Niall Taylor

I would also love to see this

Hi, So if you want a potential fix that goes around the "load_plm" function from OpenPrompt, you can load each component in separately and then merge: Actually I have one...

Fixed it ... Afterall I stumbled upon the subtle change required for the config: Honestly thought I had tried this several times, so perhaps it was fatigue. But the following...

Hi @JohnGiorgi, So I did have to make a few changes - in line with the guidance found here: https://github.com/allenai/allennlp/discussions/4933. Whilst I seemed to have been successful in modifying the...

Hi @JohnGiorgi . Sorry I ended up so quiet on this, got swamped with other things... I am still planning to find a day to action on this - am...

Sorry - just found this; https://colab.research.google.com/drive/1uAhgAdc8Qr42UKYDlgUv0f7W1-gAFwGo?usp=sharing I will see how this works and once satisfied will close issue

Having looked through the provided notebook - https://colab.research.google.com/drive/1uAhgAdc8Qr42UKYDlgUv0f7W1-gAFwGo?usp=sharing, it appears to show how you can load in different delta models as "serving multi-tasks" but it doesn't show how you can...

Hi Betty, Thanks for the super reply. Very helpful. I will try to re run some of my training with the same parameters as you have described. For reasons of...

Hi again, Sorry to email again but do you remember what batch_size you were using? Or did you let FARM decide that for you? Best, Niall ________________________________ From: Betty van...

Perfect - thanks! So with the gradient accumulation, an effective batch_size of 200 I guess? As the optimizers only update parameters every 10 batches? Best, Niall ________________________________ From: Betty van...