manifest.json

{
    "title": "LSTM-VAE",
    "description": "self implementation of: https://ieeexplore.ieee.org/document/8279425",
    "inputDimensionality": "univariate",
    "version": "0.3.0",
    "authors": "Siddeshkanth Logonathan",
    "learningType": "Semi-Supervised",
    "language": "Python",
    "type": "Detector",
    "mainFile": "algorithm.py",
    "trainingStep": {
        "parameters": [
            {
                "name": "rnn_hidden_size",
                "type": "int",
                "defaultValue": 5,
                "optional": "false",
                "description": "LTSM cells hidden dimension"
            },
            {
                "name": "latent_size",
                "type": "int",
                "defaultValue": 5,
                "optional": "true",
                "description": "dimension of latent space"
            },
            {
                "name": "learning_rate",
                "type": "float",
                "defaultValue": 0.001,
                "optional": "false",
                "description": "rate at which the gradients are updated"
            },
            {
                "name": "batch_size",
                "type": "int",
                "defaultValue": 32,
                "optional": "false",
                "description": "size of batch given for each iteration"
            },
            {
                "name": "epochs",
                "type": "int",
                "defaultValue": 10,
                "optional": "false",
                "description": "number of iterations we train the model"
            },
            {
                "name": "window_size",
                "type": "int",
                "defaultValue": 10,
                "optional": "false",
                "description": "number of datapoints that the model takes once"
            },
            {
                "name": "lstm_layers",
                "type": "int",
                "defaultValue": 10,
                "optional": "false",
                "description": "number of layers in lstm"
            },
            {
              "name": "early_stopping_delta",
              "type": "float",
              "defaultValue": 0.05,
              "optional": "true",
              "description": "If 1 - (loss / last_loss) is less than `delta` for `patience` epochs, stop"
            },
            {
              "name": "early_stopping_patience",
              "type": "int",
              "defaultValue": 10,
              "optional": "true",
              "description": "If 1 - (loss / last_loss) is less than `delta` for `patience` epochs, stop"
            }
        ],
        "modelInput": "none"
    },
    "executionStep": {
        "parameters": [
            {
                "name": "batch_size",
                "type": "int",
                "defaultValue": 32,
                "optional": "false",
                "description": "size of batch given for each iteration"
            }
        ],
        "modelInput": "required"
    }
}