Fit a Model

Uploads and fits a TabPFN model on your training data. The API automatically handles preprocessing and stores a reference to your trained context (not the model weights). You can use either a single dataset file (with the target column included) or separate feature and label files.

POST /v1/fit

Python

Single CSV upload

import os, json, requests # Define your dataset path train_path = "train.csv" # Get your API key from the environment api_key = os.getenv("PRIORLABS_API_KEY") headers = {"Authorization": f"Bearer {api_key}"} # Upload your training dataset to /v1/fit payload = { "task": "classification", "schema": { "target": "churn", "description": "Customer churn dataset" } } files = { "data": (None, json.dumps(payload), "application/json"), "dataset_file": (train_path, open(train_path, "rb")), } fit_response = requests.post( "https://api.priorlabs.ai/v1/fit", headers=headers, files=files, ) model_id = fit_response.json().get("model_id") print(f"✅ Model trained: {model_id}")

200

Example

{ "model_id": "123e4567-e89b-12d3-a456-426614174000", "task": "classification" }

Authorizations

Authorization · string · header · required

Bearer token for authentication, obtained after signing up and generating an API key.

Body

multipart/form-data

Option 1

Option 2

data · string · required

A JSON string defining the training configuration.

Supported Systems:

• ["preprocessing"] - Applies skrub preprocessing,
• ["text"] - Adds text embeddings for text columns.

Default: ["preprocessing", "text"].

Supported Config Parameters:

• n_estimators (int, 1-10) - Number of ensemble estimators,
• softmax_temperature (float) - Temperature for softmax scaling,
• average_before_softmax (bool) - Average before softmax,
• ignore_pretraining_limits (bool) - Ignore pretraining limits,
• random_state (int) - Random seed for reproducibility.

dataset_file · file · required

Option 1: CSV file containing both features (X_train) and labels (y_train). Use this when you have all data in a single file.

features_file · file

Option 2:: CSV file containing only feature columns (X_train). Must be used together with labels_file.

labels_file · file

Option 2: CSV file containing only the target/label column (y_train). Must be used together with features_file.

Response

200

application/json

Model fitted successfully — returns a model ID for later prediction calls.

model_id · string<uuid> · required

Unique identifier for the fitted model (used for prediction calls).

task · enum<string> · required

Specifies the type of task to perform — either classification or regression.

Available options: classification, regression