Natural Language Search

Natural language search allows you to search for information using natural language queries, offering considerable performance gains above traditional search algorithms. The TrueState Python SDK provides a high-level API for making predictions on your searchable datasets.

To create a searchable dataset, visit the Text Embedding documentation.

from truestate.inference import search

query = "lightweight camping gear for backpacking"

dataset_id = "your-dataset-id"

results = search(query, dataset_id)

print(results)
# [
#     {
#         "id": "UL001",
#         "name": "Featherlight Pro X2 Tent",
#         "description": "Ultra-lightweight 2-person tent weighing 1.75 lbs (794g). Made with Dyneema Composite Fabric for ultimate strength-to-weight ratio. Ideal for thru-hikers on long-distance trails."
#     },
#     {
#         "id": "UL002",
#         "name": "AeroDown Ultralight 5°F Sleeping Bag",
#         "description": "Premium 5°F (-15°C) rated sleeping bag weighing only 1.2 lbs (544g). Features 950+ fill power down with water-repellent treatment. Perfect for long-distance backpacking in varied conditions."
#     },
#     {
#         "id": "UL003",
#         "name": "TitaniumFlame Micro Stove System",
#         "description": "Integrated stove system weighing 5.6 oz (159g) including pot. Boils water in 100 seconds and nests with a fuel canister. Engineered for ultralight long-distance hikers."
#     }
# ]