Florence-2-FT-JP-TF

This model is a fine-tuned version of microsoft/Florence-2-base-ft on SakanaAI/JA-VG-VQA-500 dataset. It achieves the following results on the evaluation set:

  • Loss: 1.3826

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: 7e-06
  • train_batch_size: 1
  • eval_batch_size: 1
  • seed: 42
  • optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • num_epochs: 5

Training results

Training Loss Epoch Step Validation Loss
2.0416 1.0 1000 1.6576
1.3098 2.0 2000 1.4314
0.9951 3.0 3000 1.4185
0.8056 4.0 4000 1.3705
0.6916 5.0 5000 1.3826

Framework versions

  • Transformers 4.46.3
  • Pytorch 2.3.0
  • Datasets 2.20.0
  • Tokenizers 0.20.3
Downloads last month
79
Safetensors
Model size
271M params
Tensor type
F32
·
Inference Examples
Inference API (serverless) does not yet support model repos that contain custom code.

Model tree for aipib/Florence-2-FT-JP-TF

Finetuned
(15)
this model

Dataset used to train aipib/Florence-2-FT-JP-TF