itslukeypookie commited on
Commit
40771d5
·
verified ·
1 Parent(s): df2cf07

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +12 -62
README.md CHANGED
@@ -1,62 +1,12 @@
1
- # Train CIFAR10 with PyTorch
2
-
3
- I'm playing with [PyTorch](http://pytorch.org/) on the CIFAR10 dataset.
4
-
5
- ## Prerequisites
6
- - Python 3.6+
7
- - PyTorch 1.0+
8
-
9
- - Install the dependencies from `requirements.txt` by running `pip install -r requirements.txt`
10
- - Update the `requirements.txt` with `pip list --format=freeze > requirements.txt`
11
-
12
- ## Dependencies for Weights and Biases
13
- You must install weights and biases using the command below:
14
- `pip install wandb`
15
-
16
- Afterwards, you must run this command to login into weights and biases. A GitHub account can be used to login:
17
- `wandb login`
18
-
19
- Paste the API key into terminal.
20
-
21
- Additionally, change this line in `main.py`:
22
- `wandb.init(entity="balica15", project="tutorial")`
23
-
24
- Where "balica15" is replaced by your username that you used to login with into Weights and Biases.
25
-
26
- ## Training
27
- ```
28
- # Start training with:
29
- python main.py
30
-
31
- # You can manually resume the training with:
32
- python main.py --resume --lr=0.01
33
- ```
34
-
35
- ## Features to Add
36
- | Name | Feature |
37
- | ----------------------------- | ----------- |
38
- | Evelyn Atkins and Ethan White | Input and Error Protection |
39
- | Keiane Balicanta | TorchVision Model Dropdown |
40
- | Henry Conde | Weights and Biases API |
41
- | Matthew Gerace | Iteration and Batch Size Sliders |
42
- | Luke Wilkins | Image Classification |
43
-
44
-
45
- ## Accuracy
46
- | Model | Acc. |
47
- | ----------------- | ----------- |
48
- | [VGG16](https://arxiv.org/abs/1409.1556) | 92.64% |
49
- | [ResNet18](https://arxiv.org/abs/1512.03385) | 93.02% |
50
- | [ResNet50](https://arxiv.org/abs/1512.03385) | 93.62% |
51
- | [ResNet101](https://arxiv.org/abs/1512.03385) | 93.75% |
52
- | [RegNetX_200MF](https://arxiv.org/abs/2003.13678) | 94.24% |
53
- | [RegNetY_400MF](https://arxiv.org/abs/2003.13678) | 94.29% |
54
- | [MobileNetV2](https://arxiv.org/abs/1801.04381) | 94.43% |
55
- | [ResNeXt29(32x4d)](https://arxiv.org/abs/1611.05431) | 94.73% |
56
- | [ResNeXt29(2x64d)](https://arxiv.org/abs/1611.05431) | 94.82% |
57
- | [SimpleDLA](https://arxiv.org/abs/1707.064) | 94.89% |
58
- | [DenseNet121](https://arxiv.org/abs/1608.06993) | 95.04% |
59
- | [PreActResNet18](https://arxiv.org/abs/1603.05027) | 95.11% |
60
- | [DPN92](https://arxiv.org/abs/1707.01629) | 95.16% |
61
- | [DLA](https://arxiv.org/pdf/1707.06484.pdf) | 95.47% |
62
-
 
1
+ ---
2
+ title: CIFAR-10 Model Training GUI
3
+ emoji: 🌍
4
+ colorFrom: red
5
+ colorTo: blue
6
+ sdk: gradio
7
+ sdk_version: 4.16.0
8
+ app_file: app.py
9
+ pinned: false
10
+ ---
11
+
12
+ Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference