
    'form_guid': 'form-123',
    'user_guid': 'user-456',
    'practice': 'Handwashing',
    'barrier': 'Lack of soap',
    'generated_message': 'Remember to wash your hands with soap.',
    'is_liked': True,
    'preferred_message': 'Please wash hands before meals.',
    'test_mode': False  # Set to True to test without persisting
})

Update a feedback (full update)

from rtv_workmate.sbcc import update_message_feedback, set_config

set_config(base_url="http://localhost:8000", api_key="your_api_key")
updated = update_message_feedback(
    pk=1,
    feedback_data={
        'form_guid': 'form-123',
        'user_guid': 'user-456',
        'practice': 'Handwashing',
        'barrier': 'Lack of soap',
        'generated_message': 'Updated message.',
        'is_liked': False,
        'preferred_message': 'Updated preferred message.'
    }
)

Partial update

from rtv_workmate.sbcc import partial_update_message_feedback, set_config

set_config(base_url="http://localhost:8000", api_key="your_api_key")
updated = partial_update_message_feedback(
    pk=1,
    feedback_data={
        'is_liked': True,
        'preferred_message': 'New preferred message'
    }
)

Delete a feedback

from rtv_workmate.sbcc import delete_message_feedback, set_config

set_config(base_url="http://localhost:8000", api_key="your_api_key")
delete_message_feedback(pk=1, test_mode=False)

Test Mode

All create/update/delete operations support test mode, which allows you to test requests without persisting data:

# Test mode - data won't be persisted
create_message_feedback({
    'form_guid': 'form-123',
    'user_guid': 'user-456',
    'practice': 'Handwashing',
    'barrier': 'Lack of soap',
    'generated_message': 'Test message',
    'test_mode': True
})

API Reference

Configuration

• set_config(base_url, api_key=None, jwt_token=None) - Set global API configuration

Message Feedback Functions

• list_message_feedbacks(filters=None) - List all feedbacks. filters dict can contain: form_guid, user_guid, practice, barrier, is_liked
• get_message_feedback(pk) - Get a specific feedback by ID
• create_message_feedback(feedback_data) - Create a new feedback. feedback_data dict requires: form_guid, user_guid, practice, barrier, generated_message. Optional: is_liked, preferred_message, test_mode
• update_message_feedback(pk, feedback_data) - Update all fields of a feedback (PUT). feedback_data dict requires: form_guid, user_guid, practice, barrier, generated_message, is_liked. Optional: preferred_message, test_mode
• partial_update_message_feedback(pk, feedback_data) - Partially update a feedback (PATCH). feedback_data dict can contain any fields to update. Optional: test_mode
• delete_message_feedback(pk, test_mode=False) - Delete a feedback

Exceptions

• APIError - Exception raised for API errors. Contains message and status_code attributes.