edadaltocg/resnet50_supcon_cifar10

Model Card for resnet50supconcifar10

This model is a small resnet50 trained on cifar10.

• —Test Accuracy: 0.9518000000000001
• —License: MIT

How to Get Started with the Model

Use the code below to get started with the model.

import detectors
import timm

model = timm.create_model("resnet50_supcon_cifar10", pretrained=True)

Training Data

Training data is cifar10.

Training Hyperparameters

• —config: None

• —model: resnet50_supcon_cifar10

• —batch_size: 512

• —epochs: 501

• —lr: 0.5

• —warmup_epochs: 10

• —validation_frequency: 50

• —output_features_dim: 128

• —seed: 1

• —debug: False

• —dataset: cifar10

• —training_mode: supcon

Testing Data

Testing data is cifar10.

This model card was created by Eduardo Dadalto.