Documentation

Welcome to the developer documentation for SigOpt. If you have a question you can’t answer, feel free to contact us!
Welcome to the new SigOpt docs! If you're looking for the classic SigOpt documentation then you can find that here. Otherwise, happy optimizing!

Record SigOpt Runs with Python IDE + SigOpt CLI

With SigOpt installed and your Python environment set up, let's take a look at how to record a SigOpt Run in a Python IDE with the SigOpt CLI.

Instrument the Run

import tensorflow as tf
import sigopt
import os

os.environ["SIGOPT_API_TOKEN"] = # SIGOPT_API_TOKEN
os.environ["SIGOPT_PROJECT"] = "run-examples"


class KerasNNModel:
  def __init__(self, hidden_layer_size, activation_fn):
    model = tf.keras.Sequential(
      [
        tf.keras.layers.Flatten(input_shape=(28, 28)),
        tf.keras.layers.Dense(hidden_layer_size, activation=activation_fn),
        tf.keras.layers.Dense(10),
      ]
    )
    self.model = model

  def get_keras_nn_model(self):
    return self.model

  def train_model(self, train_images, train_labels, optimizer_type, metrics_list, num_epochs):
    self.model.compile(
      optimizer=optimizer_type,
      loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
      metrics=metrics_list,
    )
    self.model.fit(train_images, train_labels, epochs=num_epochs)

  def evaluate_model(self, test_images, test_labels):
    metrics_dict = self.model.evaluate(test_images, test_labels, verbose=2, return_dict=True)
    return metrics_dict


def load_data_train_model():
  sigopt.log_dataset(name="mnist")
  (train_images, train_labels), (test_images, test_labels) = tf.keras.datasets.mnist.load_data()

  # set model training, architecture parameters and hyperparameters
  sigopt.params.num_epochs = 2
  sigopt.params.hidden_layer_size = 200
  sigopt.params.activation_fn = "relu"

  # create the model
  keras_nn_model = KerasNNModel(
    hidden_layer_size=sigopt.params.hidden_layer_size, activation_fn=sigopt.params.activation_fn
  )
  sigopt.log_model("Keras NN Model with 1 Hidden layer")

  # train the model
  keras_nn_model.train_model(train_images, train_labels, "adam", ["accuracy"], sigopt.params.num_epochs)
  sigopt.log_metadata("sgd optimizer", "adam")
  metrics_dict = keras_nn_model.evaluate_model(test_images, test_labels)

  # log performance metrics
  sigopt.log_metric("accuracy", metrics_dict["accuracy"])
  sigopt.log_metric("loss", metrics_dict["loss"])


if __name__ == "__main__":
  load_data_train_model()

Run the Code with the SigOpt CLI

$ sigopt run python keras_model.py
Run started, view it on the SigOpt dashboard at https://app.sigopt.com/run/1234
Epoch 1/2 
 1875/1875 [==============================] - 5s 2ms/step - loss: 2.7513 - accuracy: 0.8826 
Epoch 2/2 
 1875/1875 [==============================] - 4s 2ms/step - loss: 0.3313 - accuracy: 0.9265 
313/313 - 0s - loss: 0.2941 - accuracy: 0.9478 
Run finished, view it on the SigOpt dashboard at https://app.sigopt.com/run/1234

Orchestrate a Run with the SigOpt CLI

SigOpt Orchestrate allows you to confidently scale your modeling experimentation on either your existing Kubernetes infrastructure or a Kubernetes cluster managed by SigOpt. To set up SigOpt Orchestrate, follow this installation guide. With SigOpt Orchestrate installed, all you have to do is set up yourrun.yml file and execute the following command locally to orchestrate your Run on your Kubernetes cluster:

#run.yml 
name: My Run 
run: python keras_model.py 
resources: 
  limits: 
    cpu: 2 
    memory: 4Gi 
  gpus: 1 
image: my-run 
$ sigopt cluster run --run-file run.yml

When you execute the above, SigOpt Orchestrate will look into the specified run.yml file. It will allocate 2 CPUs, 4 GiB memory, and 1 GPU of your kubernetes cluster for your Run, and will execute the command provided for run.

Run started, view it on the SigOpt dashboard at https://app.sigopt.com/run/1234
Epoch 1/2 
 1875/1875 [==============================] - 5s 2ms/step - loss: 2.7513 - accuracy: 0.8826 
Epoch 2/2 
 1875/1875 [==============================] - 4s 2ms/step - loss: 0.3313 - accuracy: 0.9265 
313/313 - 0s - loss: 0.2941 - accuracy: 0.9478 
Run finished, view it on the SigOpt dashboard at https://app.sigopt.com/run/1234