The PyPi package torch_utils changed maintainers
The previous maintainer sfzhou gave the pacakge name to AYO.
Versions 0.X.X are the legacy package published until 2019
All version from 1.0.0 are a totally different project.
Pin you version to the legacy versions if you need it.
torch-utils 
Common Utils for PyTorch.
Installation
Need Python 3.6+.
pip install torch-utils
Usage
- Accuracy
import torch_utils
# ...
top_1, top_5 = torch_utils.accuracy(output=..., target=..., top_k=(1, 5))
- Meter
import torch_utils
loss_meter = torch_utils.AverageMeter(name='Meter', length=10)
loss_meter.update(val=...)
print(loss_meter.avg, loss_meter.val)
print(loss_meter)
#> Test 0.00 (0.00)
progress_meter = torch_utils.ProgressMeter(total_steps=100, total_epochs=10)
progress_meter.update(step=10)
assert progress_meter.step == 10
assert progress_meter.ratio == 0.1
assert progress_meter.epoch == 1
print(progress_meter)
#> Step 10/100=10.0% (1/10)
2 maintainers
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
torch_utils-0.1.3.tar.gz
(7.7 kB
view details)
File details
Details for the file torch_utils-0.1.3.tar.gz.
File metadata
- Download URL: torch_utils-0.1.3.tar.gz
- Upload date:
- Size: 7.7 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via:
twine/7.0.0 CPython/3.14.7
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
c6e62333ee7276f8b09e6a8f89378b7f5ac916ee4ddc131707384707b18b65a6
|
|
| MD5 |
b258995be6f1228604bfc2aba5d549d0
|
|
| BLAKE2b-256 |
1d9cd578b644a1bc47de785758396e5cadc24342e0fc358b467023c3be3be590
|
