def predict(self, conf):
"""
Run linear regression predict task
:param conf: configuration of algorithm and resource
:return:
"""
conf.set_int("angel.worker.matrix.transfer.request.timeout.ms", 60000)
conf.set(
AngelConf.ANGEL_TASK_USER_TASKCLASS,
'com.tencent.angel.ml.regression.linear.LinearRegPredictTask')
# Create an angel job client
client = AngelClientFactory.get(conf)
# Submit this application
client.startPSServer()
# Create a model
model = conf._jvm.com.tencent.angel.ml.regression.linear.LinearRegModel(
conf._jconf)
# Add the model meta to client
client.loadModel(model)
# Run user task
client.runTask(
'com.tencent.angel.ml.regression.linear.LinearRegPredictTask')
# Wait for completion,
# User task is set in AngelConf.ANGEL_TASK_USER_TASKCLASS
client.waitForCompletion()
# Stop
client.stop()
def train(self, conf):
"""
Training job to obtain a model
:param conf: configuration for parameter settings
"""
conf.set(AngelConf.ANGEL_TASK_USER_TASKCLASS,
'com.tencent.angel.ml.matrixfactorization.MFTrainTask')
# Create an angel job client
client = AngelClientFactory.get(conf)
# Submit this application
client.startPSServer()
# Create a MFModel
mfModel = conf._jvm.com.tencent.angel.ml.matrixfactorization.MFModel(
conf._jconf, None)
# Load model meta to client
client.loadModel(mfModel)
# Start
client.runTask('com.tencent.angel.ml.matrixfactorization.MFTrainTask')
# Run user task and wait for completion
# User task is set in AngelConf.ANGEL_TASK_USER_TASKCLASS
client.waitForCompletion()
# Save the trained model to HDFS
client.saveModel(mfModel)
# Stop
client.stop()
def fm_classification(self):
input_path = "./src/test/data/fm/a9a.train"
LOCAL_FS = LocalFileSystem.DEFAULT_FS
TMP_PATH = tempfile.gettempdir()
save_path = LOCAL_FS + TMP_PATH + "/model"
log_path = LOCAL_FS + TMP_PATH + "/LRlog"
# Set trainning data path
self.conf.set(AngelConf.ANGEL_TRAIN_DATA_PATH, input_path)
# Set save model path
self.conf.set(AngelConf.ANGEL_SAVE_MODEL_PATH, save_path)
# Set log path
self.conf.set(AngelConf.ANGEL_LOG_PATH, log_path)
# Set actionType train
self.conf.set(AngelConf.ANGEL_ACTION_TYPE, MLConf.ANGEL_ML_TRAIN)
# Set learnType
self.conf.set(MLConf.ML_FM_LEARN_TYPE, "c")
# Set feature number
self.conf.set(MLConf.ML_FEATURE_NUM, str(124))
runner = FMRunner()
runner.train(self.conf)
angel_client = AngelClientFactory.get(self.conf)
angel_client.stop()
def train(self, conf):
conf.set(AngelConf.ANGEL_TASK_USER_TASKCLASS, 'com.tencent.angel.ml.clustering.kmeans.KMeansTrainTask')
# Create an angel job client
client = AngelClientFactory.get(conf)
# Submit this application
client.startPSServer()
# Create a KMeans model
kmeansModel = conf._jvm.com.tencent.angel.ml.clustering.kmeans.KMeansModel(conf._jconf, None)
# Load model meta to client
client.loadModel(kmeansModel)
# Start
client.runTask('com.tencent.angel.ml.clustering.kmeans.KMeansTrainTask')
# Run user task and wait for completion,
# User task is set in AngelConf.ANGEL_TASK_USER_TASKCLASS
client.waitForCompletion()
# Save the trained model to HDFS
client.saveModel(kmeansModel)
# Stop
client.stop()
def predict(self, conf):
"""
Using a model to predict with unobserved samples
"""
conf.set_int("angel.worker.matrix.transfer.request.timeout.ms", 60000)
conf.set(AngelConf.ANGEL_TASK_USER_TASKCLASS, 'com.tencent.angel.ml.clustering.kmeans.KMeansPredictTask')
# Create an angel job client
client = AngelClientFactory.get(conf)
# Submit this application
client.startPSServer()
# Create KMeans model
model = conf._jvm.com.tencent.angel.ml.clustering.kmeans.KMeansModel(conf._jconf, None)
# Add the model meta to client
client.loadModel(model)
# Start
client.runTask('com.tencent.angel.ml.clustering.kmeans.KMeansPredictTask')
# Run user task and wait for completion,
# User task is set in AngelConf.ANGEL_TASK_USER_TASKCLASS
client.waitForCompletion()
# Stop
client.stop()
def inc_train(self, conf):
"""
Run incremental train task
:param conf: configuration of aalgorithm and resource
:return:
"""
conf['angel.worker.matrix.transfer.request.timeout.ms'] = 60000
conf[AngelConf.ANGEL_TASK_USER_TASKCLASS] = 'com.tencent.angel.ml.regression.linear.LinearRegTrainTask'
# Create an angel job client
client = AngelClientFactory.get(conf)
# Submit this application
client.startPSServer()
# Create a model
model = conf._jvm.com.tencent.angel.ml.regression.linear.LinearRegModel(conf._jconf)
# Load model meta to client
client.loadModel(model)
# Run user task
client.runTask('com.tencent.angel.ml.regression.linear.LinearRegTrainTask')
# Wait for completion,
# User task is set in AngelConf.ANGEL_TASK_USER_TASKCLASS
client.waitForCompletion()
# Save the incremental trained model to HDFS
client.saveModel(model)
# Stop
client.stop()
def train(self, conf):
conf.set(AngelConf.ANGEL_TASK_USER_TASKCLASS,
'com.tencent.angel.ml.clustering.kmeans.KMeansTrainTask')
# Create an angel job client
client = AngelClientFactory.get(conf)
# Submit this application
client.startPSServer()
# Create a KMeans model
kmeansModel = conf._jvm.com.tencent.angel.ml.clustering.kmeans.KMeansModel(
conf._jconf, None)
# Load model meta to client
client.loadModel(kmeansModel)
# Start
client.runTask(
'com.tencent.angel.ml.clustering.kmeans.KMeansTrainTask')
# Run user task and wait for completion,
# User task is set in AngelConf.ANGEL_TASK_USER_TASKCLASS
client.waitForCompletion()
# Save the trained model to HDFS
client.saveModel(kmeansModel)
# Stop
client.stop()
def fm_classification(self):
input_path = "./src/test/data/fm/a9a.train"
LOCAL_FS = LocalFileSystem.DEFAULT_FS
TMP_PATH = tempfile.gettempdir()
save_path = LOCAL_FS + TMP_PATH + "/model"
log_path = LOCAL_FS + TMP_PATH + "/LRlog"
# Set trainning data path
self.conf.set(AngelConf.ANGEL_TRAIN_DATA_PATH, input_path)
# Set save model path
self.conf.set(AngelConf.ANGEL_SAVE_MODEL_PATH, save_path)
# Set log path
self.conf.set(AngelConf.ANGEL_LOG_PATH, log_path)
# Set actionType train
self.conf.set(AngelConf.ANGEL_ACTION_TYPE, MLConf.ANGEL_ML_TRAIN)
# Set learnType
self.conf.set(MLConf.ML_FM_LEARN_TYPE, "c")
# Set feature number
self.conf.set(MLConf.ML_FEATURE_NUM, str(124))
runner = FMRunner()
runner.train(self.conf)
angel_client = AngelClientFactory.get(self.conf)
angel_client.stop()
def predict(self, conf):
"""
Using a model to predict with unobserved samples
"""
conf.set_int("angel.worker.matrix.transfer.request.timeout.ms", 60000)
conf.set(AngelConf.ANGEL_TASK_USER_TASKCLASS,
'com.tencent.angel.ml.clustering.kmeans.KMeansPredictTask')
# Create an angel job client
client = AngelClientFactory.get(conf)
# Submit this application
client.startPSServer()
# Create KMeans model
model = conf._jvm.com.tencent.angel.ml.clustering.kmeans.KMeansModel(
conf._jconf, None)
# Add the model meta to client
client.loadModel(model)
# Start
client.runTask(
'com.tencent.angel.ml.clustering.kmeans.KMeansPredictTask')
# Run user task and wait for completion,
# User task is set in AngelConf.ANGEL_TASK_USER_TASKCLASS
client.waitForCompletion()
# Stop
client.stop()
def train(self, conf):
client = AngelClientFactory.get(conf)
model = conf._jvm.com.tencent.angel.ml.factorizationmachines.FMModel(conf._jconf, None)
client.startPSServer()
client.loadModel(model)
client.runTask('com.tencent.angel.ml.factorizationmachines.FMTrainTask')
client.waitForCompletion()
client.saveModel(model)
client.stop()
def predict(self, dconf, model, class_name):
"""
Default predict method with standard predcit process. Don't try to override this Method
:param conf:
:return:
"""
jmap = Configuration._dict_to_jmap(dconf)
client = AngelClientFactory.get(jmap)
client.startPSServer()
client.loadModel(model)
client.runTask(class_name)
client.waitForCompletion()
client.stop()
def train(self, conf):
client = AngelClientFactory.get(conf)
model = conf._jvm.com.tencent.angel.ml.factorizationmachines.FMModel(
conf._jconf, None)
client.startPSServer()
client.loadModel(model)
client.runTask(
'com.tencent.angel.ml.factorizationmachines.FMTrainTask')
client.waitForCompletion()
client.saveModel(model)
client.stop()
def predict(self, dconf, model, class_name):
"""
Default predict method with standard predcit process. Don't try to override this Method
:param conf:
:return:
"""
jmap = Configuration._dict_to_jmap(dconf)
client = AngelClientFactory.get(jmap)
client.startPSServer()
client.loadModel(model)
client.runTask(class_name)
client.waitForCompletion()
client.stop()
def predict_on_local_cluster(self):
self.set_conf()
LOCAL_FS = LocalFileSystem.DEFAULT_FS
TMP_PATH = tempfile.gettempdir()
# Set trainning data path
self.conf[AngelConf.ANGEL_PREDICT_DATA_PATH] = input_path
# Set load model path
self.conf[AngelConf.ANGEL_LOAD_MODEL_PATH] = LOCAL_FS + TMP_PATH + "/model"
# Set predict result path
self.conf[AngelConf.ANGEL_PREDICT_PATH] = LOCAL_FS + TMP_PATH + "/predict"
# Set actionType prediction
self.conf[AngelConf.ANGEL_ACTION_TYPE] = MLConf.ANGEL_ML_PREDICT
runner = KMeansRunner()
runner.predict(self.conf)
angel_client = AngelClientFactory.get(self.conf)
angel_client.stop()
def train(self, conf, model, class_name):
"""
Default train method with standard training process. Don't Override this method
:param conf: Configuration for angel application
:param model: Data training model
:param train_class: train task class name
:return:
"""
# Get Java HashMap instance which converted from a python dict
jmap = conf.dict_to_jmap()
client = AngelClientFactory.get(jmap, conf)
client.startPSServer()
client.loadModel(model)
client.runTask(class_name)
client.waitForCompletion()
client.saveModel(model)
client.stop()
def train(self, conf, model, class_name):
"""
Default train method with standard training process. Don't Override this method
:param conf: Configuration for angel application
:param model: Data training model
:param train_class: train task class name
:return:
"""
# Get Java HashMap instance which converted from a python dict
jmap = conf.dict_to_jmap()
client = AngelClientFactory.get(jmap, conf)
client.startPSServer()
client.loadModel(model)
client.runTask(class_name)
client.waitForCompletion()
client.saveModel(model)
client.stop()
def train(self):
self.set_conf()
input_path = "data/exampledata/clusteringLocalExampleData/iris"
LOCAL_FS = LocalFileSystem.DEFAULT_FS
TMP_PATH = tempfile.gettempdir()
# Set trainning data path
self.conf.set(AngelConf.ANGEL_TRAIN_DATA_PATH, input_path)
# Set save model path
self.conf.set(AngelConf.ANGEL_SAVE_MODEL_PATH,
LOCAL_FS + TMP_PATH + "/model")
# Set log sava path
self.conf.set(AngelConf.ANGEL_LOG_PATH,
LOCAL_FS + TMP_PATH + "/kmeansLog/log")
# Set actionType train
self.conf.set(AngelConf.ANGEL_ACTION_TYPE, MLConf.ANGEL_ML_TRAIN)
runner = KMeansRunner()
runner.train(self.conf)
angel_client = AngelClientFactory.get(self.conf)
angel_client.stop()
def train_on_local_cluster(self):
self.set_conf()
input_path = "./src/test/data/fm/food_fm_libsvm"
LOCAL_FS = LocalFileSystem.DEFAULT_FS
TMP_PATH = tempfile.gettempdir()
save_path = LOCAL_FS + TMP_PATH + "/model"
log_path = LOCAL_FS + TMP_PATH + "/LRlog"
# Set trainning data path
self.conf.set(AngelConf.ANGEL_TRAIN_DATA_PATH, input_path)
# Set save model path
self.conf.set(AngelConf.ANGEL_SAVE_MODEL_PATH, save_path)
# Set log path
self.conf.set(AngelConf.ANGEL_LOG_PATH, log_path)
# Set actionType train
self.conf.set(AngelConf.ANGEL_ACTION_TYPE, MLConf.ANGEL_ML_TRAIN())
runner = FMRunner()
runner.train(self.conf)
angel_client = AngelClientFactory.get(self.conf)
angel_client.stop()
def train_on_local_cluster(self):
self.set_conf()
input_path = "./src/test/data/fm/food_fm_libsvm"
LOCAL_FS = LocalFileSystem.DEFAULT_FS
TMP_PATH = tempfile.gettempdir()
save_path = LOCAL_FS + TMP_PATH + "/model"
log_path = LOCAL_FS + TMP_PATH + "/LRlog"
# Set trainning data path
self.conf.set(AngelConf.ANGEL_TRAIN_DATA_PATH, input_path)
# Set save model path
self.conf.set(AngelConf.ANGEL_SAVE_MODEL_PATH, save_path)
# Set log path
self.conf.set(AngelConf.ANGEL_LOG_PATH, log_path)
# Set actionType train
self.conf.set(AngelConf.ANGEL_ACTION_TYPE, MLConf.ANGEL_ML_TRAIN())
runner = FMRunner()
runner.train(self.conf)
angel_client = AngelClientFactory.get(self.conf)
angel_client.stop()