Download lib/core/evaluate.py from galaxygummybear/hrnet_pytorch: direct link, hf CLI and curl.
- Browser
- Download file 925 Bytes
-
https://huggingface.co/galaxygummybear/hrnet_pytorch/resolve/main/lib/core/evaluate.py
- Command line
-
hf download hf://galaxygummybear/hrnet_pytorch/lib/core/evaluate.py
-
curl -L -o evaluate.py https://huggingface.co/galaxygummybear/hrnet_pytorch/resolve/main/lib/core/evaluate.py
925 Bytes
| # ------------------------------------------------------------------------------ | |
| # Copyright (c) Microsoft | |
| # Licensed under the MIT License. | |
| # Written by Bin Xiao (Bin.Xiao@microsoft.com) | |
| # ------------------------------------------------------------------------------ | |
| from __future__ import absolute_import | |
| from __future__ import division | |
| from __future__ import print_function | |
| import torch | |
| def accuracy(output, target, topk=(1,)): | |
| """Computes the precision@k for the specified values of k""" | |
| with torch.no_grad(): | |
| maxk = max(topk) | |
| batch_size = target.size(0) | |
| _, pred = output.topk(maxk, 1, True, True) | |
| pred = pred.t() | |
| correct = pred.eq(target.reshape(1, -1).expand_as(pred)) | |
| res = [] | |
| for k in topk: | |
| correct_k = correct[:k].reshape(-1).float().sum(0, keepdim=True) | |
| res.append(correct_k.mul_(100.0 / batch_size)) | |
| return res | |