Avinash Madasu

Finetuning time vis MSRVTT

3

Hi, How much times does it take to fine-tune on MSRVTT on 8 v100 GPUs? Thank you.

Datapath

5

Hi, Congrats on the amazing work. I downloaded data using bash scripts/download_tvr.sh $PATH_TO_STORAGE (I gave my own $PATH_TO_STORAGE). How can I mention this data path in the fine-tuning script both...

[feature] Add Sentence Classification tutorial on a custom dataset

1

Hi, I have found that current examples are on existing pytorch datasets. There are no examples on a custom dataset. I have a tutorial of sentence classification on a custom...

Entity F1 without KB

1

Hi, Congratulations for the amazing paper. I have a doubt on how to measure Entity F1 Score without KB input. I have a prediction file (system responses) and ground truth...

Added Sentence Classification tutorial

3

As discussed in #715 , I have added sentence classification tutorial using CNN on TREC dataset. cc @yf225 @soumith

Electra Vocabulary

1

Hi, I have a query about Electra's vocabulary. In contrast to other models where vocabulary words are clearly seen, why are words in Electra vocab ```[unused0], [unused1].....``` except for a...

Test data in vocabulary preparation

2

Hi, Congratulations on your amazing work. I have a doubt in vocabulary preparation in line 47 in utils.py. Testing data is also used in vocabulary preparation. However, testing data should...

Hi, Congratulations on your amazing work. I would like to know which script needs to be run for Blitzer dataset. Because BLSE_domain_all.py produces results for all the domains. Should I...

Custom dataset format

1

Hi, Congrats on the amazing work!! I plan to fine-tune BLIP for image captioning on a custom dataset. What is the input format of the files and changes required in...