deeplearning4j/deeplearning4j · error · java.lang.IllegalArgumentException
No constructor found with parameter count < 3! Rule name
Error message
No constructor found with parameter count < 3! Rule name ${rule.ruleName} What it means
Thrown by AbstractMappingProcessLoader.createProcess when instantiating an AttributeMappingRule from a serialized rule definition. It looks up the rule class in attributeRuleRegistry() and searches for a constructor with 1 or 2 parameters; if the registered class has no such constructor (parameterCount >= 3 or zero), it throws IllegalArgumentException naming the rule.
Solutions
- Add a 1- or 2-parameter constructor to the rule class (Kotlin primary constructor with only inputToOutputMap, plus optional second param) so the loader can instantiate it.
- Check attributeRuleRegistry() to confirm rule.ruleName maps to the intended class, not a same-named different rule.
- If a custom rule needs more config, keep the small constructor and supply the rest via setMappingTransformerArgs / transformer args.
- Regenerate the rule registry after refactoring rule classes so names and classes stay in sync.
Example fix
// before
class MyRule(val map: Map<String,String>, val ctx: Context, val flag: Boolean) : AttributeMappingRule<...>
// after
class MyRule(val map: Map<String,String>) : AttributeMappingRule<...> { ... } // <=2 param constructor for loader reflection Defensive patterns
Strategy: try-catch
Validate before calling
val cls = attributeRuleRegistry()[rule.ruleName]
requireNotNull(cls) { "Rule ${rule.ruleName} not registered" }
require(cls.constructors.any { it.parameterCount == 1 || it.parameterCount == 2 }) {
"Rule ${rule.ruleName} needs a 1- or 2-parameter constructor"
} Try / catch
try {
loader.createProcess(...)
} catch (e: IllegalArgumentException) {
if (e.message?.contains("No constructor found") == true) throw IllegalStateException("Rule class incompatible with loader: ${e.message}", e) else throw e
} Prevention
- Give every custom AttributeMappingRule a primary constructor of at most 2 parameters.
- Pass extra configuration via setMappingTransformerArgs instead of constructor arguments.
- Keep the rule-name-to-class registry generated/updated in the same change as rule class refactors.
When it happens
Trigger: Loading a mapping process from a serialized config (e.g. protocol-buffer-defined process files) where rule.ruleName resolves to a class whose constructors all take >= 3 arguments, or the registry maps the rule name to the wrong class.
Common situations: Adding a custom AttributeMappingRule with a multi-arg constructor and registering it without a 1- or 2-arg constructor; rule name collision causing the registry to resolve a different class; stale/mismatched generated rule registry after refactor; loading old serialized configs against new rule classes.
Understand the failure class
Background: "missing required argument" and "the following required arguments were not provided": what required-argument errors mean and how to fix them — this error's family across 20 libraries.
Related errors
- Found invalid input tensor named
- Found invalid output tensor named
- Illegal type
- Rule for framework $inputFramework does not accept input…
- Rule for framework $inputFramework with input framework…
AI-assisted analysis of deeplearning4j/deeplearning4j@4c22ac5fe4 (2026-09-07).
Data as JSON: /api/errors/805542bae0dd2e53.
Report an issue: GitHub.
Appendix: source
Thrown at nd4j/samediff-import/samediff-import-api/src/main/kotlin/org/nd4j/samediff/frameworkimport/process/AbstractMappingProcessLoader.kt:115
"tensor" -> {
val clazz = tensorRuleRegistry()[rule.ruleName]!!.getConstructor(dictClass,transformerArgsClass)
val transformerArgs = ConcurrentHashMap<String,List<OpNamespace.ArgDescriptor>>()
rule.transformerArgsList.forEach { arg ->
transformerArgs[arg.key] = arg.transformerArgsList
}
val instance = clazz.newInstance(rule.inputToOutputMap,transformerArgs)
listOfTensorRules.add(instance)
}
"attribute" -> {
val transformerArgs = ConcurrentHashMap<String,List<OpNamespace.ArgDescriptor>>()
rule.transformerArgsList.forEach { arg ->
transformerArgs[arg.key] = arg.transformerArgsList
}
val constructor = attributeRuleRegistry()[rule.ruleName]!!.constructors.firstOrNull { constructor -> constructor.parameterCount == 1 || constructor.parameterCount == 2
} ?: throw IllegalArgumentException("No constructor found with parameter count < 3! Rule name ${rule.ruleName}")
when (constructor!!.parameterCount) {
1 -> {
val instance = constructor!!.newInstance(rule.inputToOutputMap) as AttributeMappingRule<GRAPH_TYPE, OP_DEF_TYPE, NODE_DEF_TYPE, ATTRIBUTE_TYPE, ATTRIBUTE_VALUE_TYPE, TENSOR_TYPE, DATA_TYPE>
instance.setMappingTransformerArgs(transformerArgs)
instance.setMappingTransformerArgs(transformerArgs)
instance.modifyName(rule.ruleName)
instance.modifyInputFrameworkOpName(rule.inputFrameworkOpName)
listOfAttributeRules.add(instance)
}
2 -> {
val instance = constructor.newInstance(rule.inputToOutputMap,transformerArgs) as AttributeMappingRule<GRAPH_TYPE, OP_DEF_TYPE, NODE_DEF_TYPE, ATTRIBUTE_TYPE, ATTRIBUTE_VALUE_TYPE, TENSOR_TYPE, DATA_TYPE>
instance.setMappingTransformerArgs(transformerArgs)
instance.modifyName(rule.ruleName)
instance.modifyInputFrameworkOpName(rule.inputFrameworkOpName)
View on GitHub (pinned to 4c22ac5fe4)