Adagrad
Customizes the Adagrad optimizer.
Prototype
1 | def create_hash_optimizer(learning_rate=0.001, initial_accumulator_value=0.9, use_locking=False, name="Adagrad") |
Parameters
Parameter |
Type |
Mandatory/Optional |
Description |
|---|---|---|---|
learning_rate |
float/tf.Tensor |
Optional |
Learning rate. Default value: 0.001 Value range: [0.0, 10.0] |
initial_accumulator_value |
float |
Optional |
Initial value of the accumulator Value range: (0.0, 1.0] Default value: 0.9 |
use_locking |
bool |
Optional |
Prevents concurrent updates to variables in the optimizer. Default value: False Value range:
|
name |
string |
Optional |
Optimizer name. Default value: Adagrad Name length range: [1, 200] |
Return Value
An instance object of the CustomizedAdagrad optimizer.
Example
1 2 | from mx_rec.optimizers.adagrad import create_hash_optimizer hashtable_optimizer = create_hash_optimizer(0.001) |
Parent topic: Optimizers