nnl4

package

v0.1.7 Latest Latest Go to latest Published: Mar 25, 2019 License: MIT Imports: 6 Imported by: 0

Details

Valid go.mod file

The Go module system was introduced in Go 1.11 and is the official dependency management solution for Go.
Redistributable license

Redistributable licenses place minimal restrictions on how software can be used, modified, and redistributed.
Tagged version

Modules with tagged versions give importers more predictable builds.
Stable version

When a project reaches major version v1 it is considered stable.
Learn more about best practices

Repository

github.com/sdeoras/comp

README ¶

train

this is a simple go wrapper on TensorFlow trainer.

installation

install Go API for TensorFlow https://www.tensorflow.org/install/lang_go

$ go get github.com/sdeoras/comp/nn/nnl4

details

it implements a simple linear model in the form:

y1 = x * W1 + B1  // with relu
y2 = y1 * W2 + B2 // with relu
y3 = y2 * W3 + B3 // with relu
y = y3 * W4 + B4 // with softmax
where, x is input, Wx's are weights matrices and Bx's are biases matrices

a new trainer can be initialized as follows:

func New(dim []int, learningRate float64) (*Op, error) {...}

Documentation ¶

Index ¶

type Op
- func New(dim []int, learningRate float64, name, logdir string, ...) (*Op, error)
- func NewOperator(options *tf.SessionOptions) (*Op, error)

Constants ¶

This section is empty.

Variables ¶

This section is empty.

Functions ¶

This section is empty.

Types ¶

type Op ¶

type Op struct {
	// It embeds common op
	common.Op
	// contains filtered or unexported fields
}

func New ¶

func New(dim []int, learningRate float64, name, logdir string, options *tf.SessionOptions) (*Op, error)

New provides a new trainer operator that can be used to feed taining data and step through training process.

func NewOperator ¶

func NewOperator(options *tf.SessionOptions) (*Op, error)

NewOperator provides an instance of new operator for interfacing with TF.

func (*Op) Load ¶

func (op *Op) Load(checkPoint *nn.CheckPoint) error

Load loads a given checkpoitn into current training session of receiver.

func (*Op) Predict ¶

func (op *Op) Predict(data *nn.Data) (*nn.PredictionOutput, error)

Predict does inference on given data.

func (*Op) Save ¶

func (op *Op) Save() (*nn.CheckPoint, error)

Save saves model checkpoint.

func (*Op) Step ¶

func (op *Op) Step(trainingData, validationData *nn.Data) (*nn.TrainingOutput, error)

Step steps through iteration process.

Source Files ¶

View all Source files

?	: This menu
/	: Search site
f or F	: Jump to
y or Y	: Canonical URL