Tabular/Structured Data

Guides

Build models for structured/tabular data prediction.

What You’ll Build

AI models for tabular data:

Classification - Predict categories from features
Regression - Predict numeric values
Time series - Forecast future values

Prerequisites

A SeeMe.ai account (sign up)
CSV or structured data
(Optional) Python environment with seeme SDK installed

Supported Algorithms

SeeMe.ai supports multiple frameworks:

Framework	Best For
XGBoost	General purpose, competitions
LightGBM	Large datasets, speed
CatBoost	Categorical features
fastai	Deep learning on tabular

Steps

1. Prepare Data 2. Train Model 3. Optimize 4. Deploy 5. Monitor

Quick Start

from seeme import Client
import pandas as pd

client = Client()

## Create tabular dataset
dataset = client.create_dataset(
    name="Sales Prediction",
    description="Predict sales from features",
    content_type="tabular"
)

# Upload data from DataFrame
df = pd.read_csv("sales_data.csv")
client.upload_df(
    dataset_id=dataset.id,
    df=df,
    target_column="sales"
)

# Train model
model = client.create_model(
    name="Sales Predictor",
    application_id="xgboost_2_0_2"
)

job = client.create_job(
    job_type="training",
    model_id=model.id,
    dataset_id=dataset.id
)

Data Requirements

CSV format
Headers in first row
One row per sample
Target column for supervised learning