readme change about batch size

README.md

@@ -65,7 +65,7 @@ The DETR model was trained on [SKU110K Dataset](https://github.com/eg4000/SKU110
 ## Training procedure
 ### Training
 
-The model was trained for 140 epochs on 1 RTX 4060 Ti GPU(Finetuning decoder only) with batch size of 8 and 70 epochs(finetuning the whole network) with batch size of 3 and accumulating gradients for 3 steps.
+The model was trained for 60 epochs on 1 RTX 4060 Ti GPU(Finetuning decoder only) with batch size of 1 and gradient_accumulation set to 8 and 60 epochs(finetuning the whole network) with batch size of 1 and accumulating gradients for 8 steps.
 
 ## Evaluation results