Model¶
Create, load/save, and use MXNet models¶
Create a MXNet Feedforward neural net model with the specified training |
|
Train RNN with bucket support |
|
Parameter initialization |
|
Load model checkpoint from file |
|
Save model checkpoint into file |
|
Predict the outputs given a model and dataset |
|
Check if the model has been serialized into RData-compatiable format |
|
Serialize MXNet model into RData-compatiable format |
|
Unserialize MXNet model from Robject |
|
Convenience interface for multiple layer perceptron |
|
Inference of RNN model |
|
Inference for one-to-one fusedRNN (CUDA) models |
|
Inference for one-to-one unroll models |
|
Applies recurrent layers to input data |
|
Applies recurrent layers to input data |
|
Generate a RNN symbolic model - requires CUDA |
|
Unroll representation of RNN running on non CUDA device |