# Copyright (C) Microsoft Corporation. All rights reserved.
# Licensed under the MIT License. See LICENSE in project root for information.
import sys
if sys.version >= '3':
basestring = str
from pyspark import SparkContext, SQLContext
from pyspark.sql import DataFrame
from pyspark.ml.param.shared import *
from pyspark import keyword_only
from pyspark.ml.util import JavaMLReadable, JavaMLWritable
from mmlspark.core.serialize.java_params_patch import *
from pyspark.ml.wrapper import JavaTransformer, JavaEstimator, JavaModel
from pyspark.ml.common import inherit_doc
from mmlspark.core.schema.Utils import *
[docs]@inherit_doc
class RecognizeText(ComplexParamsMixin, JavaMLReadable, JavaMLWritable, JavaTransformer):
"""
Args:
backoffs (list): array of backoffs to use in the handler (default: [I@6ce2f9f8)
concurrency (int): max number of concurrent calls (default: 1)
concurrentTimeout (double): max number seconds to wait on futures if concurrency >= 1 (default: 100.0)
errorCol (str): column to hold http errors (default: [self.uid]_error)
imageBytes (object): bytestream of the image to use
imageUrl (object): the url of the image to use
maxPollingRetries (int): number of times to poll (default: 1000)
mode (object): If this parameter is set to 'Printed', printed text recognition is performed. If 'Handwritten' is specified, handwriting recognition is performed
outputCol (str): The name of the output column (default: [self.uid]_output)
pollingDelay (int): number of milliseconds to wait between polling (default: 300)
subscriptionKey (object): the API key to use
timeout (double): number of seconds to wait before closing the connection (default: 60.0)
url (str): Url of the service
"""
@keyword_only
def __init__(self, backoffs=[100,500,1000], concurrency=1, concurrentTimeout=100.0, errorCol=None, imageBytes=None, imageUrl=None, maxPollingRetries=1000, mode=None, outputCol=None, pollingDelay=300, subscriptionKey=None, timeout=60.0, url=None):
super(RecognizeText, self).__init__()
self._java_obj = self._new_java_obj("com.microsoft.ml.spark.cognitive.RecognizeText")
self.backoffs = Param(self, "backoffs", "backoffs: array of backoffs to use in the handler (default: [I@6ce2f9f8)")
self._setDefault(backoffs=[100,500,1000])
self.concurrency = Param(self, "concurrency", "concurrency: max number of concurrent calls (default: 1)")
self._setDefault(concurrency=1)
self.concurrentTimeout = Param(self, "concurrentTimeout", "concurrentTimeout: max number seconds to wait on futures if concurrency >= 1 (default: 100.0)")
self._setDefault(concurrentTimeout=100.0)
self.errorCol = Param(self, "errorCol", "errorCol: column to hold http errors (default: [self.uid]_error)")
self._setDefault(errorCol=self.uid + "_error")
self.imageBytes = Param(self, "imageBytes", "imageBytes: bytestream of the image to use")
self.imageUrl = Param(self, "imageUrl", "imageUrl: the url of the image to use")
self.maxPollingRetries = Param(self, "maxPollingRetries", "maxPollingRetries: number of times to poll (default: 1000)")
self._setDefault(maxPollingRetries=1000)
self.mode = Param(self, "mode", "mode: If this parameter is set to 'Printed', printed text recognition is performed. If 'Handwritten' is specified, handwriting recognition is performed")
self.outputCol = Param(self, "outputCol", "outputCol: The name of the output column (default: [self.uid]_output)")
self._setDefault(outputCol=self.uid + "_output")
self.pollingDelay = Param(self, "pollingDelay", "pollingDelay: number of milliseconds to wait between polling (default: 300)")
self._setDefault(pollingDelay=300)
self.subscriptionKey = Param(self, "subscriptionKey", "subscriptionKey: the API key to use")
self.timeout = Param(self, "timeout", "timeout: number of seconds to wait before closing the connection (default: 60.0)")
self._setDefault(timeout=60.0)
self.url = Param(self, "url", "url: Url of the service")
if hasattr(self, "_input_kwargs"):
kwargs = self._input_kwargs
else:
kwargs = self.__init__._input_kwargs
self.setParams(**kwargs)
[docs] @keyword_only
def setParams(self, backoffs=[100,500,1000], concurrency=1, concurrentTimeout=100.0, errorCol=None, imageBytes=None, imageUrl=None, maxPollingRetries=1000, mode=None, outputCol=None, pollingDelay=300, subscriptionKey=None, timeout=60.0, url=None):
"""
Set the (keyword only) parameters
Args:
backoffs (list): array of backoffs to use in the handler (default: [I@6ce2f9f8)
concurrency (int): max number of concurrent calls (default: 1)
concurrentTimeout (double): max number seconds to wait on futures if concurrency >= 1 (default: 100.0)
errorCol (str): column to hold http errors (default: [self.uid]_error)
imageBytes (object): bytestream of the image to use
imageUrl (object): the url of the image to use
maxPollingRetries (int): number of times to poll (default: 1000)
mode (object): If this parameter is set to 'Printed', printed text recognition is performed. If 'Handwritten' is specified, handwriting recognition is performed
outputCol (str): The name of the output column (default: [self.uid]_output)
pollingDelay (int): number of milliseconds to wait between polling (default: 300)
subscriptionKey (object): the API key to use
timeout (double): number of seconds to wait before closing the connection (default: 60.0)
url (str): Url of the service
"""
if hasattr(self, "_input_kwargs"):
kwargs = self._input_kwargs
else:
kwargs = self.__init__._input_kwargs
return self._set(**kwargs)
[docs] def getBackoffs(self):
"""
Returns:
list: array of backoffs to use in the handler (default: [I@6ce2f9f8)
"""
return self.getOrDefault(self.backoffs)
[docs] def getConcurrency(self):
"""
Returns:
int: max number of concurrent calls (default: 1)
"""
return self.getOrDefault(self.concurrency)
[docs] def getConcurrentTimeout(self):
"""
Returns:
double: max number seconds to wait on futures if concurrency >= 1 (default: 100.0)
"""
return self.getOrDefault(self.concurrentTimeout)
[docs] def getErrorCol(self):
"""
Returns:
str: column to hold http errors (default: [self.uid]_error)
"""
return self.getOrDefault(self.errorCol)
[docs] def getImageBytes(self):
"""
Returns:
object: bytestream of the image to use
"""
return self._cache.get("imageBytes", None)
[docs] def getImageUrl(self):
"""
Returns:
object: the url of the image to use
"""
return self._cache.get("imageUrl", None)
[docs] def getMaxPollingRetries(self):
"""
Returns:
int: number of times to poll (default: 1000)
"""
return self.getOrDefault(self.maxPollingRetries)
[docs] def getMode(self):
"""
Returns:
object: If this parameter is set to 'Printed', printed text recognition is performed. If 'Handwritten' is specified, handwriting recognition is performed
"""
return self._cache.get("mode", None)
[docs] def getOutputCol(self):
"""
Returns:
str: The name of the output column (default: [self.uid]_output)
"""
return self.getOrDefault(self.outputCol)
[docs] def getPollingDelay(self):
"""
Returns:
int: number of milliseconds to wait between polling (default: 300)
"""
return self.getOrDefault(self.pollingDelay)
[docs] def getSubscriptionKey(self):
"""
Returns:
object: the API key to use
"""
return self._cache.get("subscriptionKey", None)
[docs] def getTimeout(self):
"""
Returns:
double: number of seconds to wait before closing the connection (default: 60.0)
"""
return self.getOrDefault(self.timeout)
[docs] def getUrl(self):
"""
Returns:
str: Url of the service
"""
return self.getOrDefault(self.url)
[docs] def setBackoffs(self, value):
"""
Args:
backoffs: array of backoffs to use in the handler (default: [I@6ce2f9f8)
"""
self._set(backoffs=value)
return self
[docs] def setConcurrency(self, value):
"""
Args:
concurrency: max number of concurrent calls (default: 1)
"""
self._set(concurrency=value)
return self
[docs] def setConcurrentTimeout(self, value):
"""
Args:
concurrentTimeout: max number seconds to wait on futures if concurrency >= 1 (default: 100.0)
"""
self._set(concurrentTimeout=value)
return self
[docs] def setErrorCol(self, value):
"""
Args:
errorCol: column to hold http errors (default: [self.uid]_error)
"""
self._set(errorCol=value)
return self
[docs] def setImageBytes(self, value):
"""
Args:
imageBytes: bytestream of the image to use
"""
self._java_obj = self._java_obj.setImageBytes(value)
return self
[docs] def setImageBytesCol(self, value):
"""
Args:
imageBytes: bytestream of the image to use
"""
self._java_obj = self._java_obj.setImageBytesCol(value)
return self
[docs] def setImageUrl(self, value):
"""
Args:
imageUrl: the url of the image to use
"""
self._java_obj = self._java_obj.setImageUrl(value)
return self
[docs] def setImageUrlCol(self, value):
"""
Args:
imageUrl: the url of the image to use
"""
self._java_obj = self._java_obj.setImageUrlCol(value)
return self
[docs] def setMaxPollingRetries(self, value):
"""
Args:
maxPollingRetries: number of times to poll (default: 1000)
"""
self._set(maxPollingRetries=value)
return self
[docs] def setMode(self, value):
"""
Args:
mode: If this parameter is set to 'Printed', printed text recognition is performed. If 'Handwritten' is specified, handwriting recognition is performed
"""
self._java_obj = self._java_obj.setMode(value)
return self
[docs] def setModeCol(self, value):
"""
Args:
mode: If this parameter is set to 'Printed', printed text recognition is performed. If 'Handwritten' is specified, handwriting recognition is performed
"""
self._java_obj = self._java_obj.setModeCol(value)
return self
[docs] def setOutputCol(self, value):
"""
Args:
outputCol: The name of the output column (default: [self.uid]_output)
"""
self._set(outputCol=value)
return self
[docs] def setPollingDelay(self, value):
"""
Args:
pollingDelay: number of milliseconds to wait between polling (default: 300)
"""
self._set(pollingDelay=value)
return self
[docs] def setSubscriptionKey(self, value):
"""
Args:
subscriptionKey: the API key to use
"""
self._java_obj = self._java_obj.setSubscriptionKey(value)
return self
[docs] def setSubscriptionKeyCol(self, value):
"""
Args:
subscriptionKey: the API key to use
"""
self._java_obj = self._java_obj.setSubscriptionKeyCol(value)
return self
[docs] def setTimeout(self, value):
"""
Args:
timeout: number of seconds to wait before closing the connection (default: 60.0)
"""
self._set(timeout=value)
return self
[docs] def setUrl(self, value):
"""
Args:
url: Url of the service
"""
self._set(url=value)
return self
[docs] def setLocation(self, value):
self._java_obj = self._java_obj.setLocation(value)
return self
[docs] @classmethod
def read(cls):
""" Returns an MLReader instance for this class. """
return JavaMMLReader(cls)
[docs] @staticmethod
def getJavaPackage():
""" Returns package name String. """
return "com.microsoft.ml.spark.cognitive.RecognizeText"
@staticmethod
def _from_java(java_stage):
module_name=RecognizeText.__module__
module_name=module_name.rsplit(".", 1)[0] + ".RecognizeText"
return from_java(java_stage, module_name)