Miscellaneous

float_epsilon

Python

Syntax:

n.float_epsilon = 1e-11

Description:

The default value is 1e-11

Allows somewhat safer NEURON logical comparisons and integer truncation for floating point numbers. Most NEURON comparisons are treated as true if they are within float_epsilon of being true. e.g.,

from neuron import n

n("""
proc count_to_1() {for (i = 0; i < 1; i += 0.1) $o1.__call__(i)}
""")

def print_i(i):
    print(f'{i:3g} {int(10*i)}')

rv = n.count_to_1(print_i)

n.float_epsilon = 0

# two bugs due to roundoff
rv = n.count_to_1(print_i)

Warning

This has no effect on Python comparisons, which is why the example uses a HOC procedure.

Warning

We certainly haven’t gotten every floating comparison in the program to use float_epsilon but NEURON has most of them including all HOC interpreter logical operations, int, array indices, and Vector logic methods.

HOC

Syntax:

float_epsilon = 1e-11
Description:
The default value is 1e-11

Allows somewhat safer logical comparisons and integer truncation for floating point numbers. Most comparisons are treated as true if they are within float_epsilon of being true. e.g.,
for (i = 0; i < 1; i += .1) {
    print i, int(10*i)
}
float_epsilon = 0
// two bugs due to roundoff
for (i = 0; i < 1; i += .1) {
    print i, int(10*i)
}

Warning

MATLAB

Syntax:

n.float_epsilon = 1e-11

Description:

The default value is 1e-11

Allows somewhat safer NEURON logical comparisons and integer truncation for floating point numbers. Most NEURON comparisons are treated as true if they are within float_epsilon of being true. e.g.,

n = neuron.launch();

n('print 1.01 == 1.0');  % 0 i.e., false
n.float_epsilon = 0.1;
n('print 1.01 == 1.0');  % 1 i.e., true

Warning

This has no effect on MATLAB comparisons, which is why the example does the comparison using a HOC statement.

Warning