mirror of
https://github.com/vladmandic/automatic
synced 2026-08-27 23:51:00 +02:00
vladmandic
126 lines
4.5 KiB
Python
126 lines
4.5 KiB
Python
#!/usr/bin/env python
|
|
|
|
import sys
|
|
import os
|
|
from collections import Counter
|
|
|
|
|
|
script_dir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
|
|
os.chdir(script_dir)
|
|
|
|
|
|
# library
|
|
fn = r'./modules/styles.py'
|
|
# tested function
|
|
funcname = 'select_from_weighted_list'
|
|
# random needed
|
|
ns = {'Dict': dict, 'random': __import__('random')}
|
|
# number of samples to test
|
|
tries = 10000
|
|
# allowed deviation in percentage points
|
|
tolerance_pct = 2.0
|
|
# tests
|
|
tests = [
|
|
# - empty
|
|
["", { '': 100 } ],
|
|
# - no weights
|
|
[ "red|blonde|black", { 'black': 33, 'red': 33, 'blonde': 33 } ],
|
|
# - list with empty entry
|
|
[ "red||black", { 'black': 33, 'red': 33, '': 33 } ],
|
|
# - list with repeated entry
|
|
[ "red|blonde|blonde", { 'red': 33, 'blonde': 66 } ],
|
|
# - full weights <= 1
|
|
[ "red:0.1|blonde:0.9", { 'red': 10, 'blonde': 90 } ],
|
|
# - weights > 1 to test normalization
|
|
[ "red:1|blonde:2|black:5", { 'red': 12.5, 'blonde': 25, 'black': 62.5 } ],
|
|
# - disabling 0 weights to force one result
|
|
[ "red:0|blonde|black:0", { 'blonde': 100 } ],
|
|
# - weights <= 1 with distribution of the leftover
|
|
[ "red:0.5|blonde|black:0.3|brown", { 'red': 0.5, 'blonde': 1.0, 'black': 0.3, 'brown': 1.0 } ],
|
|
# - weights > 1, unweightes should get default of 1
|
|
[ "red:2|blonde|black", { 'red': 50, 'blonde': 25, 'black': 25 } ],
|
|
# - ignore content of ()
|
|
[ "red:0.5|(blonde:1.3)", { 'red': 50, '(blonde:1.3)': 100 } ],
|
|
# - ignore content of ()
|
|
[ "red:0.5|(blonde:1.3):0.5", { 'red': 50, '(blonde:1.3)': 50 } ],
|
|
# - ignore content of []
|
|
[ "red:0.5|[stuff:1.3]", { 'red': 50, '[stuff:1.3]': 100 } ],
|
|
# - ignore content of <>
|
|
[ "red:0.5|<lora:1.0>", { 'red': 50, '<lora:1.0>': 100 } ],
|
|
# - simple list, 1 entry with lora with weights
|
|
[ "red|<lora:test:1.0>|black", { 'black': 33, 'red': 33, '<lora:test:1.0>': 33 } ],
|
|
# - simple list, 1 entry with loraand comma
|
|
[ "red|blonde <lora:test:1.0>|black", { 'black': 33, 'red': 33, 'blonde <lora:test:1.0>': 33 } ],
|
|
# - simple list, 1 entry with lora and comma
|
|
[ "red|blonde, <lora:test:1.0>|black", { 'black': 33, 'red': 33, 'blonde, <lora:test:1.0>': 33 } ],
|
|
# - simple list, 1 entry with lora and comma
|
|
[ "red|blonde, <lora:test:1.0>|black:2", { 'black': 50, 'red': 25, 'blonde, <lora:test:1.0>': 25 } ],
|
|
]
|
|
|
|
with open(fn, 'r', encoding='utf-8') as f:
|
|
src = f.read()
|
|
start = src.find('def ' + funcname)
|
|
if start == -1:
|
|
print('Function not found')
|
|
sys.exit(1)
|
|
# find next top-level def or class after start
|
|
next_def = src.find('\ndef ', start+1)
|
|
next_class = src.find('\nclass ', start+1)
|
|
end_candidates = [i for i in (next_def, next_class) if i != -1]
|
|
end = min(end_candidates) if end_candidates else len(src)
|
|
func_src = src[start:end]
|
|
|
|
exec(func_src, ns) # pylint: disable=exec-used
|
|
func = ns.get(funcname)
|
|
if func is None:
|
|
print('Failed to extract function')
|
|
sys.exit(1)
|
|
|
|
print('Running' , tries, 'isolated quick tests for ' + funcname + ':\n')
|
|
|
|
for t in tests:
|
|
print('Input:', t[0])
|
|
samples = [func(t[0]) for _ in range(tries)]
|
|
c = Counter(samples)
|
|
print(" Expected:", dict(t[1]))
|
|
print(" Samples:", dict(c))
|
|
|
|
# validation
|
|
expected_pct = t[1]
|
|
expected_keys = set(expected_pct.keys())
|
|
actual_keys = set(c.keys())
|
|
missing = expected_keys - actual_keys
|
|
unexpected = actual_keys - expected_keys
|
|
|
|
if missing or unexpected:
|
|
if missing:
|
|
print(" Missing: ", sorted(missing))
|
|
if unexpected:
|
|
print(" Unexpected: ", sorted(unexpected))
|
|
print(" Result: FAIL keys")
|
|
continue
|
|
|
|
failures = []
|
|
W = sum(expected_pct.values())
|
|
deviations = {}
|
|
for k, pct in expected_pct.items():
|
|
pct = pct / W # normalize
|
|
expected_count = tries * pct
|
|
actual_count = c.get(k, 0)
|
|
actual_pct = actual_count / tries
|
|
deviation_pct = abs(actual_pct - pct)
|
|
deviations[k] = deviation_pct
|
|
if deviation_pct > tolerance_pct / 100.0:
|
|
failures.append(
|
|
f"{k}: expected {pct:.1f}%, got {actual_pct:.1f}% "
|
|
f"({actual_count}/{tries})"
|
|
)
|
|
|
|
print(" Deviations:", {k: f"{v*100:.2f}%" for k, v in deviations.items()})
|
|
if failures:
|
|
for line in failures:
|
|
print(" " + line)
|
|
print(" Result: FAIL distribution")
|
|
else:
|
|
print(" Result: PASS")
|