TensorFlow: Tabular Regression

Predicting Exoplanet Surface Temperature Using Kepler Satellite Sensor Data.

💾 Data

Reference Example Datasets for more information.

This dataset is comprised of:

• Features = characteristics of the planet in the context of its solar system.

• Label = the temperature of the planet.

from aiqc import datum
df = datum.to_pandas('exoplanets.parquet')
df.sample(5)

	TypeFlag	PlanetaryMassJpt	PeriodDays	SurfaceTempK	DistFromSunParsec	HostStarMassSlrMass	HostStarRadiusSlrRad	HostStarMetallicity	HostStarTempK
5	0	0.2500	19.224180	707.2	650.00	1.070	1.0200	0.12	5777.0
6	0	0.1700	39.031060	557.9	650.00	1.070	1.0200	0.12	5777.0
7	0	0.0220	1.592851	1601.5	650.00	1.070	1.0200	0.12	5777.0
15	0	1.2400	2.705782	2190.0	200.00	1.630	2.1800	0.12	6490.0
16	0	0.0195	1.580404	604.0	14.55	0.176	0.2213	0.10	3250.0

from aiqc.orm import Dataset
shared_dataset = Dataset.Tabular.from_df(df)

---

🚰 Pipeline

Reference High-Level API Docs for more information.

from aiqc.mlops import Pipeline, Input, Target, Stratifier
from sklearn.preprocessing import StandardScaler, RobustScaler, OneHotEncoder

pipeline = Pipeline(
    Input(
        dataset  = shared_dataset,
        encoders = [
            Input.Encoder(
                RobustScaler(),
                dtypes = ['float64']
            ),
            Input.Encoder(
                OneHotEncoder(),
                dtypes = ['int64']
            )
        ]
    ),

    Target(
        dataset   = shared_dataset
        , column  = 'SurfaceTempK'
        , encoder = Target.Encoder(StandardScaler())
    ),

    Stratifier(
        size_test         = 0.12
        , size_validation = 0.22
        , fold_count      = None
        , bin_count       = 4
    )
)

└── Info - System overriding user input to set `sklearn_preprocess.copy=False`.
        This saves memory when concatenating the output of many encoders.


└── Info - System overriding user input to set `sklearn_preprocess.copy=False`.
        This saves memory when concatenating the output of many encoders.


└── Info - System overriding user input to set `sklearn_preprocess.sparse=False`.
        This would have generated 'scipy.sparse.csr.csr_matrix', causing Keras training to fail.

---

🧪 Experiment

Reference High-Level API Docs for more information.

from aiqc.mlops import Experiment, Architecture, Trainer
import tensorflow as tf
from tensorflow.keras import layers as l
from aiqc.utils.tensorflow import TrainingCallback

def fn_build(features_shape, label_shape, **hp):
    m = tf.keras.models.Sequential()
    m.add(l.Input(shape=features_shape))

    # Example of using hyperparameters to tweak topology.
    # with 'block' for each layer.
    for block in range(hp['blocks']):
        # Example of using hyperparameters to tweak topology.
        m.add(l.Dense(hp['neuron_count']))

        # Example of using hyperparameters to tweak topology.
        # BatchNorm, Activation, Dropout (B.A.D.)
        if (hp['batch_norm'] == True):
            m.add(l.BatchNormalization())

        m.add(l.Activation('relu'))
        m.add(l.Dropout(0.2))

    m.add(l.Dense(label_shape[0]))
    return m

def fn_train(
    model, loser, optimizer,
    train_features, train_label,
    eval_features, eval_label,
    **hp
):
    model.compile(
        loss        = loser
        , optimizer = optimizer
        , metrics   = ['mean_squared_error']
    )

    metrics_cuttoffs = [
        dict(metric='val_loss', cutoff=0.025, above_or_below='below'),
        dict(metric='loss', cutoff=0.025, above_or_below='below'),
    ]
    cutoffs = TrainingCallback.MetricCutoff(metrics_cuttoffs)

    model.fit(
        train_features, train_label
        , validation_data = (eval_features, eval_label)
        , verbose         = 0
        , batch_size      = hp['batch_size']
        , callbacks       = [tf.keras.callbacks.History(), cutoffs]
        , epochs          = hp['epoch_count']
    )
    return model

hyperparameters = dict(
    batch_size      = [3]
    , blocks        = [2]
    , batch_norm    = [True, False]
    , epoch_count   = [75]
    , neuron_count  = [24, 36]
    , learning_rate = [0.01]
)

experiment = Experiment(
    Architecture(
        library           = "keras"
        , analysis_type   = "regression"
        , fn_build        = fn_build
        , fn_train        = fn_train
        , hyperparameters = hyperparameters
    ),

    Trainer(
        pipeline       = pipeline
        , repeat_count = 2
    )
)

experiment.run_jobs()

📦 Caching Splits 📦: 100%|██████████████████████████████████████████| 3/3 [00:00<00:00, 382.83it/s]
🔮 Training Models 🔮: 100%|██████████████████████████████████████████| 8/8 [01:39<00:00, 12.43s/it]

---

📊 Visualization & Interpretation

For more information on visualization of performance metrics, reference the Dashboard documentation.