RundialDocs

Appearance

Keras ​

Log Keras training metrics to a Rundial run.

Start Rundial and create an API key with the Quickstart. Keep the stack running and set RUNDIAL_API_KEY.

Install the packages:

bash
pip install rundial tensorflow

Add RundialKerasCallback to fit:

python
import numpy as np
from tensorflow import keras
from rundial.integrations import RundialKerasCallback

x = np.arange(32, dtype="float32").reshape(-1, 1) / 32
model = keras.Sequential([keras.layers.Input(shape=(1,)), keras.layers.Dense(1)])
model.compile(optimizer="adam", loss="mse")
model.fit(x, 2 * x, epochs=3, callbacks=[
    RundialKerasCallback(workspace="default-workspace", project="default-project", name="keras")
])

Open the run's Metrics tab to view epoch loss. Set log_batch_metrics=True on the callback to log batch metrics too.

Next steps ​