serialized representation of a graph
The default batchPerThread is 4.
The default batchPerThread is 4. For DistributedTextSet, the total batchSize is batchPerThread * rdd.getNumPartitions. For LocalTextSet, the total batchSize is batchPerThread * numOfCores.
Prediction data, TextSet.
Use a model to do prediction on TextSet.
Use a model to do prediction on TextSet.
Prediction data, TextSet.
The total batch size is batchPerThread * rdd.getNumPartitions(distributed mode) or batchPerThread * numOfCores(local mode)
The default batchPerThread is 4.
The default batchPerThread is 4. For DistributedImageSet, the total batchSize is batchPerThread * rdd.getNumPartitions. For LocalImageSet, the total batchSize is batchPerThread * numOfCores.
Prediction data, ImageSet.
Use a model to do prediction on ImageSet.
Use a model to do prediction on ImageSet.
Prediction data, ImageSet.
The total batch size is batchPerThread * rdd.getNumPartitions(distributed mode) or batchPerThread * numOfCores(local mode)
Use a model to do prediction in local mode.
Use a model to do prediction in local mode. The total batch size is batchPerThread * numOfCores, and batchPerThread is 4 by default.
Prediction data, array of Sample.
Use a model to do prediction in local mode.
Use a model to do prediction in local mode.
Prediction data, array of Sample.
The total batchSize is batchPerThread * numOfCores.
Use a model to do prediction in local mode.
Use a model to do prediction in local mode. The total batch size is batchPerThread * numOfCores, and batchPerThread is 4 by default.
Prediction data, LocalDataSet.
Use a model to do prediction in local mode.
Use a model to do prediction in local mode.
Prediction data, LocalDataSet.
The total batchSize is batchPerThread * numOfCores.
Use a model to do prediction for RDD.
Use a model to do prediction for RDD. The default batchPerThread is 4, and the total batchSize is batchPerThread * rdd.getNumPartitions.
Prediction data, RDD of Sample.
Use a model to do prediction for RDD.
Use a model to do prediction for RDD.
Prediction data, RDD of Sample.
The total batchSize is batchPerThread * rdd.getNumPartitions.
Use a model to predict for classes.
Use a model to predict for classes. By default, label predictions start from 0.
Prediction data, RDD of Sample.
The default batchPerThread is 4, and the total batchSize is batchPerThread * rdd.getNumPartitions.
Boolean. Whether result labels start from 0. Default is true. If false, result labels start from 1.
(Since version 0.3.0) please use recommended saveModule(path, overWrite)
TFNet wraps a tensorflow subgraph as a layer, and use tensorflow to calculate the layer's output.
This subgraph should not contain any tensorflow Variable and the input/output must be numeric types
When used with other layers for training, there should be no trainable layer before this one, as the gradInput of this layer is always zero.