No adapter with name 'sentiment/hinglish-twitter-sentiment@nirantk' was found in the adapter index.
whisere opened this issue · 5 comments
Hello, will you be able to help with this issue:
No adapter with name 'sentiment/hinglish-twitter-sentiment@nirantk' was found in the adapter index.
inception-project/inception-external-recommender#28
Thanks!
Hi @whisere,
I tried to reproduce the error using the sample code from our website, i.e.:
model = BertModelWithHeads.from_pretrained("bert-base-multilingual-uncased")
model.load_adapter("sentiment/hinglish-twitter-sentiment@nirantk", "text_task", config="pfeiffer")but this runs without problems, so there doesn't seem to be a problem on our side. I'm not quite familiar with Inception but I'll try to see if there maybe is a problem with the integration.
Thank you very much @calpt I will look into Inception too, perhaps something to do with the other options, "text_task" and "pfeiffer" but I can't find them in inception, will keep looking. Thanks!
It seems this has been fixed at the Inception side. I will keep testing. How do we know the labels used in the Adapters? I can only find this adapter states they use 'postive, netural, negative' as tags in their description https://adapterhub.ml/explore/sentiment/hinglish-twitter-sentiment/ but not sure if the other Sentiment Analysis adapters also use 'netural' or have other tags? Thank you!
Great to hear. The labels of an adapter are determined by the task they're trained on, so looking up the respective task should give information about the label classes, e.g.:
SST-2: https://huggingface.co/datasets/glue#sst2
IMDb: https://huggingface.co/datasets/imdb
Rotten Tomatoes: https://huggingface.co/datasets/rotten_tomatoes
Unfortunately, not all adapters provide this information within their description currently, we're trying to improve that in the future.
Thank you so much!! That's very helpful!