Hierarchy classification

In some instances it's necessary to classify text data into a hierarchy of categories. The TrueState Python SDK provides a high-level API for making predictions on your hierarchy classification models.

from truestate.inference import hierarchy_classificaiton

sample_text = "Amazon is a multinational technology company that focuses on e-commerce, cloud computing, digital streaming, and artificial intelligence. The primary business of Amazon.com is the sale of consumer goods and subscriptions."

hierarchy = ClassHierarchy(
    name="company-categorisation",
    choices=[
        Category(
            label="Software company",
            criteria="this text contains a description of a software company",
            subcategories=[
                Category(
                    label="Ecommerce",
                    criteria="this text contains a description of an ecommerce company",
                ),
                Category(
                    label="Blockchain",
                    criteria="this text contains a description of an web3.0 / blockchain company",
                ),
            ],
        ),
    ],
)

result = hierarchy_classificaiton(sample_text, hierarchy)
print(result)
# {
#     "text": "Amazon is a multinational technology company that focuses on e-commerce, cloud computing, digital streaming, and artificial intelligence. The primary business of Amazon.com is the sale of consumer goods and subscriptions.",
#    "categories": {
#       "category_1": "Software company",
#       "category_2": "Ecommerce"
# }