⚡️ Speed up function state_dict_prefix_replace by 127%

[aseembits93]

📄 127% (1.27x) speedup for state_dict_prefix_replace in comfy/utils.py

⏱️ Runtime : 1.61 millisecond → 710 microseconds (best of 398 runs)

📝 Explanation and details

Here's an optimized version of your Python function. The primary changes are to minimize the creation of intermediate lists and to use dictionary comprehensions for more efficient data manipulation.

Changes and Optimizations

1. Avoid Unneeded List Creation:

   • Instead of mapping and filtering the keys in a separate step (map and filter), it is done directly in the list comprehension.

2. Dictionary Comprehension:

   • By directly assigning out to {} or state_dict, it forgoes unnecessary intermediate steps in the conditional initialization.

3. In-Loop Item Assignment.

   • Keys to be replaced and corresponding operations are now handled directly within loops, reducing intermediate variable assignments.

This rewritten function should perform better, especially with large dictionaries, due to reduced overhead from list operations and more efficient key manipulation.

✅ Correctness verification report:

Test	Status
⚙️ Existing Unit Tests	🔘 None Found
🌀 Generated Regression Tests	✅ 29 Passed
⏪ Replay Tests	🔘 None Found
🔎 Concolic Coverage Tests	🔘 None Found
📊 Tests Coverage	100.0%

🌀 Generated Regression Tests Details

import logging

# imports
import pytest  # used for our unit tests
import torch
from _codecs import encode
from comfy.utils import state_dict_prefix_replace
from numpy import dtype
from numpy.core.multiarray import scalar
from numpy.dtypes import Float64DType

# unit tests

def test_basic_single_prefix_replacement():
    state_dict = {"layer1.weight": 1, "layer1.bias": 2}
    replace_prefix = {"layer1.": "layer2."}
    expected_output = {"layer2.weight": 1, "layer2.bias": 2}
    codeflash_output = state_dict_prefix_replace(state_dict.copy(), replace_prefix)

def test_basic_multiple_prefix_replacement():
    state_dict = {"layer1.weight": 1, "layer2.bias": 2}
    replace_prefix = {"layer1.": "new_layer1.", "layer2.": "new_layer2."}
    expected_output = {"new_layer1.weight": 1, "new_layer2.bias": 2}
    codeflash_output = state_dict_prefix_replace(state_dict.copy(), replace_prefix)

def test_no_replacement_needed():
    state_dict = {"layer1.weight": 1, "layer1.bias": 2}
    replace_prefix = {"layer2.": "layer3."}
    expected_output = {"layer1.weight": 1, "layer1.bias": 2}
    codeflash_output = state_dict_prefix_replace(state_dict.copy(), replace_prefix)

def test_empty_prefix_dictionary():
    state_dict = {"layer1.weight": 1, "layer1.bias": 2}
    replace_prefix = {}
    expected_output = {"layer1.weight": 1, "layer1.bias": 2}
    codeflash_output = state_dict_prefix_replace(state_dict.copy(), replace_prefix)

def test_empty_state_dictionary():
    state_dict = {}
    replace_prefix = {"layer1.": "layer2."}
    expected_output = {}
    codeflash_output = state_dict_prefix_replace(state_dict.copy(), replace_prefix)

def test_prefix_overlap():
    state_dict = {"layer1.weight": 1, "layer1.bias": 2, "layer1.weight_extra": 3}
    replace_prefix = {"layer1.": "layer2.", "layer1.weight": "layer3.weight"}
    expected_output = {"layer2.weight": 1, "layer2.bias": 2, "layer2.weight_extra": 3}
    codeflash_output = state_dict_prefix_replace(state_dict.copy(), replace_prefix)

def test_filter_keys_enabled_basic_replacement():
    state_dict = {"layer1.weight": 1, "layer1.bias": 2}
    replace_prefix = {"layer1.": "layer2."}
    expected_output = {"layer2.weight": 1, "layer2.bias": 2}
    codeflash_output = state_dict_prefix_replace(state_dict.copy(), replace_prefix, filter_keys=True)

def test_filter_keys_enabled_no_matching_prefix():
    state_dict = {"layer1.weight": 1, "layer1.bias": 2}
    replace_prefix = {"layer2.": "layer3."}
    expected_output = {}
    codeflash_output = state_dict_prefix_replace(state_dict.copy(), replace_prefix, filter_keys=True)

def test_large_state_dictionary():
    state_dict = {f"layer{i}.weight": i for i in range(1000)}
    replace_prefix = {"layer": "new_layer"}
    expected_output = {f"new_layer{i}.weight": i for i in range(1000)}
    codeflash_output = state_dict_prefix_replace(state_dict.copy(), replace_prefix)

def test_complex_nested_prefixes():
    state_dict = {f"block{i}.layer{j}.weight": i*j for i in range(10) for j in range(10)}
    replace_prefix = {"block": "module", "layer": "submodule"}
    expected_output = {f"module{i}.submodule{j}.weight": i*j for i in range(10) for j in range(10)}
    codeflash_output = state_dict_prefix_replace(state_dict.copy(), replace_prefix)

def test_partial_matching_prefixes():
    state_dict = {"layer1.weight": 1, "layer1.bias": 2, "layer2.weight": 3}
    replace_prefix = {"layer1.": "layerA.", "layer2.": "layerB."}
    expected_output = {"layerA.weight": 1, "layerA.bias": 2, "layerB.weight": 3}
    codeflash_output = state_dict_prefix_replace(state_dict.copy(), replace_prefix)

def test_overlapping_prefixes():
    state_dict = {"layer1.weight": 1, "layer1.bias": 2, "layer1.weight_extra": 3}
    replace_prefix = {"layer1.": "layer2.", "layer1.weight": "layer3.weight"}
    expected_output = {"layer2.weight": 1, "layer2.bias": 2, "layer2.weight_extra": 3}
    codeflash_output = state_dict_prefix_replace(state_dict.copy(), replace_prefix)

def test_nested_dictionaries():
    state_dict = {"block1.layer1.weight": 1, "block1.layer1.bias": 2, "block2.layer1.weight": 3}
    replace_prefix = {"block1.layer1.": "blockA.layerA.", "block2.layer1.": "blockB.layerA."}
    expected_output = {"blockA.layerA.weight": 1, "blockA.layerA.bias": 2, "blockB.layerA.weight": 3}
    codeflash_output = state_dict_prefix_replace(state_dict.copy(), replace_prefix)
# codeflash_output is used to check that the output of the original code is the same as that of the optimized code.

import logging

# imports
import pytest  # used for our unit tests
import torch
from _codecs import encode
from comfy.utils import state_dict_prefix_replace
from numpy import dtype
from numpy.core.multiarray import scalar
from numpy.dtypes import Float64DType

# unit tests

# Basic Functionality Tests
def test_single_prefix_replacement():
    state_dict = {'a1': 1, 'a2': 2}
    replace_prefix = {'a': 'b'}
    codeflash_output = state_dict_prefix_replace(state_dict, replace_prefix, filter_keys=False); result = codeflash_output

def test_multiple_prefix_replacement():
    state_dict = {'a1': 1, 'b2': 2}
    replace_prefix = {'a': 'c', 'b': 'd'}
    codeflash_output = state_dict_prefix_replace(state_dict, replace_prefix, filter_keys=False); result = codeflash_output

# Edge Case Tests
def test_empty_state_dict():
    state_dict = {}
    replace_prefix = {'a': 'b'}
    codeflash_output = state_dict_prefix_replace(state_dict, replace_prefix, filter_keys=False); result = codeflash_output

def test_empty_replace_prefix():
    state_dict = {'a1': 1, 'a2': 2}
    replace_prefix = {}
    codeflash_output = state_dict_prefix_replace(state_dict, replace_prefix, filter_keys=False); result = codeflash_output

def test_prefix_not_in_state_dict():
    state_dict = {'a1': 1, 'a2': 2}
    replace_prefix = {'b': 'c'}
    codeflash_output = state_dict_prefix_replace(state_dict, replace_prefix, filter_keys=False); result = codeflash_output

# Filtering Keys Tests
def test_filter_keys_true_with_matching_prefix():
    state_dict = {'a1': 1, 'a2': 2}
    replace_prefix = {'a': 'b'}
    codeflash_output = state_dict_prefix_replace(state_dict, replace_prefix, filter_keys=True); result = codeflash_output

def test_filter_keys_true_with_non_matching_prefix():
    state_dict = {'a1': 1, 'a2': 2}
    replace_prefix = {'b': 'c'}
    codeflash_output = state_dict_prefix_replace(state_dict, replace_prefix, filter_keys=True); result = codeflash_output

# Complex Scenarios Tests
def test_nested_prefix_replacement():
    state_dict = {'a1': 1, 'a2': 2, 'b3': 3}
    replace_prefix = {'a': 'b', 'b': 'c'}
    codeflash_output = state_dict_prefix_replace(state_dict, replace_prefix, filter_keys=False); result = codeflash_output

def test_overlapping_prefixes():
    state_dict = {'a1': 1, 'ab2': 2}
    replace_prefix = {'a': 'b', 'ab': 'c'}
    codeflash_output = state_dict_prefix_replace(state_dict, replace_prefix, filter_keys=False); result = codeflash_output

# Large Scale Test Cases
def test_large_state_dict():
    state_dict = {f'a{i}': i for i in range(1000)}
    replace_prefix = {'a': 'b'}
    codeflash_output = state_dict_prefix_replace(state_dict, replace_prefix, filter_keys=False); result = codeflash_output
    expected = {f'b{i}': i for i in range(1000)}

def test_large_replace_prefix_dict():
    state_dict = {'a1': 1, 'b2': 2}
    replace_prefix = {f'a{i}': f'b{i}' for i in range(1000)}
    codeflash_output = state_dict_prefix_replace(state_dict, replace_prefix, filter_keys=False); result = codeflash_output

# Special Characters in Prefixes Tests
def test_special_characters_in_state_dict_keys():
    state_dict = {'a!1': 1, 'a@2': 2}
    replace_prefix = {'a': 'b'}
    codeflash_output = state_dict_prefix_replace(state_dict, replace_prefix, filter_keys=False); result = codeflash_output

def test_special_characters_in_replace_prefix():
    state_dict = {'a1': 1, 'a2': 2}
    replace_prefix = {'a': 'b!', 'a': 'b@'}
    codeflash_output = state_dict_prefix_replace(state_dict, replace_prefix, filter_keys=False); result = codeflash_output

# Mixed Data Types Tests
def test_mixed_data_types():
    state_dict = {'a1': 1, 'a2': 'value', 'a3': [1, 2, 3]}
    replace_prefix = {'a': 'b'}
    codeflash_output = state_dict_prefix_replace(state_dict, replace_prefix, filter_keys=False); result = codeflash_output

# Prefixes with Different Lengths Tests
def test_shorter_prefix_replacement():
    state_dict = {'abc1': 1, 'abc2': 2}
    replace_prefix = {'abc': 'a'}
    codeflash_output = state_dict_prefix_replace(state_dict, replace_prefix, filter_keys=False); result = codeflash_output

def test_longer_prefix_replacement():
    state_dict = {'a1': 1, 'a2': 2}
    replace_prefix = {'a': 'abc'}
    codeflash_output = state_dict_prefix_replace(state_dict, replace_prefix, filter_keys=False); result = codeflash_output
# codeflash_output is used to check that the output of the original code is the same as that of the optimized code.

To edit these changes git checkout codeflash/optimize-state_dict_prefix_replace-m9js6ztz and push.

[ltdrdata]

This code should be skipped if filter_keys is false.

[aseembits93]

double confirming this as the for rp in replace_prefix: loop is outside the if-else block

[aseembits93]

I have made the necessary changes, it's ready for review @ltdrdata

[aseembits93]

@comfyanonymous ready for review