class Parameters:
inputCol = Parameter(str,
'tokens',
'Input column',
input_column=True,
input_dtype=Parameter.T_ARRAY_STRING)
outputCol = Parameter(str, 'tf', 'Output column', output_column=True)
numFeatures = Parameter(int, 1 << 18, 'Number of features')
binary = Parameter(bool, False, 'Binary')
class Parameters:
inputCol = Parameter(str,
'text',
'Input column',
input_column=True,
input_dtype=Parameter.T_STRING)
outputCol = Parameter(str,
'tokens',
'Output column',
output_column=True)
class Parameters:
inputCols = Parameter(str,
'vectors',
'Input columns',
input_column=True,
input_dtype=Parameter.T_VECTOR)
outputCol = Parameter(str,
'assembled_vector',
'Output1 column',
output_column=True)
class Parameters:
inputCol = Parameter(str,
'category',
'Input column',
input_column=True,
input_dtype=Parameter.T_STRING)
outputCol = Parameter(str,
'category_index',
'Output column',
output_column=True)
class Parameters:
n = Parameter(int, 2, 'N')
inputCol = Parameter(str,
'text',
'Input column',
input_column=True,
input_dtype=Parameter.T_ARRAY_STRING)
outputCol = Parameter(str,
'tokens',
'Output column',
output_column=True)
class Parameters:
inputCol = Parameter(str,
'tf',
'Input column',
input_column=True,
input_dtype=Parameter.T_VECTOR)
outputCol = Parameter(str,
'tfidf',
'Output column',
output_column=True)
minDocFreq = Parameter(int, 0, 'Minimum document frequency')
class Parameters:
inputCol = Parameter(str,
'text',
'Input column',
input_column=True,
input_dtype=Parameter.T_ARRAY_STRING)
outputCol = Parameter(str,
'tokens',
'Output column',
output_column=True)
# 'stopWords': Parameter(list, None, 'Stopwords list')
caseSensitive = Parameter(bool, False, 'Case sensitive')
class Parameters:
featuresCol = Parameter(str,
'features',
'Feature column',
input_column=True)
labelCol = Parameter(str, 'label', 'Label column', input_column=True)
# predictionCol = Parameter(str, 'prediction', 'Prediction column')
# probabilityCol = Parameter(str, 'probability', 'Probability column')
# rawPredictionCol = Parameter(str, 'rawPrediction', 'Raw probability column')
weightCol = Parameter(str, 'weight', 'Weight Column')
smoothing = Parameter(float, 1.0, 'Smoothing')
class Parameters:
cutoff = Parameter(int, 50, 'Top scores to display:')
inputCol = Parameter(str,
'tokens',
'Input column',
input_column=True,
input_dtype=Parameter.T_STRING)
sortingType = Parameter(
str,
'Term_Freq',
'Choose a sorting type',
items=['Term Frequency', 'Chi', 't-score', 'Non perfect Chi'])
class Parameters:
featuresCol = Parameter(str,
'features',
'Feature column',
input_column=True)
labelCol = Parameter(str, 'label', 'Label column', input_column=True)
# weightCol = Parameter(str, 'weight', 'Weight Column')
maxIter = Parameter(int, 100, 'Maximal iteration')
regParam = Parameter(float, 0.0, 'Regression Parameter')
elasticNetParam = Parameter(float, 0.0, 'Elastic Net Parameter')
tol = Parameter(float, 0.000001, 'tol')
fitIntercept = Parameter(bool, True, 'Fit intercept')
standardization = Parameter(bool, False, 'Standardization')
solver = Parameter(str, 'auto', 'Solver')
aggregationDepth = Parameter(int, 2, 'Aggregation depth')
class Parameters:
inputCol = Parameter(str,
'tokens',
'Input column',
input_column=True,
input_dtype=Parameter.T_ARRAY_STRING)
outputCol = Parameter(str,
'vector',
'Output1 column',
output_column=True)
vectorSize = Parameter(int, 100, 'Vector size')
minCount = Parameter(int, 5, 'Minimum count')
numPartitions = Parameter(int, 1, 'Number of partitions ')
stepSize = Parameter(float, 0.025, 'Step size')
maxIter = Parameter(int, 1, 'Maximum Iteration')
seed = Parameter(int, None, 'Seed')
windowSize = Parameter(int, 5, 'Window size')
maxSentenceLength = Parameter(int, 1000, 'Maximum sentence length')
class Parameters:
inputCol = Parameter(str, 'tokens', 'Input column', input_column=True, input_dtype=Parameter.T_ARRAY_STRING)
outputCol = Parameter(str, 'vector', 'Output1 column', output_column=True)
minTF = Parameter(float, 1.0, 'Minimum term frequency'),
minDF = Parameter(float, 1.0, 'Minimum document frequency')
vocabSize = Parameter(int, 1 << 18, 'Vocabulary size')
binary = Parameter(bool, False, 'Binary')
class Parameters:
inputCol = Parameter(str,
'text',
'Input column',
input_column=True,
input_dtype=Parameter.T_STRING)
outputCol = Parameter(str,
'tokens',
'Output column',
output_column=True)
minTokenLength = Parameter(int, 1, 'Minimum token length')
gaps = Parameter(bool, True, 'Gaps?')
pattern = Parameter(str, '\\s+', 'Pattern')
toLowercase = Parameter(bool, True, 'Convert to lower case?')
class Parameters:
inputCols = Parameter(list, [], 'Input columns', input_column=True, input_multiple=True)
outputCol = Parameter(str, 'features', 'Output column', output_column=True)
class Parameters:
featuresCol = Parameter(str, 'features', 'Features column', input_column=True)
labelCol = Parameter(str, 'label', 'Label column', input_column=True)
# predictionCol = Parameter(str, 'prediction', 'Prediction column')
# probabilityCol = Parameter(str, "probability", 'Probability column')
# rawPredictionCol = Parameter(str, 'rawPrediction', 'Raw probability column')
weightCol = Parameter(str, 'weight', 'Weight column')
maxIter = Parameter(int, 100, 'Maximum iteration')
regParam = Parameter(float, 0.0, 'Regression Parameter')
elasticNetParam = Parameter(float, 0.0, 'Elastic Net Parameter')
tol = Parameter(float, 0.000001, 'tol')
fitIntercept = Parameter(bool, True, 'Fit intercept')
threshold = Parameter(float, 0.5, 'Threshold')
# thresholds = Parameter(list, None, 'Thresholds') # list[float]
standardization = Parameter(bool, True, 'Standardization')
aggregationDepth = Parameter(int, 2, 'Aggregation depth')
family = Parameter(str, 'auto', 'Family')
class Parameters:
inputCol = Parameter(str, 'input', 'Input column', input_column=True)
class Parameters:
featuresCol = Parameter(str,
'features',
'Features column',
input_column=True)
labelCol = Parameter(str, 'label', 'Label column', input_column=True)
predictionCol = Parameter(str,
'prediction',
'Prediction column',
input_column=True)
probabilityCol = Parameter(str,
"probability",
'Probability column',
input_column=True)
rawPredictionCol = Parameter(str,
'rawPrediction',
'Raw prediction column',
input_column=True)
maxDepth = Parameter(int, 5, 'Maximal depth')
maxBins = Parameter(int, 32, 'Maximal bins')
minInstancesPerNode = Parameter(int, 1, 'Minimum instance per node')
minInfoGain = Parameter(float, 0.0, 'Minimum Information gain')
maxMemoryInMB = Parameter(int, 256, 'Maximal Memory (MB)')
cacheNodeIds = Parameter(bool, False, 'Cache node ids')
checkpointInterval = Parameter(int, 10, 'Checkpoint interval')
impurity = Parameter(str, 'gini', 'Impurity')
seed = Parameter(int, None, 'Seed')
class Parameters:
select_ratio = Parameter(int, 1,
'Selection ratio for unlinked reports')
class Parameters:
groupCol = Parameter(str, 'id', 'Group column', input_column=True)
outputCol = Parameter(str,
'list',
'List column after grouping',
output_column=True)
class Parameters:
id = Parameter(str, '_id', 'ID column to join on', input_column=True)
class Parameters:
train_weight = Parameter(float, 0.9, 'Train weight of split ratio')
test_weight = Parameter(float, 0.1, 'Test weight of split ratio')
class Parameters:
dropLast = Parameter(bool, True, 'Drop the last category')
inputCol = Parameter(str, 'tokens', 'Input column (%s)', input_column=True, input_dtype=Parameter.T_ARRAY_STRING)
outputCol = Parameter(str, 'features', 'Output column', output_column=True)