Yidan Sun

Running funetuning on GPU

2

Thanks for releasing your great work. I was wondering if there is a way to run the finetuning and zero-shot inference code on GPU rather than TPU? What king of...

I downloaded the S3D feature, it seems the features for every video has been mean/max pooled. can I get the features for each clips? (without pooling)

How to extract unified repersentation for object and text caption

1

Thankyou for releasing this great work. I read in your paper the the image, text, and object are representated with tokens from a unified vocabulary. I was wondering how can...

Error with batch inference

3

Assertion error or out of bounds error when doing inference with batch size greater than 1: ../aten/src/ATen/native/cuda/Indexing.cu:1093: indexSelectSmallIndex: block: [15,0,0], thread: [64,0,0] Assertion `srcIndex < srcSelectDimSize` failed. ../aten/src/ATen/native/cuda/Indexing.cu:1093: indexSelectSmallIndex: block:...

Zero-shot inference my own images

5

Thank you for this amazing work. I was wondering, is there a way to use CPT on my own data? There is a few point in the .json file and...

train config for retrieval

1

Hi, amazing work. When you have the time, could you release the config file for finetuning on MSRVTT with Xretrieval.py Thanks.

Hi, great work, But I run into a bit a problem while using drop dtw with manually set drop cost. Namely I have two sequences for different length, I found...