id,category1,category2,feature1,target
1,A,X,0.3373961604172684,1
2,B,Z,0.6481718720511972,0
3,A,Y,0.36824153984054797,1
4,B,Z,0.9571551589530464,1
5,B,Z,0.14035078041264515,1
6,C,X,0.8700872583584364,1
7,A,Y,0.4736080452737105,0
8,C,Y,0.8009107519796442,1
9,A,Y,0.5204774795512048,0
10,A,Y,0.6788795301189603,0
.
.
.

❯ autotrain tabular --help
usage: autotrain <command> [<args>] tabular [-h] [--train] [--deploy] [--inference] [--username USERNAME]
                                            [--backend {local-cli,spaces-a10gl,spaces-a10gs,spaces-a100,spaces-t4m,spaces-t4s,spaces-cpu,spaces-cpuf}]
                                            [--token TOKEN] [--push-to-hub] --model MODEL --project-name PROJECT_NAME [--data-path DATA_PATH]
                                            [--train-split TRAIN_SPLIT] [--valid-split VALID_SPLIT] [--batch-size BATCH_SIZE] [--seed SEED]
                                            --target-columns TARGET_COLUMNS [--categorical-columns CATEGORICAL_COLUMNS]
                                            [--numerical-columns NUMERICAL_COLUMNS] --id-column ID_COLUMN --task {classification,regression}
                                            [--num-trials NUM_TRIALS] [--time-limit TIME_LIMIT] [--categorical-imputer {most_frequent,None}]
                                            [--numerical-imputer {mean,median,None}] [--numeric-scaler {standard,minmax,normal,robust}]

✨ Run AutoTrain Tabular Data Training

options:
  -h, --help            show this help message and exit
  --train               Command to train the model
  --deploy              Command to deploy the model (limited availability)
  --inference           Command to run inference (limited availability)
  --username USERNAME   Hugging Face Hub Username
  --backend {local-cli,spaces-a10gl,spaces-a10gs,spaces-a100,spaces-t4m,spaces-t4s,spaces-cpu,spaces-cpuf}
                        Backend to use: default or spaces. Spaces backend requires push_to_hub & username. Advanced users only.
  --token TOKEN         Your Hugging Face API token. Token must have write access to the model hub.
  --push-to-hub         Push to hub after training will push the trained model to the Hugging Face model hub.
  --model MODEL         Base model to use for training
  --project-name PROJECT_NAME
                        Output directory / repo id for trained model (must be unique on hub)
  --data-path DATA_PATH
                        Train dataset to use. When using cli, this should be a directory path containing training and validation data in appropriate
                        formats
  --train-split TRAIN_SPLIT
                        Train dataset split to use
  --valid-split VALID_SPLIT
                        Validation dataset split to use
  --batch-size BATCH_SIZE
                        Training batch size to use
  --seed SEED           Random seed for reproducibility
  --target-columns TARGET_COLUMNS
                        Specify the names of the target or label columns separated by commas if multiple. These columns are what the model will
                        predict. Required for defining the output of the model.
  --categorical-columns CATEGORICAL_COLUMNS
                        List the names of columns that contain categorical data, useful for models that need explicit handling of such data.
                        Categorical data is typically processed differently from numerical data, such as through encoding. If not specified, the
                        model will infer the data type.
  --numerical-columns NUMERICAL_COLUMNS
                        Identify columns that contain numerical data. Proper specification helps in applying appropriate scaling and normalization
                        techniques, which can significantly impact model performance. If not specified, the model will infer the data type.
  --id-column ID_COLUMN
                        Specify the column name that uniquely identifies each row in the dataset. This is critical for tracking samples through the
                        model pipeline and is often excluded from model training. Required field.
  --task {classification,regression}
                        Define the type of machine learning task, such as 'classification', 'regression'. This parameter determines the model's
                        architecture and the loss function to use. Required to properly configure the model.
  --num-trials NUM_TRIALS
                        Set the number of trials for hyperparameter tuning or model experimentation. More trials can lead to better model
                        configurations but require more computational resources. Default is 100 trials.
  --time-limit TIME_LIMIT
                        mpose a time limit (in seconds) for training or searching for the best model configuration. This helps manage resource
                        allocation and ensures the process does not exceed available computational budgets. The default is 3600 seconds (1 hour).
  --categorical-imputer {most_frequent,None}
                        Select the method or strategy to impute missing values in categorical columns. Options might include 'most_frequent',
                        'None'. Correct imputation can prevent biases and improve model accuracy.
  --numerical-imputer {mean,median,None}
                        Choose the imputation strategy for missing values in numerical columns. Common strategies include 'mean', & 'median'.
                        Accurate imputation is vital for maintaining the integrity of numerical data.
  --numeric-scaler {standard,minmax,normal,robust}
                        Determine the type of scaling to apply to numerical data. Examples include 'standard' (zero mean and unit variance), 'min-
                        max' (scaled between given range), etc. Scaling is essential for many algorithms to perform optimally