test_automl_when_clean_df.py

FAILED        [100%]Recreating table test_when_clean with data from file
Done
Task: cls
|   iter    |  target   | algo_i... | prepro... |
-------------------------------------------------
Child Iteration accuracy: 0.77528
Child Iteration accuracy: 0.786516
|  1        |  0.7865   |  2.063    |  1.061    |
Child Iteration accuracy: 0.786516
Child Iteration accuracy: 0.808988
|  2        |  0.809    |  6.983    |  0.008468 |
Child Iteration accuracy: 0.786516
Child Iteration accuracy: 0.808988
|  3        |  0.809    |  7.0      |  2.0      |
Child Iteration accuracy: 0.786516
Child Iteration accuracy: 0.808988
|  4        |  0.809    |  6.982    |  0.04241  |
Child Iteration accuracy: 0.786516
100%|██████████| 1/1 [00:00<00:00,  5.56it/s]

test_automl_when_clean_df.py:7 (test_main[BayesianOptimizer])
self = <bayes_opt.target_space.TargetSpace object at 0x11ea70f70>
params = {'algo_index_tuned': 6.99999761791255, 'preprocess_method': 1.9999986923442028}

    def probe(self, params):
        """
        Evaulates a single point x, to obtain the value y and then records them
        as observations.
    
        Notes
        -----
        If x has been previously seen returns a cached value of y.
    
        Parameters
        ----------
        x : ndarray
            a single point, with len(x) == self.dim
    
        Returns
        -------
        y : float
            target function value.
        """
        x = self._as_array(params)
    
        try:
>           target = self._cache[_hashable(x)]
E           KeyError: (6.99999761791255, 1.9999986923442028)

../../venv/lib/python3.9/site-packages/bayes_opt/target_space.py:191: KeyError

During handling of the above exception, another exception occurred:

self = <bayes_opt.target_space.TargetSpace object at 0x11eaeb520>
params = {'learning_rate': 0.5100947196305086, 'max_depth': 25.19044276398371, 'min_sample_weight_leaf': 8.62659476429784, 'n_estimators': 85.05593483656696, ...}

    def probe(self, params):
        """
        Evaulates a single point x, to obtain the value y and then records them
        as observations.
    
        Notes
        -----
        If x has been previously seen returns a cached value of y.
    
        Parameters
        ----------
        x : ndarray
            a single point, with len(x) == self.dim
    
        Returns
        -------
        y : float
            target function value.
        """
        x = self._as_array(params)
    
        try:
>           target = self._cache[_hashable(x)]
E           KeyError: (0.5100947196305086, 25.19044276398371, 8.62659476429784, 85.05593483656696, 1.6246289857511451)

../../venv/lib/python3.9/site-packages/bayes_opt/target_space.py:191: KeyError

During handling of the above exception, another exception occurred:

optimizer = 'BayesianOptimizer'
tmpdir = local('/private/var/folders/s_/9f0c9ct96xv0p4b0dvvd4xtr0000gn/T/pytest-of-danonchik/pytest-12/test_main_BayesianOptimizer_0')

    @pytest.mark.parametrize("optimizer", ["OptunaSearch", "BayesianOptimizer"])
    def test_main(optimizer, tmpdir):
        m = AutoML(connection_context)
    
>       m.fit(
            table_name="test_when_clean",
            file_path="../../data/cleaned_train.csv",
            target="Survived",
            id_column="PassengerId",
            steps=5,
            categorical_features=["Survived"],
            optimizer=optimizer,
        )

test_automl_when_clean_df.py:12: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
../../automl.py:96: in fit
    self.opt = pipe.train(
../../pipeline/pipeline.py:70: in train
    self.opt.tune()
../../optimizers/bayes.py:149: in tune
    opt.maximize(n_iter=self.iter, init_points=1)
../../venv/lib/python3.9/site-packages/bayes_opt/bayesian_optimization.py:185: in maximize
    self.probe(x_probe, lazy=False)
../../venv/lib/python3.9/site-packages/bayes_opt/bayesian_optimization.py:116: in probe
    self._space.probe(params)
../../venv/lib/python3.9/site-packages/bayes_opt/target_space.py:194: in probe
    target = self.target_func(**params)
../../optimizers/bayes.py:88: in objective
    opt.maximize(n_iter=1, init_points=1)
../../venv/lib/python3.9/site-packages/bayes_opt/bayesian_optimization.py:185: in maximize
    self.probe(x_probe, lazy=False)
../../venv/lib/python3.9/site-packages/bayes_opt/bayesian_optimization.py:116: in probe
    self._space.probe(params)
../../venv/lib/python3.9/site-packages/bayes_opt/target_space.py:194: in probe
    target = self.target_func(**params)
../../optimizers/bayes.py:115: in child_objective
    acc = algorithm.score(self.inner_data, self.inner_data.test)
../../algorithms/base_algo.py:23: in score
    return self.model.score(df, key=data.id_colm, label=data.target)
../../venv/lib/python3.9/site-packages/hana_ml/algorithms/pal/trees.py:4362: in score
    prediction = self.predict(data=data, key=key, features=features,
../../venv/lib/python3.9/site-packages/hana_ml/algorithms/pal/trees.py:4292: in predict
    return super(HybridGradientBoostingClassifier, self)._predict(
../../venv/lib/python3.9/site-packages/hana_ml/algorithms/pal/sqlgen.py:318: in function_with_sql_tracing
    return func(*args, **kwargs)
../../venv/lib/python3.9/site-packages/hana_ml/algorithms/pal/trees.py:3749: in _predict
    call_pal_auto(conn,
../../venv/lib/python3.9/site-packages/hana_ml/algorithms/pal/pal_base.py:338: in call_pal_auto
    if try_exec(cur, sql):
../../venv/lib/python3.9/site-packages/hana_ml/algorithms/pal/pal_base.py:309: in try_exec
    execute_logged(cur, sql, conn.sql_tracer) # SQLTRACE added sql_tracer
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 

cursor = <pyhdbcli.Cursor object at 0x11ea2fc90>
statement = 'DO (IN in_0 TABLE ("PassengerId" INT, "Pclass" INT, "Age" DOUBLE, "SibSp" INT, "Parch" INT, "Fare" DOUBLE, "Sex_femal... COLUMN TABLE "#PAL_HGBT_PREDICT_RESULT_TBL_69_0658FF16_9EF5_11EB_9E5C_6C4008A8C5E6" AS (SELECT * FROM :out_0);\nEND\n'
sql_tracer = <hana_ml.dataframe.SqlTrace object at 0x110671430>

    def execute_logged(cursor, statement, sql_tracer=None): # SQLTRACE added sql_tracer
        """
        Execute a SQL statement and log that we did it.
    
        Parameters
        ----------
        cursor : hdbcli.dbapi.Cursor
            Database cursor to execute the statement through.
        statement : str
            SQL statement to execute.
        """
        logger.info("Executing SQL: %s", statement)
    
        # SQLTRACE
        if sql_tracer:
            sql_tracer.trace_sql(statement)
    
>       cursor.execute(statement)
E       hdbcli.dbapi.Error: (256, 'sql processing error: "DEVELOPER"."(DO statement)": line 9 col 1 (at pos 592): search table error: _SYS_AFL.AFLPAL:HGBTPREDICT_ANY: [102] (range 1) TypeMismatch exception: Column \'ID\' does not exist')

../../venv/lib/python3.9/site-packages/hana_ml/ml_base.py:339: Error