gagan3012
/

ViTGPT2I2A

Image-Text-to-Text

vision-encoder-decoder

image-captioning

Generated from Trainer

Model card Files Files and versions

ViTGPT2I2A / eval_results.json

gagan3012's picture

End of training

45a3289 about 4 years ago

160 Bytes

	{
	"epoch": 5.0,
	"eval_loss": 0.0708099901676178,
	"eval_runtime": 220.4842,
	"eval_samples_per_second": 35.15,
	"eval_steps_per_second": 8.79
	}