| Model | Download | Download (with sample test data) | ONNX version | Opset version |
|---|---|---|---|---|
| CaffeNet | 238 MB | 244 MB | 1.1 | 3 |
| CaffeNet | 238 MB | 244 MB | 1.1.2 | 6 |
| CaffeNet | 238 MB | 244 MB | 1.2 | 7 |
| CaffeNet | 238 MB | 244 MB | 1.3 | 8 |
| CaffeNet | 238 MB | 244 MB | 1.4 | 9 |
CaffeNet a variant of AlexNet. AlexNet is the name of a convolutional neural network for classification, which competed in the ImageNet Large Scale Visual Recognition Challenge in 2012.
Differences:
- not training with the relighting data-augmentation;
- the order of pooling and normalization layers is switched (in CaffeNet, pooling is done before normalization).
ImageNet Classification with Deep Convolutional Neural Networks
Caffe BVLC CaffeNet ==> Caffe2 CaffeNet ==> ONNX CaffeNet
data_0: float[1, 3, 224, 224]
prob_1: float[1, 1000]
random generated sampe test data:
- test_data_set_0
- test_data_set_1
- test_data_set_2
- test_data_set_3
- test_data_set_4
- test_data_set_5
This model is snapshot of iteration 310,000. The best validation performance during training was iteration 313,000 with validation accuracy 57.412% and loss 1.82328. This model obtains a top-1 accuracy 57.4% and a top-5 accuracy 80.4% on the validation set, using just the center crop. (Using the average of 10 crops, (4 + 1 center) * 2 mirror, should obtain a bit higher accuracy still.)