clear query| facets| time Search criteria: .   Results from 1 to 10 from 50 (0.0s).
Loading phrases to help you
refine your search...
[expand - 4 more] - "LLR with time" - Mahout - [mail # user]
...✓On Mon, Nov 13, 2017 at 3:32 AM, Ted Dunning  wrote:> Regarding overfitting, don't forget dithering. That can be the most> important single step you take in building a good recom...
   Author: Johannes Schulte , 2017-11-14, 20:02
[MAHOUT-1385] Caching Encoders don't cache - Mahout - [issue]
...The Caching... line of encoders contains code of caching the hash code terms added to the vector. However, the method "hashForProbe" inside this classes is never called as the signature has ...
http://issues.apache.org/jira/browse/MAHOUT-1385    Author: Johannes Schulte , 2015-04-13, 10:21
Regularize calls in AbstractOnlineLogisticRegression - Mahout - [mail # user]
...Hi,i see that one single training instance is used for regularization twice,first in train:    // push coefficients back to zero based on the prior    regularize(instance...
   Author: Johannes Schulte , 2015-01-16, 08:56
[expand - 1 more] - Text clustering with hashing vector encoders - Mahout - [mail # user]
...Hi frank,no, no collocation job. You just take a big enough sample of documents andassign it to it's cluster with the learned ClusterClassifier. Parallel tothat you count the total words in ...
   Author: Johannes Schulte , 2014-03-21, 20:30
OutOfMemoryError: Java Heap Space in DocumentProcessor.tokenizeDocuments - Mahout - [mail # user]
...1I would pass the memory parameters in the args array directly. The hadoopspecific arguments must come before your custom arguments, so like thisString[] args = new String[]{"-Dmapreduce.map...
   Author: Johannes Schulte , 2014-02-22, 20:22
SGD classifier demo app - Mahout - [mail # user]
...Hi Frank,you are using the feature vector encoders which hash a combination offeature name and feature value to 2 (default) locations in the vector. Thevector size you configured is 11 and t...
   Author: Johannes Schulte , 2014-02-03, 22:41
[MAHOUT-1357] InteractionValueEncoder produces wrong traceDictionary entries - Mahout - [issue]
...In the trace code the byte values of the terms being hashed are not converted back to string but just concatenated in their raw form with Arrays.asString()This makes the reverse engineering ...
http://issues.apache.org/jira/browse/MAHOUT-1357    Author: Johannes Schulte , 2014-02-03, 07:57
Item recommendation w/o users or preferences - Mahout - [mail # user]
...Hey,  since you are already using basket analysis terms like support, confidence and lift it might be easier for you to think of the llr score as a "better lift" since it automatically ...
   Author: Johannes Schulte , 2014-01-14, 07:46
[expand - 7 more] - Streaming KMeans clustering - Mahout - [mail # dev]
...Right. Up until now i'am helping myself with some minDf truncation since i am using tf idf weighted vectors anyway and have the idf counts at hand. But having a true loss driven sparsificati...
   Author: Johannes Schulte , 2013-12-30, 14:47
Setting up a recommender - Mahout - [mail # user]
...we have a "cross recommender" in production for about 3 month now, with the difference that we use lucene to build indices from map reduce directly plus we do the same thing for 30+ customer...
   Author: Johannes Schulte , 2013-08-05, 21:55