googlenet

GoogLeNet

GoogLeNet: Going Deeper with Convolutions

Introduction

GoogLeNet is a new deep learning structure proposed by Christian Szegedy in 2014. Prior to this, AlexNet, VGG and other structures achieved better training effects by increasing the depth (number of layers) of the network, but the increase in the number of layers It will bring many negative effects, such as overfit, gradient disappearance, gradient explosion, etc. The proposal of inception improves the training results from another perspective: it can use computing resources more efficiently, and can extract more features under the same amount of computing, thereby improving the training results.[1]

Figure 1. Architecture of GoogLeNet [1]

Requirements

mindspore	ascend driver	firmware	cann toolkit/kernel
2.3.1	24.1.RC2	7.3.0.1.231	8.0.RC2.beta1

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

• Distributed 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 NPU devices
msrun --bind_core=True --worker_num 8 python train.py --config configs/googlenet/googlenet_ascend.yaml --data_dir /path/to/imagenet

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.

• Standalone Training

If you want to train or finetune the model on a smaller dataset without distributed training, please run:

# standalone training on single NPU device
python train.py --config configs/googlenet/googlenet_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/googlenet/googlenet_ascend.yaml --data_dir /path/to/imagenet --ckpt_path /path/to/ckpt

Performance

Our reproduced model performance on ImageNet-1K is reported as follows.

Experiments are tested on ascend 910* with mindspore 2.3.1 graph mode.

model name	params(M)	cards	batch size	resolution	jit level	graph compile	ms/step	img/s	acc@top1	acc@top5	recipe	weight
googlenet	6.99	8	32	224x224	O2	113s	23.5	10893.62	72.89	90.89	yaml	weights

Experiments are tested on ascend 910 with mindspore 2.3.1 graph mode.

model name	params(M)	cards	batch size	resolution	jit level	graph compile	ms/step	img/s	acc@top1	acc@top5	recipe	weight
googlenet	6.99	8	32	224x224	O2	72s	21.40	11962.62	72.68	90.89	yaml	weights

Notes

• top-1 and top-5: Accuracy reported on the validation set of ImageNet-1K.

References

[1] Szegedy C, Liu W, Jia Y, et al. Going deeper with convolutions[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2015: 1-9.