# Copyright (c) 2023, NVIDIA CORPORATION. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from typing import Dict
from .config import ExchangeFormat
[docs]def numerical_params_diff(original: Dict, new: Dict) -> Dict:
"""Calculates the numerical parameter difference.
Args:
original: A dict of numerical values.
new: A dict of numerical values.
Returns:
A dict with common keys that exist in both original dict and new dict,
values are the difference between original and new.
"""
diff_dict = {}
for k in original:
if k not in new:
continue
if isinstance(new[k], list) and isinstance(original[k], list):
diff = [new[k][i] - original[k][i] for i in range(len(new[k]))]
else:
diff = new[k] - original[k]
diff_dict[k] = diff
if diff_dict == {}:
raise RuntimeError("no common keys between original and new dict, parameters difference are empty.")
return diff_dict
DIFF_FUNCS = {ExchangeFormat.PYTORCH: numerical_params_diff, ExchangeFormat.NUMPY: numerical_params_diff}