SqueezeNet
SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size
Introduction
SqueezeNet is a smaller CNN architectures which is comprised mainly of Fire modules and it achieves AlexNet-level
accuracy on ImageNet with 50x fewer parameters. SqueezeNet can offer at least three advantages: (1) Smaller CNNs require
less communication across servers during distributed training. (2) Smaller CNNs require less bandwidth to export a new
model from the cloud to an autonomous car. (3) Smaller CNNs are more feasible to deploy on FPGAs and other hardware with
limited memory. Additionally, with model compression techniques, SqueezeNet is able to be compressed to less than
0.5MB (510× smaller than AlexNet). Blow is macroarchitectural view of SqueezeNet architecture. Left: SqueezeNet ;
Middle: SqueezeNet with simple bypass; Right: SqueezeNet with complex bypass.
Figure 1. Architecture of SqueezeNet [1]
Results
Our reproduced model performance on ImageNet-1K is reported as follows.
Model |
Context |
Top-1 (%) |
Top-5 (%) |
Params (M) |
Recipe |
Download |
squeezenet1_0 |
D910x8-G |
58.67 |
80.61 |
1.25 |
yaml |
weights |
squeezenet1_0 |
GPUx8-G |
58.83 |
81.08 |
1.25 |
yaml |
weights |
squeezenet1_1 |
D910x8-G |
58.44 |
80.84 |
1.24 |
yaml |
weights |
squeezenet1_1 |
GPUx8-G |
59.18 |
81.41 |
1.24 |
yaml |
weights |
Notes
- Context: Training context denoted as {device}x{pieces}-{MS mode}, where mindspore mode can be G - graph mode or F - pynative mode with ms function. For example, D910x8-G is for training on 8 pieces of Ascend 910 NPU using graph mode.
- Top-1 and Top-5: Accuracy reported on the validation set of ImageNet-1K.
Quick Start
Preparation
Installation
Please refer to the installation instruction in MindCV.
Dataset Preparation
Please download the ImageNet-1K dataset for model training and validation.
Training
It is easy to reproduce the reported results with the pre-defined training recipe. For distributed training on multiple Ascend 910 devices, please run
# distributed training on multiple GPU/Ascend devices
mpirun -n 8 python train.py --config configs/squeezenet/squeezenet_1.0_ascend.yaml --data_dir /path/to/imagenet
Similarly, you can train the model on multiple GPU devices with the above mpirun
command.
For detailed illustration of all hyper-parameters, please refer to config.py.
Note: As the global batch size (batch_size x num_devices) is an important hyper-parameter, it is recommended to keep the global batch size unchanged for reproduction or adjust the learning rate linearly to a new global batch size.
If you want to train or finetune the model on a smaller dataset without distributed training, please run:
# standalone training on a CPU/GPU/Ascend device
python train.py --config configs/squeezenet/squeezenet_1.0_ascend.yaml --data_dir /path/to/dataset --distribute False
Validation
To validate the accuracy of the trained model, you can use validate.py
and parse the checkpoint path with --ckpt_path
.
python validate.py -c configs/squeezenet/squeezenet_1.0_ascend.yaml --data_dir /path/to/imagenet --ckpt_path /path/to/ckpt
Deployment
To deploy online inference services with the trained model efficiently, please refer to the deployment tutorial.
References
[1] Iandola F N, Han S, Moskewicz M W, et al. SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and< 0.5 MB model size[J]. arXiv preprint arXiv:1602.07360, 2016.