azureml + ray (#344)

* examples and documentation about how to use azureml + ray

test/ray/distribute_tune.py

@@ -0,0 +1,49 @@
+import ray
+import lightgbm as lgb
+import numpy as np
+import sklearn.datasets
+import sklearn.metrics
+from sklearn.model_selection import train_test_split
+from flaml import tune
+from flaml.model import LGBMEstimator
+
+data, target = sklearn.datasets.load_breast_cancer(return_X_y=True)
+train_x, test_x, train_y, test_y = train_test_split(data, target, test_size=0.25)
+
+
+def train_breast_cancer(config):
+    params = LGBMEstimator(**config).params
+    num_boost_round = params.pop("n_estimators")
+    train_set = lgb.Dataset(train_x, label=train_y)
+    gbm = lgb.train(params, train_set, num_boost_round)
+    preds = gbm.predict(test_x)
+    pred_labels = np.rint(preds)
+    tune.report(
+        mean_accuracy=sklearn.metrics.accuracy_score(test_y, pred_labels), done=True
+    )
+
+
+if __name__ == "__main__":
+    ray.init(address="auto")
+    flaml_lgbm_search_space = LGBMEstimator.search_space(train_x.shape)
+    config_search_space = {
+        hp: space["domain"] for hp, space in flaml_lgbm_search_space.items()
+    }
+    low_cost_partial_config = {
+        hp: space["low_cost_init_value"]
+        for hp, space in flaml_lgbm_search_space.items()
+        if "low_cost_init_value" in space
+    }
+
+    analysis = tune.run(
+        train_breast_cancer,
+        metric="mean_accuracy",
+        mode="max",
+        config=config_search_space,
+        num_samples=-1,
+        time_budget_s=60,
+        use_ray=True,
+    )
+
+    # print("Best hyperparameters found were: ", analysis.best_config)
+    print("The best trial's result: ", analysis.best_trial.last_result)