danieljf24

The code using python3 has been also [released](https://github.com/danieljf24/dual_encoding/tree/python3). Please checkout the `python3` branch.

What is the version of python and pytorch you used?

Please refer to [here](https://github.com/danieljf24/simpleknn/blob/master/toydata/FeatureData/f1/id.feature.txt). The format of each line is an id followed by a feature vector. ps: We have already released our feature extraction [code](https://github.com/xuchaoxi/video-cnn-feat).

I guess it's the problem with your cuda or cudnn.

Sorry for the confusion. Indeed, we directly used the Dual encoding model in Table 1 without any finetune, so the model was trained in the training set (6,513 video clips)...

On MSVD, we tested on the test set of 670 video clips, and using all the corresponding sentences. It is worth noting that in order to evaluate the generalization ability...

In the previous version of our w2vv [paper](https://arxiv.org/pdf/1604.06838.pdf), in Table 5, we used all the corresponding sentences instead of randomly sampled 5 sentences for each test video ( Results using...

The MSVD data with extracted video features have been updated. https://github.com/danieljf24/dual_encoding#dual-encoding-on-msvd

> I am sorry to bother you, can you share the entails of code in MSVD？ The corresponding code has also been released.

The source code for TPAMI2021 version will be released this week.