tapas

tapas copied to clipboard

Hi, I am trying to reproduce the results directly listed in the paper: https://arxiv.org/pdf/2103.12011.pdf  Specifically, I am trying to reproduce the answer to the question "which element is named...

apurvak

Pre-Training TAPAS on new data

1

Hello, I am trying to pre-train a tapas model on my data (a set of queries and a table). My first question is that is there any function that can...

naserahmadi

TAPAS unable to use weak supervision labels to finetune

1

I am trying to Fine-tune the pretrained TAPAS WTQ model on a custom dataset. I have used both Hugging face Pytorch code. My dataset has majority of samples with arithmetic...

shabbie

run the pipeline locally only to get empty floders

2

Thanks for releasing the code. When I run the pipeline locally, only to get three empty folders like "beam-temp-test.tfrecord-3e99843c1d3411ec8a4cfa163ee22da8". And the logs shows "I0925 18:54:12.767069 139992182503232 number_annotation_utils.py:149] Can't consolidate types:...

airsyuan

Hi Niels!

3

Hi Niels! In the case of WTQ we have some special logic that tries to find the answer text in the table or that populates the `float_value` field if the...

ManasiPat

Hyperparameter setting for Table Retrieval Model

1

Dear developers, Could you please share the hyper-parameters used for the table retrieval model to replicate the results [here](https://github.com/google-research/tapas/blob/master/DENSE_TABLE_RETRIEVER.md)? In the paper, you mentioned that you used warmup ratio 0.2...

Doreenruirui

Showing cell selection / snaphot of reference answer

2

Hi Team , Thanks for releasing TAPAS , really appreciate your efforts towards the new way of semantic parsing using NLP .I ran your colab demo and it showed quite...

ajay01994

Due to lack of GPU memory，I set batch_size to 32 and gradient_accumulation_steps to 32 when fine-tunning in SQA. The scripts are as follows. ``` SQA_run_cmd="\ CUDA_VISIBLE_DEVICES=1 python ../tapas/run_task_main.py \ --task="SQA"...

Facico

Thanks for submitting the notebook about the use of retrieval models on tables (DTR). I'm wondering if is it possible to make predictions with our own tables instead of those...

bpquestion

WTQ TableQuestionAnswering Task: answer_coordinates for weakly supervision task

1

I am using the pytorch hugginface model (https://huggingface.co/transformers/model_doc/tapas.html) for tablequestion answering task. My data has only the final answer for supervision and no answer co-ordinates or aggregation labels. I have...

ManasiPat