陷入Python困境，在H2O的XGBoost上使用网格搜索

Question

通过Python编码在xgboost中使用Gridsearch时，我不会遇到问题。但今天当我尝试在H2O的xgboost（也使用H2O的Gridsearch函数）中使用Gridsearch时，它并没有让我通过。以下是代码：

xgboost_hyperparameters ={ 'max_depth' : range(2,10)
               ,'min_rows' : range(1,9)                                      #min_child_weight
               ,'sample_rate' : [i/10 for i in range (5,10)]}                 #subsample  
               ,'col_sample_rate_per_tree' : [i/10 for i in range (5,10)]}    #colsample_bytree


param = {'booster': 'gbtree', 
     'col_sample_rate': 1,                     #colsample_bylevel
     'keep_cross_validation_predictions': True,
     'learn_rate' : 0.1,         
     'max_abs_leafnode_pred': 1.0,        
     'nfolds': 10,
     'ntrees' : 24,
     'reg_alpha': 0.0,
     'reg_lambda': 5.0

    }

xgboost_grid1 = H2OGridSearch(model = H2OXGBoostEstimator(**param),
                         grid_id = 'xgboost_grid1',
                         hyper_params = xgboost_hyperparameters)

它是在Jupyter Notebook中传递的，但当我开始使用下面的代码训练模型时，报告错误：

xgboost_grid1.train(x=x, y=y,
           training_frame=train,
           validation_frame=valid)

错误的消息：

H2OResponseError                          Traceback (most recent call last)
<ipython-input-15-b1393b94399c> in <module>()
      1 xgboost_grid1.train(x=x, y=y,
      2                    training_frame=train,
----> 3                    validation_frame=valid)
      4 

~/anaconda3/lib/python3.6/site-packages/h2o/grid/grid_search.py in train(self, x, y, training_frame, offset_column, fold_column, weights_column, validation_frame, **params)
    206         x = list(xset)
    207         parms["x"] = x
--> 208         self.build_model(parms)
    209 
    210 

~/anaconda3/lib/python3.6/site-packages/h2o/grid/grid_search.py in build_model(self, algo_params)
    221         if is_auto_encoder and y is not None: raise ValueError("y should not be specified for autoencoder.")
    222         if not is_unsupervised and y is None: raise ValueError("Missing response")
--> 223         self._model_build(x, y, training_frame, validation_frame, algo_params)
    224 
    225 

~/anaconda3/lib/python3.6/site-packages/h2o/grid/grid_search.py in _model_build(self, x, y, tframe, vframe, kwargs)
    243         rest_ver = kwargs.pop("_rest_version") if "_rest_version" in kwargs else None
    244 
--> 245         grid = H2OJob(h2o.api("POST /99/Grid/%s" % algo, data=kwargs), job_type=(algo + " Grid Build"))
    246 
    247         if self._future:

~/anaconda3/lib/python3.6/site-packages/h2o/h2o.py in api(endpoint, data, json, filename, save_to)
    101     # type checks are performed in H2OConnection class
    102     _check_connection()
--> 103     return h2oconn.request(endpoint, data=data, json=json, filename=filename, save_to=save_to)
    104 
    105 

~/anaconda3/lib/python3.6/site-packages/h2o/backend/connection.py in request(self, endpoint, data, json, filename, save_to)
    400                                     auth=self._auth, verify=self._verify_ssl_cert, proxies=self._proxies)
    401             self._log_end_transaction(start_time, resp)
--> 402             return self._process_response(resp, save_to)
    403 
    404         except (requests.exceptions.ConnectionError, requests.exceptions.HTTPError) as e:

~/anaconda3/lib/python3.6/site-packages/h2o/backend/connection.py in _process_response(response, save_to)
    723         # Client errors (400 = "Bad Request", 404 = "Not Found", 412 = "Precondition Failed")
    724         if status_code in {400, 404, 412} and isinstance(data, (H2OErrorV3, H2OModelBuilderErrorV3)):
--> 725             raise H2OResponseError(data)
    726 
    727         # Server errors (notably 500 = "Server Error")

H2OResponseError: Server error water.exceptions.H2OIllegalArgumentException:
  Error: Can't parse the hyper_parameters dictionary; got error: com.google.gson.stream.MalformedJsonException: Expected ':' at line 1 column 28 path $. for raw value: {'max_depth': range(2, 10), 'min_rows': range(1, 9), 'sample_rate': [0.5, 0.6, 0.7, 0.8, 0.9]}
  Request: POST /99/Grid/xgboost
    data: {'hyper_parameters': "{'max_depth': range(2, 10), 'min_rows': range(1, 9), 'sample_rate': [0.5, 0.6, 0.7, 0.8, 0.9]}", 'booster': 'gbtree', 'col_sample_rate': '1', 'keep_cross_validation_predictions': 'True', 'learn_rate': '0.1', 'max_abs_leafnode_pred': '1.0', 'nfolds': '10', 'ntrees': '24', 'reg_alpha': '0.0', 'reg_lambda': '5.0', 'training_frame': 'py_4_sid_80f1', 'validation_frame': 'py_5_sid_80f1', 'response_column': 'label', 'grid_id': 'xgboost_grid1'}

需要帮助，因为我在H2O的网站和这里找到的文件很少。

Answer 1

你必须使用：

list(range(...))

代替：

range(...) --> 'max_depth' : list(range(2,10)) etc.