Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Support Support Vector Machine models #31

Merged
merged 5 commits into from
Jul 24, 2021
Merged

Support Support Vector Machine models #31

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
b1a273a
refactor: move PMML array parsing to separate and tested function
iamDecode Jul 22, 2021
54bb033
feat: add support for Support Vector Machine models
iamDecode Jul 22, 2021
734b3ff
feat: support multi-class SVM classifier
iamDecode Jul 22, 2021
01d78c6
test: add additional tests for SVM parse exceptions
iamDecode Jul 23, 2021
328f7e1
docs: add package readme for svm
iamDecode Jul 23, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
20 changes: 19 additions & 1 deletion README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,6 +31,7 @@ This library is in beta, and currently not all models are supported. The library
| [Lasso](sklearn_pmml_model/linear_model) | ✅<sup>2</sup> | ✅ | ✅<sup>3</sup> |
| [ElasticNet](sklearn_pmml_model/linear_model) | ✅<sup>2</sup> | ✅ | ✅ |
| [Gaussian Naive Bayes](sklearn_pmml_model/naive_bayes) | ✅ | | ✅<sup>3</sup> |
| [Support Vector Machines](sklearn_pmml_model/svm) | ✅ | ✅ | ✅<sup>3</sup> |

<sub><sup>1</sup> Categorical feature support using slightly modified internals, based on [scikit-learn#12866](https://github.com/scikit-learn/scikit-learn/pull/12866).</sub>

Expand All @@ -41,8 +42,12 @@ This library is in beta, and currently not all models are supported. The library
---

The following part of the [specification](http://dmg.org/pmml/v4-3/GeneralStructure.html) is covered:
- Array (including typed variants)
- SparseArray *(including typed variants)*
- Indices
- Entries *(including typed variants)*
- DataDictionary
- DataField (continuous, categorical, ordinal)
- DataField *(continuous, categorical, ordinal)*
- Value
- Interval
- TransformationDictionary / LocalTransformations
Expand All @@ -69,6 +74,19 @@ The following part of the [specification](http://dmg.org/pmml/v4-3/GeneralStruct
- PairCounts
- TargetValueCounts
- TargetValueCount
- SupportVectorMachineModel
- LinearKernelType
- PolynomialKernelType
- RadialBasisKernelType
- SigmoidKernelType
- VectorDictionary
- VectorFields
- VectorInstance
- SupportVectorMachine
- SupportVectors
- SupportVector
- Coefficients
- Coefficient

## Example
A minimal working example is shown below:
Expand Down
297 changes: 297 additions & 0 deletions models/svc-cat-pima.pmml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,297 @@
<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<PMML xmlns="http://www.dmg.org/PMML-4_4" xmlns:data="http://jpmml.org/jpmml-model/InlineTable" version="4.4">
<Header>
<Application name="JPMML-R" version="1.4.4"/>
<Timestamp>2021-07-22T11:20:48Z</Timestamp>
</Header>
<DataDictionary>
<DataField name="type" optype="categorical" dataType="string">
<Value value="No"/>
<Value value="Yes"/>
</DataField>
<DataField name="npreg" optype="continuous" dataType="double"/>
<DataField name="glu" optype="continuous" dataType="double"/>
<DataField name="bp" optype="continuous" dataType="double"/>
<DataField name="skin" optype="continuous" dataType="double"/>
<DataField name="bmi" optype="continuous" dataType="double"/>
<DataField name="ped" optype="continuous" dataType="double"/>
<DataField name="age" optype="categorical" dataType="string">
<Value value="(20,30]"/>
<Value value="(30,40]"/>
<Value value="(40,50]"/>
<Value value="(50,60]"/>
<Value value="(60,70]"/>
</DataField>
</DataDictionary>
<TransformationDictionary/>
<SupportVectorMachineModel functionName="classification" classificationMethod="OneAgainstOne">
<MiningSchema>
<MiningField name="type" usageType="target"/>
<MiningField name="age"/>
<MiningField name="npreg"/>
<MiningField name="glu"/>
<MiningField name="bp"/>
<MiningField name="skin"/>
<MiningField name="bmi"/>
<MiningField name="ped"/>
</MiningSchema>
<LocalTransformations>
<DerivedField name="scale(npreg)" optype="continuous" dataType="double">
<Apply function="/">
<Apply function="-">
<FieldRef field="npreg"/>
<Constant dataType="double">3.7884615384615383</Constant>
</Apply>
<Constant dataType="double">3.7747174337126292</Constant>
</Apply>
</DerivedField>
<DerivedField name="scale(glu)" optype="continuous" dataType="double">
<Apply function="/">
<Apply function="-">
<FieldRef field="glu"/>
<Constant dataType="double">127.6923076923077</Constant>
</Apply>
<Constant dataType="double">30.590616568175218</Constant>
</Apply>
</DerivedField>
<DerivedField name="scale(bp)" optype="continuous" dataType="double">
<Apply function="/">
<Apply function="-">
<FieldRef field="bp"/>
<Constant dataType="double">74.68858981173888</Constant>
</Apply>
<Constant dataType="double">10.513032937113188</Constant>
</Apply>
</DerivedField>
<DerivedField name="scale(skin)" optype="continuous" dataType="double">
<Apply function="/">
<Apply function="-">
<FieldRef field="skin"/>
<Constant dataType="double">31.482109488578555</Constant>
</Apply>
<Constant dataType="double">10.51639268163093</Constant>
</Apply>
</DerivedField>
<DerivedField name="scale(bmi)" optype="continuous" dataType="double">
<Apply function="/">
<Apply function="-">
<FieldRef field="bmi"/>
<Constant dataType="double">34.69397755176341</Constant>
</Apply>
<Constant dataType="double">6.232781228378887</Constant>
</Apply>
</DerivedField>
<DerivedField name="scale(ped)" optype="continuous" dataType="double">
<Apply function="/">
<Apply function="-">
<FieldRef field="ped"/>
<Constant dataType="double">0.47128846153846154</Constant>
</Apply>
<Constant dataType="double">0.3016158663675463</Constant>
</Apply>
</DerivedField>
</LocalTransformations>
<RadialBasisKernelType gamma="0.09090909090909091"/>
<VectorDictionary>
<VectorFields>
<FieldRef field="scale(npreg)"/>
<FieldRef field="scale(glu)"/>
<FieldRef field="scale(bp)"/>
<FieldRef field="scale(skin)"/>
<FieldRef field="scale(bmi)"/>
<FieldRef field="scale(ped)"/>
<CategoricalPredictor name="age" value="(20,30]" coefficient="1.0"/>
<CategoricalPredictor name="age" value="(30,40]" coefficient="1.0"/>
<CategoricalPredictor name="age" value="(40,50]" coefficient="1.0"/>
<CategoricalPredictor name="age" value="(50,60]" coefficient="1.0"/>
<CategoricalPredictor name="age" value="(60,70]" coefficient="1.0"/>
</VectorFields>
<VectorInstance id="1">
<Array type="real">-0.47380011083438744 0.0100583885586802 0.3149814338135626 0.5246942253363732 1.3807676112634821 2.495596626021958 0.0 1.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="2">
<Array type="real">2.175404809960037 -1.1667730728069075 -1.206939033449189 -2.327994991223717 -1.1381720762896912 1.5075849421907774 0.0 0.0 1.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="7">
<Array type="real">0.32096136540393994 0.36964577953149863 0.5052214922214066 0.3345149442323672 -0.496404003027736 -0.36565868654954997 1.0 0.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="9">
<Array type="real">-0.7387206029138299 -0.4149085280455598 -0.4459787998178132 -0.1409332585276478 -0.015077948081269651 0.19134118889890142 0.0 1.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="12">
<Array type="real">-0.20887961875494498 0.04274815137439097 1.6466618426684703 1.6657699119604092 0.2737176848866097 1.64683491105289 0.0 1.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="13">
<Array type="real">-1.0036410949932724 2.298341785658434 -0.8264589166335011 0.049246022576358225 1.0598835746325046 0.10182335177325738 1.0 0.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="14">
<Array type="real">-1.0036410949932724 1.9714441575013264 0.6954615506292504 -1.6623675073596957 -0.4322271957015407 0.6986089326108839 1.0 0.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="15">
<Array type="real">-0.20887961875494498 0.9907512730300032 0.12474137540571863 0.42960458478437025 -0.496404003027736 1.2589242835084329 1.0 0.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="16">
<Array type="real">-1.0036410949932724 0.7619229333200279 1.4564217842606262 1.3805009903044003 1.1882371892848962 -0.3325039320585707 1.0 0.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="17">
<Array type="real">-1.0036410949932724 -1.068703784359775 0.9808216382410163 -0.6163814612876628 0.4341597032020985 -0.7436228877467133 1.0 0.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="18">
<Array type="real">2.7052457941189223 1.5464772408970864 -1.206939033449189 -0.1409332585276478 -0.17551996639675843 -0.8596645284651405 0.0 1.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="19">
<Array type="real">0.8508023495628249 0.04274815137439097 -0.6362188582256572 1.6657699119604092 0.6106459233491364 -0.10705160151991185 0.0 0.0 1.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="20">
<Array type="real">1.1157228416422673 0.892681984582871 -1.206939033449189 -0.5212918207356598 -0.11134315907056315 0.2377578451862724 0.0 0.0 1.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="22">
<Array type="real">1.1157228416422673 -0.08801089988845211 2.027141959484158 0.22372578857205253 0.01489786606199215 -0.7933550194831821 0.0 0.0 0.0 1.0 0.0</Array>
</VectorInstance>
<VectorInstance id="23">
<Array type="real">-0.47380011083438744 -0.31683923959842747 0.5052214922214066 -0.07891157559922476 1.316590803937287 0.7350791625509607 1.0 0.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="24">
<Array type="real">1.1157228416422673 0.17350720263723404 -0.2557387414099693 0.11336682344895582 -0.28782937921760104 -0.667366952417461 0.0 1.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="25">
<Array type="real">-1.0036410949932724 0.4350253051629202 0.13682302113358621 0.38335895637305323 1.2363697947795425 -0.8828728566088261 1.0 0.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="26">
<Array type="real">-0.20887961875494498 0.4350253051629202 -0.2407884106009078 -0.06409135627114358 -0.7531112323325183 0.9605314930896198 1.0 0.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="27">
<Array type="real">-1.0036410949932724 1.2195796127399787 0.12474137540571863 1.0952320686483912 2.1188008955147306 -0.7038371823575382 1.0 0.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="30">
<Array type="real">-0.47380011083438744 -0.5783573421241136 -0.06549868300212532 -0.2360228990796508 -0.3680503883753454 0.7516565397964504 1.0 0.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="31">
<Array type="real">-0.20887961875494498 0.6638536448728956 -0.8264589166335011 -0.6163814612876628 -0.35200618654379634 -0.7137836087048319 1.0 0.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="33">
<Array type="real">-1.0036410949932724 -0.28414947678271674 -0.8264589166335011 -0.4262021801836568 0.6587785288437825 -0.7038371823575382 1.0 0.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="34">
<Array type="real">-0.7387206029138299 -1.0033242587283537 -0.6362188582256572 -0.9967400234956748 -1.202348883615887 2.0679002930883263 1.0 0.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="36">
<Array type="real">-0.47380011083438744 -0.08801089988845211 -1.3971790918570328 -1.091829664047678 -0.14343156273366137 -1.2707834841532832 0.0 1.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="37">
<Array type="real">-0.7387206029138299 -0.44759829086127056 -0.8264589166335011 0.42960458478437025 0.5464691160229411 -0.6043729188846005 1.0 0.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="38">
<Array type="real">0.5858818574833824 -1.1667730728069075 -1.206939033449189 0.049246022576358225 -0.4322271957015407 -1.280729910500577 0.0 0.0 1.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="39">
<Array type="real">-1.0036410949932724 0.2388867282686556 1.8369019010763141 1.3805009903044003 0.9475741618116631 -0.6209502961300902 1.0 0.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="40">
<Array type="real">-0.47380011083438744 -0.2187699511512952 -0.4459787998178132 0.049246022576358225 0.7069111343384299 1.3749659242268604 1.0 0.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="41">
<Array type="real">2.175404809960037 -0.2187699511512952 0.3149814338135626 -1.3770985857036868 -1.314658296436729 -0.7038371823575382 0.0 0.0 0.0 0.0 1.0</Array>
</VectorInstance>
<VectorInstance id="43">
<Array type="real">-0.20887961875494498 -2.1801557200939414 0.6954615506292504 -0.33111253963165377 -0.04716635174436787 -0.756884789543105 0.0 0.0 1.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="44">
<Array type="real">0.5858818574833824 -0.5456675793084028 -1.016698975041345 0.7148735064403793 -0.07925475540746493 -0.7005217069084403 1.0 0.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="45">
<Array type="real">-0.7387206029138299 -1.526360463779726 -0.06549868300212532 0.9050527875443852 1.8620936662099483 2.0712157685374244 0.0 1.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="46">
<Array type="real">-1.0036410949932724 0.3042662539000771 0.8857016090370944 -0.4262021801836568 -1.186304681784338 -0.79667049493228 0.0 0.0 0.0 1.0 0.0</Array>
</VectorInstance>
<VectorInstance id="49">
<Array type="real">-1.0036410949932724 -0.9052549702812214 1.2661817258527823 2.7117559580324424 1.9423146753676928 1.6269420583583023 0.0 1.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="51">
<Array type="real">-0.20887961875494498 -0.12070066270416287 0.5052214922214066 0.1443356631283612 -0.23969677372295373 -0.5513253116990338 1.0 0.0 0.0 0.0 0.0</Array>
</VectorInstance>
<VectorInstance id="52">
<Array type="real">0.32096136540393994 0.36964577953149863 -1.016698975041345 0.3345149442323672 -0.9777300579742023 -0.1998849140946538 1.0 0.0 0.0 0.0 0.0</Array>
</VectorInstance>
</VectorDictionary>
<SupportVectorMachine targetCategory="No" alternateTargetCategory="Yes">
<SupportVectors>
<SupportVector vectorId="1"/>
<SupportVector vectorId="2"/>
<SupportVector vectorId="7"/>
<SupportVector vectorId="9"/>
<SupportVector vectorId="12"/>
<SupportVector vectorId="13"/>
<SupportVector vectorId="14"/>
<SupportVector vectorId="15"/>
<SupportVector vectorId="16"/>
<SupportVector vectorId="17"/>
<SupportVector vectorId="18"/>
<SupportVector vectorId="19"/>
<SupportVector vectorId="20"/>
<SupportVector vectorId="22"/>
<SupportVector vectorId="23"/>
<SupportVector vectorId="24"/>
<SupportVector vectorId="25"/>
<SupportVector vectorId="26"/>
<SupportVector vectorId="27"/>
<SupportVector vectorId="30"/>
<SupportVector vectorId="31"/>
<SupportVector vectorId="33"/>
<SupportVector vectorId="34"/>
<SupportVector vectorId="36"/>
<SupportVector vectorId="37"/>
<SupportVector vectorId="38"/>
<SupportVector vectorId="39"/>
<SupportVector vectorId="40"/>
<SupportVector vectorId="41"/>
<SupportVector vectorId="43"/>
<SupportVector vectorId="44"/>
<SupportVector vectorId="45"/>
<SupportVector vectorId="46"/>
<SupportVector vectorId="49"/>
<SupportVector vectorId="51"/>
<SupportVector vectorId="52"/>
</SupportVectors>
<Coefficients absoluteValue="-0.050320746562785366">
<Coefficient value="-1.0"/>
<Coefficient value="-1.0"/>
<Coefficient value="-1.0"/>
<Coefficient value="-1.0"/>
<Coefficient value="-0.5644464780552596"/>
<Coefficient value="-0.8200625306356393"/>
<Coefficient value="-0.7295267913726711"/>
<Coefficient value="-0.39963111744411345"/>
<Coefficient value="-1.0"/>
<Coefficient value="-1.0"/>
<Coefficient value="-0.41240221176283304"/>
<Coefficient value="-1.0"/>
<Coefficient value="-0.9978584268568605"/>
<Coefficient value="-1.0"/>
<Coefficient value="-1.0"/>
<Coefficient value="-1.0"/>
<Coefficient value="-1.0"/>
<Coefficient value="-1.0"/>
<Coefficient value="1.0"/>
<Coefficient value="1.0"/>
<Coefficient value="1.0"/>
<Coefficient value="0.5986629800653184"/>
<Coefficient value="1.0"/>
<Coefficient value="0.817336947438522"/>
<Coefficient value="0.3880321541690222"/>
<Coefficient value="0.11989547445451426"/>
<Coefficient value="1.0"/>
<Coefficient value="1.0"/>
<Coefficient value="1.0"/>
<Coefficient value="1.0"/>
<Coefficient value="1.0"/>
<Coefficient value="1.0"/>
<Coefficient value="1.0"/>
<Coefficient value="1.0"/>
<Coefficient value="1.0"/>
<Coefficient value="1.0"/>
</Coefficients>
</SupportVectorMachine>
</SupportVectorMachineModel>
</PMML>