tech.compute.tensor.operations — techascent/tech.compute 3.0

tensor operations with syntatic sugar

* clj

(* output x1)

(* output x1 x2)

Takes and x1 and x2 multiples them together and puts the result in the mutated output

+ clj

(+ output x1)

(+ output x1 x2)

Takes and x1 and x2 adds them and puts the result in the mutated output

- clj

(- output x1)

(- output x1 x2)

Takes and x1 and x2 subtracts them and puts the result in the mutated output

/ clj

(/ output x1)

(/ output x1 x2)

Divides 1 by x2 them together and puts the result in the mutated output

< clj

(< output x1)

(< output x1 x2)

Takes and x1 and x2 returns 1 if x1 < x2 and 0 otherwise them and puts the result in the mutated output

<= clj

(<= output x1)

(<= output x1 x2)

Takes and x1 and x2 returns 1 if x1 <= x2 and 0 otherwise them and puts the result in the mutated output

> clj

(> output x1)

(> output x1 x2)

Takes and x1 and x2 returns 1 if x1 > x2 and 0 otherwise them and puts the result in the mutated output

>= clj

(>= output x1)

(>= output x1 x2)

Takes and x1 and x2 returns 1 if x1 >= x2 and 0 otherwise them and puts the result in the mutated output

bit-and clj

(bit-and output x1)

(bit-and output x1 x2)

Takes and x1 and x2 returns bit and of x1 and x2 and them and puts the result in the mutated output

bit-xor clj

(bit-xor output x1)

(bit-xor output x1 x2)

Takes and x1 and x2 returns xor x1 and x2 them and puts the result in the mutated output

ceil clj

(ceil output)

(ceil output input)

Takes an tensor returns the mutated tensor with the value with the ceiling function applied

exp clj

(exp output)

(exp output input)

Takes an tensor returns the mutated tensor with the value with the exp function applied

floor clj

(floor output)

(floor output input)

Takes an tensor returns the mutated tensor with the value with the floor function applied

logistic clj

(logistic output)

(logistic output input)

Takes an tensor returns the mutated tensor with the value with the logistic function applied

max clj

(max output max-value)

(max output input max-value)

Takes an input tensor and returns the max of x or the value given, mutates the output tensor and returns it

min clj

(min output min-value)

(min output input min-value)

Takes an input tensor and returns the min of x or the value given, mutates the output tensor and returns it

new-tensor clj

(new-tensor output)

Returns a new tensor of the same shape and type of the given output tensor

tanh clj

(tanh output)

(tanh output input)

Takes an tensor returns the mutated tensor with the value with the tanh function applied

where clj

(where output test then else)

Takes a tests tensor of 1s and 0s and a tensor of then and a tensor of else. It will multiply the test and the complement of the test to the then and else and then add the result to the output. The condition tensor acts as a mask that chooses, based on the value at each element, whether the corresponding element / row in the output should be taken from x1 (if true) or x2 (if false).