sockeye
sockeye copied to clipboard
awslabs
Sequence-to-sequence framework with a focus on Neural Machine Translation based on PyTorch
This PR includes #1051. Reviewing/merging it first is recommended. This PR is the first stage of adding [DeepSpeed](https://github.com/microsoft/DeepSpeed) support to Sockeye's `main` branch. It ports several changes from the `deepspeed`...
Hello, How can I diagnose the out-of-memory error below? It's not clear to me why it's happening at that location but I suspect something was trying to perhaps(?) create a...
Hi guys, I would like to report a "bias" in your benchmark from the last paper. I tested those scripts https://github.com/awslabs/sockeye/tree/arxiv_sockeye3/arxiv The thing is that the 3 toolkit save the...
Hi all, I just finished reading Sockey 3 paper. Nicely done, congratulations! Have you considered implementing guided alignments[1] in Sockeye 3? It is handy for formatted document translation, non-translatable entity...
1-How can I plot the transformer attention matrices of the translated text in sockeye? 2-How can I translate a test file containing about 10 lines of a language after training...
Dear all I am thinking about integrating Sockeye on the Huggingface Hub: https://huggingface.co/docs/hub/adding-a-library What are your thoughts about this? Any risks or opportunities you can see? Have a nice day...
I would like to have sockeye for X-to-text translation. (`X ∈ {text, vectors, image, video...}`) **What would the steps to do that be?** ---- I understand `prepare_data` must get the...
Hi, I'm trying out the new Sockeye-3.0.4 and I'm getting the following error ``` Traceback (most recent call last): File "/gpfs/projects/DT/mtp/WMT20/opt/miniconda3/envs/sockeye-3.0.4/bin/sockeye-train", line 8, in sys.exit(main()) File "/gpfs/projects/DT/mtp/WMT20/opt/miniconda3/envs/sockeye-3.0.4/lib/python3.8/site-packages/sockeye/train_pt.py", line 848, in...
Hi, I find the results in your paper that the translation speed with int 8 quantization can be 2 times faster than the fp32 model. However, I did not get...
This is the refactored implementation from the MTMA-2022 [kNN-based Retrieval Module for Sockeye](https://docs.google.com/document/d/1_Lea0E4g-VyqiRTqVbMXfA9iLTcJyQNVQDdkiKosoJ4/edit#heading=h.656mkzwobicc) project. It re-implements the models as in [Khandelwal et al. 2021](https://arxiv.org/abs/2010.00710). #### Pull Request Checklist ## -...
