Ahmed Adel created SOLR-13903:
---------------------------------
Summary: Classification Model Confusion Matrix Discrepancy
Key: SOLR-13903
URL: https://issues.apache.org/jira/browse/SOLR-13903
Project: Solr
Issue Type: Bug
Security Level: Public (Default Security Level. Issues are Public)
Components: streaming expressions
Affects Versions: 8.2
Reporter: Ahmed Adel
Attachments: cellphones.csv
Using features and train stream sources generate a model with TP, TN, FP, FN
fields. For some reason, the summation of the values of these fields is
sometimes less than the training set size.
How to regenerate:
# Create two collections: cellphones and cellphones-model
# Indexing the attached dataset into cellphones
# Run the following expression:
{{commit(cellphones-model,update(cellphones-model,batchSize=500,
}}{{ train(cellphones,
}}{{ features(cellphones, q="*:*", featureSet="featureSet",
field="title_t", outcome="brand_i", numTerms=25),
}}{{ q="*:*",
}}{{ name="cellphones-classification-model",
}}{{ field="title_t",
}}{{ outcome="brand_i",
}}{{ maxIterations=100)))
}}
4) Run the following query to retrieve confusion matrix:
{{search
q=*:*&collection=cellphones-model&fl=name_s,trueNegative_i,truePositive_i,falseNegative_i,falsePositive_i,iteration_i&sort=iteration_i%20desc&rows=100
}}
The summation of the metrics TP, TN, FP, FN is always less than the training
set size by one in this instance for all iterations.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org
For additional commands, e-mail: issues-h...@lucene.apache.org
