Griffin Adams

Is it possible to share the fine-tuned MedNLI classifier with a simple code snippet for how to perform inference given a text and premise pair?

Running inference on GPU

3

## 🐛 Bug ``` from lightning_transformers.task.nlp.token_classification import TokenClassificationTransformer model = TokenClassificationTransformer.load_from_checkpoint(ckpt_fn).to('cuda:0') with torch.no_grad(): model.hf_predict('this is a test sentence.') ``` Running this, you get a device mismatch since it puts the...

Hi Hayden - thanks for making this repo (it's very helpful). I'm trying to re-create the best Extractive models on my own machines so I can modify. Can you help...

The MayoSRS and UMNSRS_similarity datasets mostly contain phrases. Did you use mean pooling to get the results you reported or some other pooling mechanism for ngrams longer than 1? Thanks

Hi, I changed my preprocessing code so that some iterations take a fairly long time (1-2 minutes). I think this irregular speed is causing everything to hang at the end...

Code to go from clusters to transformed document

2

Is there code to go from "The dog wagged its tail" --> "The dog wagged the dog's tail"? Basically, go from clusters to head prediction and then replacement of mentions...

Ranking Loss Question

1

Hi - Thanks for the great code. I've been trying to re-implement BRIO in my HuggingFace fork, but unable to get it to work. I'm curious what this line in...

Pretrained model for MedNLI

1

Hi - are there weights for a pre-trained MedNLI model?

Using your own data

2

Hey David - Thanks for this repo (it's very legible and easy to read - I also like PLightning!) If I want to very quickly run predict a label for...

Error calling `_initialize_deepspeed_train`: attempted relative import beyond top-level package

1

I am using pytorch-lightning with deepspeed and getting the following error. Any ideas on how to fix? Thanks a lot! **Relevant Versions** `deepspeed==0.8.1` `pytorch-lightning==1.9.4` ``` ValueError: attempted relative import beyond...