get_frames_by_pts() RuntimeError: Expected at most 2 arguments but received 3

[mrezabayat]

🐛 Describe the bug

Hi,

I’m encountering a runtime error when running the benchmark script torchcodec/benchmarks/decoders/benchmark_decoders.py. This appears to be related to the number of arguments passed to the get_frames_by_pts function.

Here is how I am running the script:

python benchmarks/decoders/benchmark_decoders.py \
  --video-paths ../sample_videos/Barcelona_vc6.mxf,../sample_videos/Barcelona_h264.mxf \
  --decoders torchcodec_core \
  --plot-path codec_comparison.png \
  --min-run-seconds 2.0 \
  --verify-outputs \
  --verbose

Here is the full call stack:

Traceback (most recent call last):
  File "/home/mohammad.bayat/dev/torchcodec/benchmarks/decoders/benchmark_decoders.py", line 167, in <module>
    main()
  File "/home/mohammad.bayat/dev/torchcodec/benchmarks/decoders/benchmark_decoders.py", line 139, in main
    verify_outputs(decoders_to_run, video_paths, num_uniform_samples)
  File "/home/mohammad.bayat/dev/torchcodec/benchmarks/decoders/benchmark_decoders_library.py", line 1069, in verify_outputs
    frames = decode_and_adjust_frames(
  File "/home/mohammad.bayat/dev/torchcodec/benchmarks/decoders/benchmark_decoders_library.py", line 1085, in decode_and_adjust_frames
    non_seq_frames = decoder.decode_frames(video_file_path, pts_list)
  File "/home/mohammad.bayat/dev/torchcodec/benchmarks/decoders/benchmark_decoders_library.py", line 243, in decode_frames
    frames, *_ = get_frames_by_pts(
  File "/home/mohammad.bayat/dev/torchcodec/.venv/lib/python3.10/site-packages/torch/_ops.py", line 756, in __call__
    return self._op(*args, **kwargs)
RuntimeError: torchcodec_ns::get_frames_by_pts() expected at most 2 argument(s) but received 3 argument(s). Declaration: torchcodec_ns::get_frames_by_pts(Tensor(a!) decoder, *, float[] timestamps) -> (Tensor, Tensor, Tensor)

Versions

Package                  Version
------------------------ -----------
certifi                  2025.7.9
charset-normalizer       3.4.2
contourpy                1.3.2
cycler                   0.12.1
filelock                 3.18.0
fonttools                4.58.5
fsspec                   2025.5.1
idna                     3.10
Jinja2                   3.1.6
kiwisolver               1.4.8
MarkupSafe               3.0.2
matplotlib               3.10.3
mpmath                   1.3.0
networkx                 3.4.2
numpy                    2.2.6
nvidia-cublas-cu12       12.6.4.1
nvidia-cuda-cupti-cu12   12.6.80
nvidia-cuda-nvrtc-cu12   12.6.77
nvidia-cuda-runtime-cu12 12.6.77
nvidia-cudnn-cu12        9.5.1.17
nvidia-cufft-cu12        11.3.0.4
nvidia-cufile-cu12       1.11.1.6
nvidia-curand-cu12       10.3.7.77
nvidia-cusolver-cu12     11.7.1.2
nvidia-cusparse-cu12     12.5.4.2
nvidia-cusparselt-cu12   0.6.3
nvidia-nccl-cu12         2.26.2
nvidia-nvjitlink-cu12    12.6.85
nvidia-nvtx-cu12         12.6.77
packaging                25.0
pandas                   2.3.1
pillow                   11.3.0
pip                      22.0.2
pyparsing                3.2.3
python-dateutil          2.9.0.post0
pytz                     2025.2
requests                 2.32.4
setuptools               59.6.0
six                      1.17.0
sympy                    1.14.0
torch                    2.7.1
torchaudio               2.7.1
torchcodec               0.4.0
torchvision              0.22.1
triton                   3.3.1
typing_extensions        4.14.1
tzdata                   2025.2
urllib3                  2.5.0

OS: Ubuntu 22.04

Python: 3.10.12

[NicolasHug]

Thanks for the report @mrezabayat , I think this is because we are still passing the stream_index parameter to those decoding methods, but we have removed it a while ago. Would you like to submit a PR to fix this? Basically we just need to remove stream_index from the calls to get_frames_by_pts and get_frames_at_indices in https://github.com/pytorch/torchcodec/blob/main/benchmarks/decoders/benchmark_decoders_library.py

[mrezabayat]

Thanks for your swift response @NicolasHug.
Yes, I’m happy to raise a PR and will do so soon.