clip-vit-bert-coco / README.md
christti's picture
End of training
465fbee verified
|
raw
history blame
1.21 kB
metadata
tags:
  - generated_from_trainer
datasets:
  - ydshieh/coco_dataset_script
model-index:
  - name: clip-vit-bert-coco
    results: []

clip-vit-bert-coco

This model was trained from scratch on the ydshieh/coco_dataset_script 2017 dataset. It achieves the following results on the evaluation set:

  • eval_loss: 4.2294
  • eval_runtime: 287.7769
  • eval_samples_per_second: 86.921
  • eval_steps_per_second: 0.681
  • step: 0

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: 5e-05
  • train_batch_size: 128
  • eval_batch_size: 128
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 5.0
  • mixed_precision_training: Native AMP

Framework versions

  • Transformers 4.39.0.dev0
  • Pytorch 2.2.1
  • Datasets 2.18.0
  • Tokenizers 0.15.2