ResNet-18 vs ResNet-34

• CNN_Classifications

  12 6 6.4 Jupyter Notebook

  Convolutional Neural Networks for image recognition and classification

I have trained ResNet-18 and ResNet-34 from scratch using PyTorch on CIFAR-10 dataset. The validation accuracy I get for ResNet-18 is 84.01%, whereas for ResNet-34 is 82.43%. Is this a sign of ResNet-34 overfitting as compared to ResNet-18? Ideally, ResNet-34 should achieve a higher validation accuracy as compared to ResNet-18.

• pytorch-cifar

  3 5,662 0.0 Python

  95.47% on CIFAR10 with PyTorch

OP could also be underfitting. Some people have shown 93% accuracy on cifar10 using resnet18. See https://github.com/kuangliu/pytorch-cifar

• InfluxDB

  www.influxdata.com featured

  Power Real-Time Data Analytics at Scale. Get real-time insights from all types of time series data with InfluxDB. Ingest, query, and analyze billions of data points in real-time with unbounded cardinality.

  

Suggest a related project