Nanthini10

from azureml.interpret import ExplanationClient

# Get model explanation data
client = ExplanationClient.from_run(run)
global_explanation = client.download_model_explanation()
local_importance_values = global_explanation.local_importance_values
expected_values = global_explanation.expected_values

# Or you can use the saved run.id to retrive the feature importance values
client = ExplanationClient.from_run_id(ws, experiment_name, run.id)

Nanthini10

from azureml.core import Environment

environment_name = "rapids"

env = Environment(environment_name)
env.docker.enabled = True
env.docker.base_image = None
env.docker.base_dockerfile = """

Nanthini10

Nanthini10

analysis = tune.run(
    WrappedTrainable,
    name=exp_name,
    scheduler=sched,
    search_alg=search,
    stop={"training_iteration": CV_folds, "is_bad": True,},
    resources_per_trial={"cpu": cpu_per_sample, "gpu": int(compute == "GPU")},
    num_samples=num_samples,
    checkpoint_at_end=True,
    keep_checkpoints_num=1,

Nanthini10

def _train(self):
  iteration = getattr(self, "iteration", 0)

  if compute == "GPU":
      # split data
      X_train, X_test, y_train, y_test = train_test_split(
          X=self._dataset,
          y=self._y_label,
          train_size=0.8,
          shuffle=True,