Documentation
¶
Index ¶
- func LogisticRegressionMap(result *map[string]interface{}, row map[string]interface{}, ...) error
- func LogisticRegressionReduce(result *map[string]interface{}, row map[string]interface{}, ...) error
- func Max(result *map[string]interface{}, row map[string]interface{}, ...) error
- func Min(result *map[string]interface{}, row map[string]interface{}, ...) error
- func Preprocessing(result *map[string]interface{}, row map[string]interface{}, ...) error
- func Sum(result *map[string]interface{}, row map[string]interface{}, ...) error
- func SumSquare(result *map[string]interface{}, row map[string]interface{}, ...) error
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
func LogisticRegressionMap ¶
func LogisticRegressionMap(result *map[string]interface{}, row map[string]interface{}, args map[string]interface{}) error
LogisticRegressionMap returns Gradient (Resp. & Hessian) to compute SGD (Resp. Newton-Raphson) Inspired from https://papers.nips.cc/paper/3150-map-reduce-for-machine-learning-on-multicore.pdf and https://www.internalpointers.com/post/cost-function-logistic-regression In both case, gradient should be computed on the data received by the mapper
func LogisticRegressionReduce ¶
func LogisticRegressionReduce(result *map[string]interface{}, row map[string]interface{}, args map[string]interface{}) error
LogisticRegressionReduce sums up Gradient and Hessian (if Newton Raphson)
func Max ¶
func Max(result *map[string]interface{}, row map[string]interface{}, args map[string]interface{}) error
Max takes in input the data a map speciafying some parameters : - keys : Value on which compute the Max. The specified keys should be one of the keys from Data.
func Min ¶
func Min(result *map[string]interface{}, row map[string]interface{}, args map[string]interface{}) error
Min takes in input the data a map speciafying some parameters : - keys : Value on which compute the Min. The specified keys should be one of the keys from Data.
func Preprocessing ¶
func Preprocessing(result *map[string]interface{}, row map[string]interface{}, args map[string]interface{}) error
Preprocessing apply a preprocessing over data before training on it Preprocessing takes as argument the doctype to choose which preprocess to use TODO : Broke the preprocess function in pieces to create more atomic functions
func Sum ¶
func Sum(result *map[string]interface{}, row map[string]interface{}, args map[string]interface{}) error
Sum takes in input the data a map speciafying some parameters : - keys : Value on which compute the sum. The specified keys should be one of the keys from Data. - weight : specify a key to compute a weighted sum
func SumSquare ¶
func SumSquare(result *map[string]interface{}, row map[string]interface{}, args map[string]interface{}) error
SumSquare takes in input the data a map speciafying some parameters : - keys : Value on which compute the sum. The specified keys should be one of the keys from Data. - weight : specify a key to compute a weighted sum
Types ¶
This section is empty.
Source Files