Edit model card

resnet-101-finetuned_resnet101-cnn-autotags

This model is a fine-tuned version of microsoft/resnet-101 on the imagefolder dataset. It achieves the following results on the evaluation set:

  • Loss: 0.2099
  • Accuracy: 0.9229

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 0.001
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 64
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Accuracy
2.0434 0.99 65 1.6045 0.4829
0.9013 1.99 130 0.6946 0.7648
0.7097 2.99 195 0.4928 0.8295
0.4386 3.99 260 0.3632 0.8610
0.4261 4.99 325 0.3269 0.8838
0.3181 5.99 390 0.2790 0.9010
0.2349 6.99 455 0.2377 0.9190
0.1615 7.99 520 0.2416 0.9114
0.1146 8.99 585 0.2162 0.9219
0.1254 9.99 650 0.2099 0.9229

Framework versions

  • Transformers 4.25.1
  • Pytorch 1.13.1+cu117
  • Datasets 2.11.0
  • Tokenizers 0.13.2
Downloads last month
9
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Evaluation results