# Copyright (c) 2025, 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.
import torch
[docs]
def compute_loss_over_dataset(
model: torch.nn.Module | None = None,
loss: torch.nn.modules.loss._Loss | None = None,
dataloader: torch.utils.data.DataLoader | None = None,
device: torch.device | None = None,
) -> float:
"""
Compute the average loss over a dataset.
Args:
model: The model to use for predictions.
loss: The loss function to use.
dataloader: The dataloader for the dataset.
device: The device to use for computation.
Returns:
The average loss over the dataset.
"""
# Check if all required arguments are provided
if model is None or loss is None or dataloader is None:
raise ValueError("All arguments (model, loss, dataloader) must be provided.")
model.eval()
epoch_loss = 0
with torch.no_grad():
# Iterate over the dataloader
for x, y in dataloader:
# Move data to the specified device
x, y = x.to(device), y.to(device)
# Make predictions
pred = model(x)
# Compute the loss
ls = loss(pred, y)
# Accumulate the loss
epoch_loss += ls.item() * x.size(0)
# Return the average loss
return epoch_loss / len(dataloader.dataset)