Update README.md

README.md

@@ -13,7 +13,9 @@ Finetuned https://huggingface.co/mesolitica/nanot5-base-malaysian-cased using 20
 - This model natively code switching.
 - This model maintain `\n`, `\t`, `\r` as it is.
 
-**Still in training session**, Wandb at https://wandb.ai/huseinzol05/nanot5-base-malaysian-cased-translation-v4?nw=nwuserhuseinzol05
+Thanks to https://x.com/oscarnazhan for the GPU access to train this model.
+
+**Still in training session**, Wandb at https://wandb.ai/huseinzol05/nanot5-base-malaysian-cased-translation-v4-multipack?nw=nwuserhuseinzol05
 
 ## Supported prefix
 
@@ -78,4 +80,8 @@ Output,
  ' Bayangkan PH dan menang PRU-14. Terdapat pelbagai pintu belakang. Akhirnya, Ismail Sabri naik. Itulah sebabnya saya tidak lagi bercakap tentang politik. Saya bersumpah sudah berputus asa.']
 ```
 
-Input text can be any languages that speak in Malaysia, as long you use proper prefix, it should be able to translate to target language.
+Input text can be any languages that speak in Malaysia, as long you use proper prefix, it should be able to translate to target language.
+
+## how to finetune your own dataset?
+
+We finetuned using T5 SDPA multipacking forked at https://github.com/mesolitica/t5-sdpa-multipack, super undocumented, but scripts from https://github.com/huggingface/transformers/tree/main/examples/pytorch/translation should work also.