ANN class
Methods
-
applyParamsFrom(ANN other)
→ void
-
-
fit(List<List<double>> X, List<List<double>> Y, {int? batchSize, bool verbose = false, String optimizer = 'sgd', double momentum = 0.9, double beta1 = 0.9, double beta2 = 0.999, double epsilon = 1e-8, double l2 = 0.0, String lrSchedule = 'constant', int stepSize = 10, double stepDecay = 0.5, double expDecay = 0.99, int epochsOverride = -1})
→ void
-
Convenience training wrapper that calls the main fit implementation.
(This class intentionally exposes a single training API.)
-
fitAsync(List<List<double>> X, List<List<double>> Y, {int? batchSize, bool verbose = false, String optimizer = 'sgd', double momentum = 0.9, double beta1 = 0.9, double beta2 = 0.999, double epsilon = 1e-8, double l2 = 0.0, String lrSchedule = 'constant', int stepSize = 10, double stepDecay = 0.5, double expDecay = 0.99})
→ Future<void>
-
Asynchronous wrapper that runs
fit in a Future (not in a separate isolate).
-
getParams()
→ Map<String, dynamic>
-
Return a simple params map (weights and biases) without other metadata.
-
markInitialized()
→ void
-
Public utility methods
-
noSuchMethod(Invocation invocation)
→ dynamic
-
Invoked when a nonexistent method or property is accessed.
inherited
-
predict(List<List<double>> X)
→ List<List<double>>
-
-
saveToFile(String path)
→ Future<void>
-
-
toJson()
→ String
-
JSON helpers
-
toMap()
→ Map<String, dynamic>
-
Serialize model parameters to a Map (JSON-ready).
-
toString()
→ String
-
A string representation of this object.
inherited
Static Methods
-
fromJson(String s, {int? seed, int? epochs, double? lr})
→ ANN
-
-
fromMap(Map<String, dynamic> m, {int? seed, int? epochs, double? lr})
→ ANN
-
Reconstruct an ANN from a previously-serialized Map.
-
loadFromFile(String path, {int? seed, int? epochs, double? lr})
→ Future<ANN>
-