TAPAS base model fine tuned on WikiTable Questions (WTQ) This model has 2 versions which can be used. The default version corresponds to the tapas wtq wikisql sqa inter masklm base reset checkpoint of the original Github repository. This model was pre trained on MLM and an additional step which the authors call intermediate pre training, and then fine tuned in a chain on SQA, WikiSQL and finally WTQ. It uses relative position embeddings (i.e. resetting the position index at every cell of the table). The other (non default) version which can be used is: no reset , which corresponds to tapas wtq wikisql sqa inter masklm base (intermediate pre training, absolute position embeddings). Disclaimer: The team releasing TAPAS did not write a model card for this model so this model card has been written by the Hugging Face team and contributors. Results Size Reset Dev Accuracy Link LARGE noreset 0.5062 tapas large finetuned wtq (with absolute pos embeddings) LARGE reset 0.5097 tapas large finetuned wtq BASE noreset 0.4525 tapas base finetuned wtq (with absolute pos embeddings) BASE reset 0.4638 tapas base finetuned wtq MEDIUM noreset 0.4324 tapas medium finetuned wtq (with absolute pos embed…
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy