Tianshuai Hu

Same issues `answer_nums = np.array([list(map(float, x.split()))[0] for x in answer_nums]).reshape(-1, 2) ValueError: cannot reshape array of size 5 into shape (2) ` will this happen is the test server?

For those who have the same iisues. It seems that ` answer_nums = re.findall(r'\d+\.\d+', answer) ` retrieve data in \d+\.\d format but some of my answers are in \d format.

I meet the same problem and have the same need。 Is there any way to workaround this? @animesh-garg

@pengzhenghao Hi, I'm facing the same problem, I have a gpu version tensorflow but when doing evaluation, I can only see the GPU memory usage but no utility. How do...

@nicomon24 Do we have new LOGGED ORACLE results for the 2025 new metric ( the one shown in the https://arxiv.org/pdf/2305.12032 does not have traffic metrics results）

Hi, @nicomon24 , I have successfully installed the new version of waymo-open-dataset-tf-2-12-0==0.1.6.5. However when using the API to do evaluation, I see some gpu memory usage but no GPU utility....

@rdesc ， I just create a pure conda environment ``` conda create --name waymo25 python=3.10 pip install waymo-open-dataset-tf-2-12-0==1.6.5 ``` my machine information ``` nvcc --version nvcc: NVIDIA (R) Cuda compiler...

@rdesc Alternative way. Install your packages first and then ``` pip install --no-deps waymo-open-dataset-tf-2-12-0==1.6.5 ```

> Thanks [@hutslib](https://github.com/hutslib) ! Got a setup working with conda. Looks like there was something weird going on with the google colab env I was using before. > > One...

@scott-ettinger I compare v1.2.0 tfexample and tfrecord. Basically, I use tfexample.