Parameters of H2OWord2Vec¶
Affected Class¶
ai.h2o.sparkling.ml.features.H2OWord2Vec
Parameters¶
Each parameter has also a corresponding getter and setter method. (E.g.:
label->getLabel(),setLabel(...))
- epochs
Number of training iterations to run.
Default value:
5- exportCheckpointsDir
Automatically export generated models to this directory.
Scala default value:
null; Python default value:None- initLearningRate
Set the starting learning rate.
Scala default value:
0.025f; Python default value:0.025- inputCol
input column name
Default value:
"No default value"- maxRuntimeSecs
Maximum allowed runtime in seconds for model training. Use 0 to disable.
Default value:
0.0- minWordFreq
This will discard words that appear less than <int> times.
Default value:
5- modelId
Destination id for this model; auto-generated if not specified.
Scala default value:
null; Python default value:None- normModel
Use Hierarchical Softmax. Possible values are
"HSM".Default value:
"HSM"- outputCol
output column name
Default value:
"H2OWord2Vec_output"- sentSampleRate
- Set threshold for occurrence of words. Those that appear with higher frequency in the training data
will be randomly down-sampled; useful range is (0, 1e-5).
Scala default value:
0.001f; Python default value:0.001- vecSize
Set size of word vectors.
Default value:
100- windowSize
Set max skip length between words.
Default value:
5- wordModel
The word model to use (SkipGram or CBOW). Possible values are
"SkipGram","CBOW".Default value:
"SkipGram"