https://github.com/kuixu/pytorch_online_plotter
Online meter ploter for pytorch. Real time ploting Accuracy, Loss, mAP, AUC, Confusion Matrix
https://github.com/kuixu/pytorch_online_plotter
confusion-matrix lossless mean-average-precision online plot pytorch real-time
Last synced: 11 months ago
JSON representation
Online meter ploter for pytorch. Real time ploting Accuracy, Loss, mAP, AUC, Confusion Matrix
- Host: GitHub
- URL: https://github.com/kuixu/pytorch_online_plotter
- Owner: kuixu
- Created: 2017-12-20T09:23:32.000Z (over 8 years ago)
- Default Branch: master
- Last Pushed: 2018-04-05T13:01:41.000Z (about 8 years ago)
- Last Synced: 2025-03-21T15:42:04.342Z (over 1 year ago)
- Topics: confusion-matrix, lossless, mean-average-precision, online, plot, pytorch, real-time
- Language: Python
- Size: 88.9 KB
- Stars: 17
- Watchers: 1
- Forks: 4
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# pytorch_online_ploter
Online meter ploter for pytorch. Real time ploting Accuracy, Loss, mAP, AUC, Confusion Matrix
## Start Visdom on a server
```bash
python -m visdom.server
# python -m visdom.server -port 9999 # to specify port to ex, 9999
```
## Basic MNIST Example
```bash
pip install -r requirements.txt
python mnist.py
# CUDA_VISIBLE_DEVICES=1 python mnist.py # to specify GPU id to ex. 1
# if you run Visdom with 9999
# python mnist.py --port 9999
```
## Multi-meter
Easy to plot multi-meter with just one-line code:
### Plotting Accuracy, mAP
```python
mlog.update_meter(output, target, meters={'accuracy', 'map'})
```
### Plotting Loss Curve
```python
# NLL Loss
nll_loss = F.nll_loss(output, target)
mlog.update_loss(nll_loss, meter='nll_loss')
# Cross Entropy Loss
ce_loss = F.cross_entropy(output, target)
mlog.update_loss(ce_loss, meter='ce_loss')
```
## Remote Plotting
```python
mlog = MeterLogger(server="Server's IP", nclass=10, title="mnist")
```
## Real time plotting
Open your browser with URL:
```bash
http://localhost:9999
# or
http://ip:9999
```
