Merge branch 'master' of https://github.com/microsoft/nni

src/nni_manager/types/child-process-promise/index.d.ts

@@ -7,5 +7,5 @@ declare module 'child-process-promise' {
             stderr: string,
             message: string
         }
-    }    
+    }
 }
\ No newline at end of file

src/nni_manager/types/webhdfs/index.d.ts

+declare module 'webhdfs' {
+    export function createClient(arg: any): any;
+}
\ No newline at end of file

src/sdk/pynni/nni/__main__.py

@@ -154,7 +154,7 @@ def main():
         assessor = None
         if args.tuner_class_name in ModuleName:
             tuner = create_builtin_class_instance(
-                args.tuner_class_name, 
+                args.tuner_class_name,
                 args.tuner_args)
         else:
             tuner = create_customized_class_instance(

src/sdk/pynni/nni/bohb_advisor/bohb_advisor.py

@@ -81,7 +81,7 @@ def create_bracket_parameter_id(brackets_id, brackets_curr_decay, increased_id=-
 
 class Bracket():
     """
-    A bracket in BOHB, all the information of a bracket is managed by 
+    A bracket in BOHB, all the information of a bracket is managed by
     an instance of this class.
 
     Parameters
@@ -251,7 +251,7 @@ class BOHB(MsgDispatcherBase):
     BOHB performs robust and efficient hyperparameter optimization
     at scale by combining the speed of Hyperband searches with the
     guidance and guarantees of convergence of Bayesian Optimization.
-    Instead of sampling new configurations at random, BOHB uses 
+    Instead of sampling new configurations at random, BOHB uses
     kernel density estimators to select promising candidates.
 
     Parameters
@@ -335,7 +335,7 @@ class BOHB(MsgDispatcherBase):
         pass
 
     def handle_initialize(self, data):
-        """Initialize Tuner, including creating Bayesian optimization-based parametric models 
+        """Initialize Tuner, including creating Bayesian optimization-based parametric models
         and search space formations
 
         Parameters
@@ -403,7 +403,7 @@ class BOHB(MsgDispatcherBase):
 
         If this function is called, Command will be sent by BOHB:
         a. If there is a parameter need to run, will return "NewTrialJob" with a dict:
-        { 
+        {
             'parameter_id': id of new hyperparameter
             'parameter_source': 'algorithm'
             'parameters': value of new hyperparameter
@@ -458,30 +458,30 @@ class BOHB(MsgDispatcherBase):
                     var, lower=search_space[var]["_value"][0], upper=search_space[var]["_value"][1]))
             elif _type == 'quniform':
                 cs.add_hyperparameter(CSH.UniformFloatHyperparameter(
-                    var, lower=search_space[var]["_value"][0], upper=search_space[var]["_value"][1], 
+                    var, lower=search_space[var]["_value"][0], upper=search_space[var]["_value"][1],
                     q=search_space[var]["_value"][2]))
             elif _type == 'loguniform':
                 cs.add_hyperparameter(CSH.UniformFloatHyperparameter(
-                    var, lower=search_space[var]["_value"][0], upper=search_space[var]["_value"][1], 
+                    var, lower=search_space[var]["_value"][0], upper=search_space[var]["_value"][1],
                     log=True))
             elif _type == 'qloguniform':
                 cs.add_hyperparameter(CSH.UniformFloatHyperparameter(
-                    var, lower=search_space[var]["_value"][0], upper=search_space[var]["_value"][1], 
+                    var, lower=search_space[var]["_value"][0], upper=search_space[var]["_value"][1],
                     q=search_space[var]["_value"][2], log=True))
             elif _type == 'normal':
                 cs.add_hyperparameter(CSH.NormalFloatHyperparameter(
                     var, mu=search_space[var]["_value"][1], sigma=search_space[var]["_value"][2]))
             elif _type == 'qnormal':
                 cs.add_hyperparameter(CSH.NormalFloatHyperparameter(
-                    var, mu=search_space[var]["_value"][1], sigma=search_space[var]["_value"][2], 
+                    var, mu=search_space[var]["_value"][1], sigma=search_space[var]["_value"][2],
                     q=search_space[var]["_value"][3]))
             elif _type == 'lognormal':
                 cs.add_hyperparameter(CSH.NormalFloatHyperparameter(
-                    var, mu=search_space[var]["_value"][1], sigma=search_space[var]["_value"][2], 
+                    var, mu=search_space[var]["_value"][1], sigma=search_space[var]["_value"][2],
                     log=True))
             elif _type == 'qlognormal':
                 cs.add_hyperparameter(CSH.NormalFloatHyperparameter(
-                    var, mu=search_space[var]["_value"][1], sigma=search_space[var]["_value"][2], 
+                    var, mu=search_space[var]["_value"][1], sigma=search_space[var]["_value"][2],
                     q=search_space[var]["_value"][3], log=True))
             else:
                 raise ValueError(
@@ -553,7 +553,7 @@ class BOHB(MsgDispatcherBase):
             self.brackets[s].set_config_perf(
                 int(i), data['parameter_id'], sys.maxsize, value)
             self.completed_hyper_configs.append(data)
-       
+
             _parameters = self.parameters[data['parameter_id']]
             _parameters.pop(_KEY)
             # update BO with loss, max_s budget, hyperparameters

src/sdk/pynni/nni/bohb_advisor/config_generator.py

@@ -117,7 +117,7 @@ class CG_BOHB(object):
         seperated by budget. This function sample a configuration from
         largest budget. Firstly we sample "num_samples" configurations,
         then prefer one with the largest l(x)/g(x).
-        
+
         Parameters:
         -----------
         info_dict: dict

src/sdk/pynni/nni/common.py

@@ -34,7 +34,7 @@ log_level_map = {
 }
 
 _time_format = '%m/%d/%Y, %I:%M:%S %p'
-    
+
 class _LoggerFileWrapper(TextIOBase):
     def __init__(self, logger_file):
         self.file = logger_file

src/sdk/pynni/nni/curvefitting_assessor/curvefitting_assessor.py

@@ -67,7 +67,7 @@ class CurvefittingAssessor(Assessor):
 
     def trial_end(self, trial_job_id, success):
         """update the best performance of completed trial job
-        
+
         Parameters
         ----------
         trial_job_id: int
@@ -112,7 +112,7 @@ class CurvefittingAssessor(Assessor):
         curr_step = len(trial_history)
         if curr_step < self.start_step:
             return AssessResult.Good
-        
+
         if trial_job_id in self.last_judgment_num.keys() and curr_step - self.last_judgment_num[trial_job_id] < self.gap:
             return AssessResult.Good
         self.last_judgment_num[trial_job_id] = curr_step

src/sdk/pynni/nni/curvefitting_assessor/curvefunctions.py

@@ -26,7 +26,7 @@ curve_combination_models = ['vap', 'pow3', 'linear', 'logx_linear', 'dr_hill_zer
 
 def vap(x, a, b, c):
     """Vapor pressure model
-    
+
     Parameters
     ----------
     x: int
@@ -109,7 +109,7 @@ model_para_num['logx_linear'] = 2
 
 def dr_hill_zero_background(x, theta, eta, kappa):
     """dr hill zero background
- 
+
     Parameters
     ----------
     x: int
@@ -261,7 +261,7 @@ model_para_num['weibull'] = 4
 
 def janoschek(x, a, beta, k, delta):
     """http://www.pisces-conservation.com/growthhelp/janoschek.htm
-    
+
     Parameters
     ----------
     x: int

src/sdk/pynni/nni/curvefitting_assessor/model_factory.py

@@ -35,7 +35,7 @@ logger = logging.getLogger('curvefitting_Assessor')
 
 class CurveModel(object):
     """Build a Curve Model to predict the performance
-    
+
     Algorithm: https://github.com/Microsoft/nni/blob/master/src/sdk/pynni/nni/curvefitting_assessor/README.md
 
     Parameters
@@ -53,7 +53,7 @@ class CurveModel(object):
 
     def fit_theta(self):
         """use least squares to fit all default curves parameter seperately
-        
+
         Returns
         -------
         None
@@ -87,7 +87,7 @@ class CurveModel(object):
 
     def filter_curve(self):
         """filter the poor performing curve
-        
+
         Returns
         -------
         None
@@ -117,7 +117,7 @@ class CurveModel(object):
 
     def predict_y(self, model, pos):
         """return the predict y of 'model' when epoch = pos
-        
+
         Parameters
         ----------
         model: string
@@ -162,7 +162,7 @@ class CurveModel(object):
 
     def normalize_weight(self, samples):
         """normalize weight
-        
+
         Parameters
         ----------
         samples: list
@@ -184,7 +184,7 @@ class CurveModel(object):
 
     def sigma_sq(self, sample):
         """returns the value of sigma square, given the weight's sample
-        
+
         Parameters
         ----------
         sample: list
@@ -203,7 +203,7 @@ class CurveModel(object):
 
     def normal_distribution(self, pos, sample):
         """returns the value of normal distribution, given the weight's sample and target position
-        
+
         Parameters
         ----------
         pos: int
@@ -227,7 +227,7 @@ class CurveModel(object):
         ----------
         sample: list
             sample is a (1 * NUM_OF_FUNCTIONS) matrix, representing{w1, w2, ... wk}
-        
+
         Returns
         -------
         float
@@ -241,13 +241,13 @@ class CurveModel(object):
 
     def prior(self, samples):
         """priori distribution
- 
+
         Parameters
         ----------
         samples: list
             a collection of sample, it's a (NUM_OF_INSTANCE * NUM_OF_FUNCTIONS) matrix,
             representing{{w11, w12, ..., w1k}, {w21, w22, ... w2k}, ...{wk1, wk2,..., wkk}}
-        
+
         Returns
         -------
         float
@@ -264,13 +264,13 @@ class CurveModel(object):
 
     def target_distribution(self, samples):
         """posterior probability
-        
+
         Parameters
         ----------
         samples: list
             a collection of sample, it's a (NUM_OF_INSTANCE * NUM_OF_FUNCTIONS) matrix,
             representing{{w11, w12, ..., w1k}, {w21, w22, ... w2k}, ...{wk1, wk2,..., wkk}}
-        
+
         Returns
         -------
         float
@@ -319,7 +319,7 @@ class CurveModel(object):
 
     def predict(self, trial_history):
         """predict the value of target position
-        
+
         Parameters
         ----------
         trial_history: list

src/sdk/pynni/nni/evolution_tuner/evolution_tuner.py

@@ -167,7 +167,7 @@ class EvolutionTuner(Tuner):
         self.space = None
 
     def update_search_space(self, search_space):
-        """Update search space. 
+        """Update search space.
         Search_space contains the information that user pre-defined.
 
         Parameters
@@ -194,7 +194,7 @@ class EvolutionTuner(Tuner):
         Parameters
         ----------
         parameter_id : int
-    
+
         Returns
         -------
         config : dict

src/sdk/pynni/nni/hyperband_advisor/hyperband_advisor.py

@@ -43,7 +43,7 @@ _epsilon = 1e-6
 
 def create_parameter_id():
     """Create an id
-    
+
     Returns
     -------
     int
@@ -55,7 +55,7 @@ def create_parameter_id():
 
 def create_bracket_parameter_id(brackets_id, brackets_curr_decay, increased_id=-1):
     """Create a full id for a specific bracket's hyperparameter configuration
-    
+
     Parameters
     ----------
     brackets_id: int
@@ -79,7 +79,7 @@ def create_bracket_parameter_id(brackets_id, brackets_curr_decay, increased_id=-
 
 def json2parameter(ss_spec, random_state):
     """Randomly generate values for hyperparameters from hyperparameter space i.e., x.
-    
+
     Parameters
     ----------
     ss_spec:
@@ -116,7 +116,7 @@ def json2parameter(ss_spec, random_state):
 
 class Bracket():
     """A bracket in Hyperband, all the information of a bracket is managed by an instance of this class
-    
+
     Parameters
     ----------
     s: int
@@ -132,7 +132,7 @@ class Bracket():
     optimize_mode: str
         optimize mode, 'maximize' or 'minimize'
     """
-    
+
     def __init__(self, s, s_max, eta, R, optimize_mode):
         self.bracket_id = s
         self.s_max = s_max
@@ -163,7 +163,7 @@ class Bracket():
 
     def set_config_perf(self, i, parameter_id, seq, value):
         """update trial's latest result with its sequence number, e.g., epoch number or batch number
-        
+
         Parameters
         ----------
         i: int
@@ -184,7 +184,7 @@ class Bracket():
                 self.configs_perf[i][parameter_id] = [seq, value]
         else:
             self.configs_perf[i][parameter_id] = [seq, value]
-            
+
 
     def inform_trial_end(self, i):
         """If the trial is finished and the corresponding round (i.e., i) has all its trials finished,
@@ -230,7 +230,7 @@ class Bracket():
         ----------
         num: int
             the number of hyperparameter configurations
-        
+
         Returns
         -------
         list
@@ -350,7 +350,7 @@ class Hyperband(MsgDispatcherBase):
 
     def handle_update_search_space(self, data):
         """data: JSON object, which is search space
-        
+
         Parameters
         ----------
         data: int
@@ -392,9 +392,9 @@ class Hyperband(MsgDispatcherBase):
         """
         Parameters
         ----------
-        data: 
+        data:
             it is an object which has keys 'parameter_id', 'value', 'trial_job_id', 'type', 'sequence'.
-        
+
         Raises
         ------
         ValueError

src/sdk/pynni/nni/medianstop_assessor/medianstop_assessor.py

@@ -21,10 +21,10 @@ from nni.assessor import Assessor, AssessResult
 logger = logging.getLogger('medianstop_Assessor')
 
 class MedianstopAssessor(Assessor):
-    """MedianstopAssessor is The median stopping rule stops a pending trial X at step S 
-    if the trial’s best objective value by step S is strictly worse than the median value 
+    """MedianstopAssessor is The median stopping rule stops a pending trial X at step S
+    if the trial’s best objective value by step S is strictly worse than the median value
     of the running averages of all completed trials’ objectives reported up to step S
-    
+
     Parameters
     ----------
     optimize_mode: str
@@ -60,7 +60,7 @@ class MedianstopAssessor(Assessor):
 
     def trial_end(self, trial_job_id, success):
         """trial_end
-        
+
         Parameters
         ----------
         trial_job_id: int
@@ -83,7 +83,7 @@ class MedianstopAssessor(Assessor):
 
     def assess_trial(self, trial_job_id, trial_history):
         """assess_trial
-        
+
         Parameters
         ----------
         trial_job_id: int

src/sdk/pynni/nni/metis_tuner/lib_acquisition_function.py

@@ -27,7 +27,7 @@ from scipy.optimize import minimize
 import nni.metis_tuner.lib_data as lib_data
 
 
-def next_hyperparameter_expected_improvement(fun_prediction,   
+def next_hyperparameter_expected_improvement(fun_prediction,
                                              fun_prediction_args,
                                              x_bounds, x_types,
                                              samples_y_aggregation,

src/sdk/pynni/nni/networkmorphism_tuner/networkmorphism_tuner.py

@@ -69,7 +69,7 @@ class NetworkMorphismTuner(Tuner):
         optimize_mode : str
             optimize mode "minimize" or "maximize" (default: {"minimize"})
         path : str
-            default mode path to save the model file (default: {"model_path"})  
+            default mode path to save the model file (default: {"model_path"})
         verbose : bool
             verbose to print the log (default: {True})
         beta : float
@@ -154,7 +154,7 @@ class NetworkMorphismTuner(Tuner):
 
     def receive_trial_result(self, parameter_id, parameters, value):
         """ Record an observation of the objective function.
-    
+
         Parameters
         ----------
         parameter_id : int
@@ -267,7 +267,7 @@ class NetworkMorphismTuner(Tuner):
         ----------
         model_id : int
             model index
-        
+
         Returns
         -------
         load_model : Graph
@@ -297,7 +297,7 @@ class NetworkMorphismTuner(Tuner):
         ----------
         model_id : int
             model index
-        
+
         Returns
         -------
         float

src/sdk/pynni/nni/platform/local.py

@@ -67,7 +67,7 @@ def get_next_parameter():
             params_file_name = 'parameter.cfg'
         else:
             raise AssertionError('_param_index value ({}) should >=0'.format(_param_index))
-    
+
     params_filepath = os.path.join(_sysdir, params_file_name)
     if not os.path.isfile(params_filepath):
         request_next_parameter()
@@ -81,11 +81,11 @@ def get_next_parameter():
 def send_metric(string):
     if _nni_platform != 'local':
         data = (string).encode('utf8')
-        assert len(data) < 1000000, 'Metric too long'    
+        assert len(data) < 1000000, 'Metric too long'
         print('NNISDK_ME%s' % (data), flush=True)
     else:
         data = (string + '\n').encode('utf8')
-        assert len(data) < 1000000, 'Metric too long'    
+        assert len(data) < 1000000, 'Metric too long'
         _metric_file.write(b'ME%06d%b' % (len(data), data))
         _metric_file.flush()
         if sys.platform == "win32":

src/sdk/pynni/nni/smac_tuner/convert_ss_to_scenario.py

@@ -24,12 +24,12 @@ import numpy as np
 
 def get_json_content(file_path):
     """Load json file content
-    
+
     Parameters
     ----------
     file_path:
         path to the file
-    
+
     Raises
     ------
     TypeError
@@ -43,9 +43,9 @@ def get_json_content(file_path):
         return None
 
 def generate_pcs(nni_search_space_content):
-    """Generate the Parameter Configuration Space (PCS) which defines the 
+    """Generate the Parameter Configuration Space (PCS) which defines the
     legal ranges of the parameters to be optimized and their default values.
-    
+
     Generally, the format is:
     # parameter_name categorical {value_1, ..., value_N} [default value]
     # parameter_name ordinal {value_1, ..., value_N} [default value]
@@ -53,14 +53,14 @@ def generate_pcs(nni_search_space_content):
     # parameter_name integer [min_value, max_value] [default value] log
     # parameter_name real [min_value, max_value] [default value]
     # parameter_name real [min_value, max_value] [default value] log
-    
+
     Reference: https://automl.github.io/SMAC3/stable/options.html
 
     Parameters
     ----------
     nni_search_space_content: search_space
         The search space in this experiment in nni
-    
+
     Returns
     -------
     Parameter Configuration Space (PCS)
@@ -81,8 +81,8 @@ def generate_pcs(nni_search_space_content):
                         if search_space[key]['_type'] == 'choice':
                             choice_len = len(search_space[key]['_value'])
                             pcs_fd.write('%s categorical {%s} [%s]\n' % (
-                                key, 
-                                json.dumps(list(range(choice_len)))[1:-1], 
+                                key,
+                                json.dumps(list(range(choice_len)))[1:-1],
                                 json.dumps(0)))
                             if key in categorical_dict:
                                 raise RuntimeError('%s has already existed, please make sure search space has no duplicate key.' % key)
@@ -90,19 +90,19 @@ def generate_pcs(nni_search_space_content):
                         elif search_space[key]['_type'] == 'randint':
                             # TODO: support lower bound in randint
                             pcs_fd.write('%s integer [0, %d] [%d]\n' % (
-                                key, 
-                                search_space[key]['_value'][0], 
+                                key,
+                                search_space[key]['_value'][0],
                                 search_space[key]['_value'][0]))
                         elif search_space[key]['_type'] == 'uniform':
                             pcs_fd.write('%s real %s [%s]\n' % (
-                                key, 
+                                key,
                                 json.dumps(search_space[key]['_value']),
                                 json.dumps(search_space[key]['_value'][0])))
                         elif search_space[key]['_type'] == 'loguniform':
                             # use np.round here to ensure that the rounded defaut value is in the range, which will be rounded in configure_space package
                             search_space[key]['_value'] = list(np.round(np.log(search_space[key]['_value']), 10))
                             pcs_fd.write('%s real %s [%s]\n' % (
-                                key, 
+                                key,
                                 json.dumps(search_space[key]['_value']),
                                 json.dumps(search_space[key]['_value'][0])))
                         elif search_space[key]['_type'] == 'quniform' \
@@ -122,9 +122,9 @@ def generate_pcs(nni_search_space_content):
     return None
 
 def generate_scenario(ss_content):
-    """Generate the scenario. The scenario-object (smac.scenario.scenario.Scenario) is used to configure SMAC and 
+    """Generate the scenario. The scenario-object (smac.scenario.scenario.Scenario) is used to configure SMAC and
     can be constructed either by providing an actual scenario-object, or by specifing the options in a scenario file.
-    
+
     Reference: https://automl.github.io/SMAC3/stable/options.html
 
     The format of the scenario file is one option per line:
@@ -135,7 +135,7 @@ def generate_scenario(ss_content):
     Parameters
     ----------
     abort_on_first_run_crash: bool
-        If true, SMAC will abort if the first run of the target algorithm crashes. Default: True, 
+        If true, SMAC will abort if the first run of the target algorithm crashes. Default: True,
         because trials reported to nni tuner would always in success state
     algo: function
         Specifies the target algorithm call that SMAC will optimize. Interpreted as a bash-command.

src/sdk/pynni/nni/smac_tuner/smac_tuner.py

@@ -64,7 +64,7 @@ class SMACTuner(Tuner):
 
     def _main_cli(self):
         """Main function of SMAC for CLI interface
-        
+
         Returns
         -------
         instance
@@ -153,7 +153,7 @@ class SMACTuner(Tuner):
 
     def receive_trial_result(self, parameter_id, parameters, value):
         """receive_trial_result
-       
+
         Parameters
         ----------
         parameter_id: int
@@ -162,7 +162,7 @@ class SMACTuner(Tuner):
             parameters
         value:
             value
-        
+
         Raises
         ------
         RuntimeError
@@ -185,7 +185,7 @@ class SMACTuner(Tuner):
         Also, we convert categorical:
         categorical values in search space are changed to list of numbers before,
         those original values will be changed back in this function
-        
+
         Parameters
         ----------
         challenger_dict: dict
@@ -211,12 +211,12 @@ class SMACTuner(Tuner):
 
     def generate_parameters(self, parameter_id):
         """generate one instance of hyperparameters
-        
+
         Parameters
         ----------
         parameter_id: int
             parameter id
-        
+
         Returns
         -------
         list
@@ -234,12 +234,12 @@ class SMACTuner(Tuner):
 
     def generate_multiple_parameters(self, parameter_id_list):
         """generate mutiple instances of hyperparameters
-        
+
         Parameters
         ----------
         parameter_id_list: list
             list of parameter id
-        
+
         Returns
         -------
         list

src/sdk/pynni/tests/test_multi_phase_tuner.py

@@ -32,7 +32,7 @@ from nni.multi_phase.multi_phase_dispatcher import MultiPhaseMsgDispatcher
 from unittest import TestCase, main
 
 class NaiveMultiPhaseTuner(MultiPhaseTuner):
-    ''' 
+    '''
     supports only choices
     '''
     def __init__(self):

src/sdk/pynni/tests/test_trial.py

@@ -40,7 +40,7 @@ class TrialTestCase(TestCase):
 
     def test_get_sequence_id(self):
         self.assertEqual(nni.get_sequence_id(), 0)
-        
+
     def test_report_intermediate_result(self):
         nni.report_intermediate_result(123)
         self.assertEqual(test_platform.get_last_metric(), {

src/webui/package.json

@@ -9,11 +9,12 @@
     "copy-to-clipboard": "^3.0.8",
     "echarts": "^4.1.0",
     "echarts-for-react": "^2.0.14",
-    "react": "^16.4.2",
-    "react-dom": "^16.4.2",
+    "react": "^16.7.0-alpha.2",
+    "react-dom": "^16.7.0-alpha.2",
     "react-json-tree": "^0.11.0",
     "react-monaco-editor": "^0.22.0",
     "react-router": "3.2.1",
+    "react-responsive": "^7.0.0",
     "react-scripts-ts-antd": "2.17.0"
   },
   "scripts": {
@@ -28,7 +29,8 @@
     "@types/react": "^16.4.17",
     "@types/react-dom": "^16.0.7",
     "@types/react-json-tree": "^0.6.8",
+    "@types/react-responsive": "^3.0.3",
     "@types/react-router": "3.0.15",
     "typescript": "^3.0.1"
   }
-}
+}
\ No newline at end of file