datarobot-model-training

Name: datarobot-model-training
Author: datarobot-oss

22stars19forks

How to install datarobot-model-training?

datarobot-oss/datarobot-agent-skills/datarobot-model-training

$npx -y skills add datarobot-oss/datarobot-agent-skills --skill datarobot-model-training

Installs into the current project.

Summary

Comprehensive guidance for training models in DataRobot, including project creation, AutoML configuration, feature engineering, and model selection. Use when training models, creating AutoML projects, or selecting models in DataRobot.

Files · 1

View on GitHub

SKILL.md

1	# DataRobot Model Training Skill
2
3	This skill provides guidance for the complete model training workflow in DataRobot, from project creation through model selection and validation.
4
5	## Quick Start
6
7	Most common use case: Create a project and train models
8
9	1. Upload dataset: `upload_dataset(file_path, dataset_name)` to upload training data
10	2. Create project: `create_project(dataset_id, project_name)` to create new project
11	3. Start training: `start_automl(project_id, mode)` to begin AutoML training
12
13	Example: "Create a new project with sales_data.csv, set 'revenue' as target, and start Quick AutoML training"
14
15	## When to use this skill
16
17	Use this skill when you need to:
18	- Create new DataRobot projects
19	- Upload training datasets
20	- Configure AutoML experiments
21	- Monitor training progress
22	- Select and compare models
23	- Understand feature engineering results
24	- Export trained models
25
26	## Key capabilities
27
28	### 1. Project Management
29
30	- Create new projects with appropriate settings
31	- Upload datasets (CSV, Parquet, database connections)
32	- Configure project settings (target, partitioning, time series)
33	- Manage multiple projects and experiments
34
35	### 2. AutoML Configuration
36
37	- Set training modes (Quick, Manual, Comprehensive)
38	- Configure feature engineering options
39	- Set time limits and resource constraints
40	- Choose algorithms and model types
41
42	### 3. Training Execution
43
44	- Start AutoML training runs
45	- Monitor training progress
46	- Handle training errors and warnings
47	- Pause/resume training if needed
48
49	### 4. Model Analysis
50
51	- Compare model performance metrics
52	- Review feature importance
53	- Analyze model insights and explanations
54	- Select best models for deployment
55
56	## Workflow examples
57
58	### Example 1: Create and train a new project
59
60	User request: "Create a new project using my sales_data.csv file, predict 'revenue' as the target, and start AutoML training."
61
62	Agent workflow:
63	1. Upload the dataset to DataRobot
64	2. Create a new project with the dataset
65	3. Set 'revenue' as the target variable
66	4. Configure project settings (detect partitioning, handle time series if needed)
67	5. Start AutoML training with appropriate mode
68	6. Monitor training progress
69	7. Report when training completes with top model metrics
70
71	### Example 2: Configure advanced training options
72
73	User request: "Train a model with time series settings: datetime column 'date', series ID 'store_id', forecast window 1-7 days."
74
75	Agent workflow:
76	1. Create project with time series configuration
77	2. Set datetime column and series ID columns
78	3. Configure forecast window (1-7 days)
79	4. Set appropriate time series validation
80	5. Start training with time series-aware algorithms
81	6. Monitor progress and report results
82
83	## Using DataRobot SDK
84
85	This skill guides you to use the DataRobot Python SDK directly. Install the SDK if needed:
86
87	```bash
88	pip install datarobot
89	```
90
91	### Key SDK Operations
92
93	Use these DataRobot SDK methods for model training:
94
95	Projects:
96	- `dr.Project.create_from_dataset(dataset_id, project_name)` - Create project
97	- `dr.Project.get(project_id)` - Get project details
98	- `dr.Project.list()` - List all projects
99	- `project.set_target(target_column)` - Set target variable
100
101	Training:
102	- `project.start(autopilot_on=True)` - Start AutoML training
103	- `project.get_status()` - Check training status
104	- `dr.Model.list(project_id)` - List trained models
105	- `dr.Model.get(model_id)` - Get model details
106
107	Model Analysis:
108	- `model.get_metrics()` - Get performance metrics
109	- `model.get_feature_impact()` - Get feature importance
110
111	See the [Common Patterns](#common-patterns) section below for complete examples.
112
113	## Helper Scripts
114
115	This skill includes executable helper scripts that Claude can run directly:
116
117	- `scripts/create_project.py` - Create a new project from a dataset
118	- `scripts/start_training.py` - Start AutoML training
119	- `scripts/list_models.py` - List trained models with metrics
120
121	Usage example:
122	```bash
123	# Create project and set target
124	python scripts/create_project.py dataset_123 "Sales Prediction" revenue
125
126	# Start training
127	python scripts/start_training.py project_456 Quick
128
129	# List models
130	python scripts/list_models.py project_456 AUC
131	```
132
133	Claude can run these scripts directly or use them as reference when writing code.
134
135	## Best practices
136
137	1. Data preparation: Ensure data is clean and properly formatted before upload
138	2. Target selection: Choose appropriate target variable (avoid leakage)
139	3. Partitioning: Use proper partitioning for time-aware or grouped data
140	4. Feature engineering: Let AutoML handle feature engineering, but review results
141	5. Model selection: Compare multiple models, not just the top performer
142	6. Validation: Review validation strat

People who liked this also viewed

6 picks

Type

datarobot-model-training

22stars19forks

View on GitHub

How to install datarobot-model-training?

datarobot-oss/datarobot-agent-skills/datarobot-model-training

$npx -y skills add datarobot-oss/datarobot-agent-skills --skill datarobot-model-training

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# DataRobot Model Training Skill
2
3	This skill provides guidance for the complete model training workflow in DataRobot, from project creation through model selection and validation.
4
5	## Quick Start
6
7	Most common use case: Create a project and train models
8
9	1. Upload dataset: `upload_dataset(file_path, dataset_name)` to upload training data
10	2. Create project: `create_project(dataset_id, project_name)` to create new project
11	3. Start training: `start_automl(project_id, mode)` to begin AutoML training
12
13	Example: "Create a new project with sales_data.csv, set 'revenue' as target, and start Quick AutoML training"
14
15	## When to use this skill
16
17	Use this skill when you need to:
18	- Create new DataRobot projects
19	- Upload training datasets
20	- Configure AutoML experiments
21	- Monitor training progress
22	- Select and compare models
23	- Understand feature engineering results
24	- Export trained models
25
26	## Key capabilities
27
28	### 1. Project Management
29
30	- Create new projects with appropriate settings
31	- Upload datasets (CSV, Parquet, database connections)
32	- Configure project settings (target, partitioning, time series)
33	- Manage multiple projects and experiments
34
35	### 2. AutoML Configuration
36
37	- Set training modes (Quick, Manual, Comprehensive)
38	- Configure feature engineering options
39	- Set time limits and resource constraints
40	- Choose algorithms and model types
41
42	### 3. Training Execution
43
44	- Start AutoML training runs
45	- Monitor training progress
46	- Handle training errors and warnings
47	- Pause/resume training if needed
48
49	### 4. Model Analysis
50
51	- Compare model performance metrics
52	- Review feature importance
53	- Analyze model insights and explanations
54	- Select best models for deployment
55
56	## Workflow examples
57
58	### Example 1: Create and train a new project
59
60	User request: "Create a new project using my sales_data.csv file, predict 'revenue' as the target, and start AutoML training."
61
62	Agent workflow:
63	1. Upload the dataset to DataRobot
64	2. Create a new project with the dataset
65	3. Set 'revenue' as the target variable
66	4. Configure project settings (detect partitioning, handle time series if needed)
67	5. Start AutoML training with appropriate mode
68	6. Monitor training progress
69	7. Report when training completes with top model metrics
70
71	### Example 2: Configure advanced training options
72
73	User request: "Train a model with time series settings: datetime column 'date', series ID 'store_id', forecast window 1-7 days."
74
75	Agent workflow:
76	1. Create project with time series configuration
77	2. Set datetime column and series ID columns
78	3. Configure forecast window (1-7 days)
79	4. Set appropriate time series validation
80	5. Start training with time series-aware algorithms
81	6. Monitor progress and report results
82
83	## Using DataRobot SDK
84
85	This skill guides you to use the DataRobot Python SDK directly. Install the SDK if needed:
86
87	```bash
88	pip install datarobot
89	```
90
91	### Key SDK Operations
92
93	Use these DataRobot SDK methods for model training:
94
95	Projects:
96	- `dr.Project.create_from_dataset(dataset_id, project_name)` - Create project
97	- `dr.Project.get(project_id)` - Get project details
98	- `dr.Project.list()` - List all projects
99	- `project.set_target(target_column)` - Set target variable
100
101	Training:
102	- `project.start(autopilot_on=True)` - Start AutoML training
103	- `project.get_status()` - Check training status
104	- `dr.Model.list(project_id)` - List trained models
105	- `dr.Model.get(model_id)` - Get model details
106
107	Model Analysis:
108	- `model.get_metrics()` - Get performance metrics
109	- `model.get_feature_impact()` - Get feature importance
110
111	See the [Common Patterns](#common-patterns) section below for complete examples.
112
113	## Helper Scripts
114
115	This skill includes executable helper scripts that Claude can run directly:
116
117	- `scripts/create_project.py` - Create a new project from a dataset
118	- `scripts/start_training.py` - Start AutoML training
119	- `scripts/list_models.py` - List trained models with metrics
120
121	Usage example:
122	```bash
123	# Create project and set target
124	python scripts/create_project.py dataset_123 "Sales Prediction" revenue
125
126	# Start training
127	python scripts/start_training.py project_456 Quick
128
129	# List models
130	python scripts/list_models.py project_456 AUC
131	```
132
133	Claude can run these scripts directly or use them as reference when writing code.
134
135	## Best practices
136
137	1. Data preparation: Ensure data is clean and properly formatted before upload
138	2. Target selection: Choose appropriate target variable (avoid leakage)
139	3. Partitioning: Use proper partitioning for time-aware or grouped data
140	4. Feature engineering: Let AutoML handle feature engineering, but review results
141	5. Model selection: Compare multiple models, not just the top performer
142	6. Validation: Review validation strat

People who liked this also viewed

6 picks

Type