1
import data_designer.config as dd
2
from data_designer.interface import DataDesigner
3
 

1
data_designer = DataDesigner()
2
 

1
# This name is set in the model provider configuration.
2
MODEL_PROVIDER = "nvidia"
3
 
4
# The model ID is from build.nvidia.com.
5
MODEL_ID = "nvidia/nemotron-3-nano-30b-a3b"
6
 
7
# We choose this alias to be descriptive for our use case.
8
MODEL_ALIAS = "nemotron-nano-v3"
9
 
10
model_configs = [
11
    dd.ModelConfig(
12
        alias=MODEL_ALIAS,
13
        model=MODEL_ID,
14
        provider=MODEL_PROVIDER,
15
        inference_parameters=dd.ChatCompletionInferenceParams(
16
            temperature=1.0,
17
            top_p=1.0,
18
            max_tokens=2048,
19
            extra_body={"chat_template_kwargs": {"enable_thinking": False}},
20
        ),
21
    )
22
]
23
 

1
config_builder = dd.DataDesignerConfigBuilder(model_configs=model_configs)
2
 

1
config_builder.info.display("samplers")
2
 

1
config_builder.add_column(
2
    dd.SamplerColumnConfig(
3
        name="product_category",
4
        sampler_type=dd.SamplerType.CATEGORY,
5
        params=dd.CategorySamplerParams(
6
            values=[
7
                "Electronics",
8
                "Clothing",
9
                "Home & Kitchen",
10
                "Books",
11
                "Home Office",
12
            ],
13
        ),
14
    )
15
)
16
 
17
config_builder.add_column(
18
    dd.SamplerColumnConfig(
19
        name="product_subcategory",
20
        sampler_type=dd.SamplerType.SUBCATEGORY,
21
        params=dd.SubcategorySamplerParams(
22
            category="product_category",
23
            values={
24
                "Electronics": [
25
                    "Smartphones",
26
                    "Laptops",
27
                    "Headphones",
28
                    "Cameras",
29
                    "Accessories",
30
                ],
31
                "Clothing": [
32
                    "Men's Clothing",
33
                    "Women's Clothing",
34
                    "Winter Coats",
35
                    "Activewear",
36
                    "Accessories",
37
                ],
38
                "Home & Kitchen": [
39
                    "Appliances",
40
                    "Cookware",
41
                    "Furniture",
42
                    "Decor",
43
                    "Organization",
44
                ],
45
                "Books": [
46
                    "Fiction",
47
                    "Non-Fiction",
48
                    "Self-Help",
49
                    "Textbooks",
50
                    "Classics",
51
                ],
52
                "Home Office": [
53
                    "Desks",
54
                    "Chairs",
55
                    "Storage",
56
                    "Office Supplies",
57
                    "Lighting",
58
                ],
59
            },
60
        ),
61
    )
62
)
63
 
64
config_builder.add_column(
65
    dd.SamplerColumnConfig(
66
        name="target_age_range",
67
        sampler_type=dd.SamplerType.CATEGORY,
68
        params=dd.CategorySamplerParams(values=["18-25", "25-35", "35-50", "50-65", "65+"]),
69
    )
70
)
71
 
72
# Optionally validate that the columns are configured correctly.
73
data_designer.validate(config_builder)
74
 

1
config_builder.add_column(
2
    dd.SamplerColumnConfig(
3
        name="customer",
4
        sampler_type=dd.SamplerType.PERSON_FROM_FAKER,
5
        params=dd.PersonFromFakerSamplerParams(age_range=[18, 70], locale="en_US"),
6
    )
7
)
8
 
9
config_builder.add_column(
10
    dd.SamplerColumnConfig(
11
        name="number_of_stars",
12
        sampler_type=dd.SamplerType.UNIFORM,
13
        params=dd.UniformSamplerParams(low=1, high=5),
14
        convert_to="int",  # Convert the sampled float to an integer.
15
    )
16
)
17
 
18
config_builder.add_column(
19
    dd.SamplerColumnConfig(
20
        name="review_style",
21
        sampler_type=dd.SamplerType.CATEGORY,
22
        params=dd.CategorySamplerParams(
23
            values=["rambling", "brief", "detailed", "structured with bullet points"],
24
            weights=[1, 2, 2, 1],
25
        ),
26
    )
27
)
28
 
29
data_designer.validate(config_builder)
30
 

1
config_builder.add_column(
2
    dd.LLMTextColumnConfig(
3
        name="product_name",
4
        prompt=(
5
            "You are a helpful assistant that generates product names. DO NOT add quotes around the product name.\n\n"
6
            "Come up with a creative product name for a product in the '{{ product_category }}' category, focusing "
7
            "on products related to '{{ product_subcategory }}'. The target age range of the ideal customer is "
8
            "{{ target_age_range }} years old. Respond with only the product name, no other text."
9
        ),
10
        model_alias=MODEL_ALIAS,
11
    )
12
)
13
 
14
config_builder.add_column(
15
    dd.LLMTextColumnConfig(
16
        name="customer_review",
17
        prompt=(
18
            "You are a customer named {{ customer.first_name }} from {{ customer.city }}, {{ customer.state }}. "
19
            "You are {{ customer.age }} years old and recently purchased a product called {{ product_name }}. "
20
            "Write a review of this product, which you gave a rating of {{ number_of_stars }} stars. "
21
            "The style of the review should be '{{ review_style }}'. "
22
            "Respond with only the review, no other text."
23
        ),
24
        model_alias=MODEL_ALIAS,
25
    )
26
)
27
 
28
data_designer.validate(config_builder)
29
 

1
preview = data_designer.preview(config_builder, num_records=2)
2
 

1
# Run this cell multiple times to cycle through the 2 preview records.
2
preview.display_sample_record()
3
 

1
# The preview dataset is available as a pandas DataFrame.
2
preview.dataset
3
 

1
# Print the analysis as a table.
2
preview.analysis.to_report()
3
 

1
results = data_designer.create(config_builder, num_records=10, dataset_name="tutorial-1")
2
 

1
# Load the generated dataset as a pandas DataFrame.
2
dataset = results.load_dataset()
3
 
4
dataset.head()
5
 

1
# Load the analysis results into memory.
2
analysis = results.load_analysis()
3
 
4
analysis.to_report()
5