Chris Seymour

@danwarrior see https://github.com/nanoporetech/bonito/issues/4#issuecomment-562876731

@Howard-Liang no, the available training set does not fully represent that used for the pre-trained models.

Hey @rajeevsrao The problem with the export is in the last layer `GlobalNorm` which is implemented with Cupy [here]( https://github.com/davidcpage/seqdist). ``` $ bonito view bonito/models/dna_r9.4.1/config.toml Model( (encoder): Sequential( (0): Conv1d(1,...

I've identified the problematic test as `tests/operators_test.py::SingleOperatorTest::test_avg_pool`. ``` $ pytest tests/operators_test.py::SingleOperatorTest::test_avg_pool ================================ test session starts ================================ platform darwin -- Python 3.7.3, pytest-5.0.1, py-1.8.0, pluggy-0.12.0 cachedir: .pytest_cache plugins: cov-2.7.1 collected 1...

No, sorry, we don't have support for language models.

I'd start by looking at how https://github.com/parlance/ctcdecode do it.

I'm very happy to upstream improvements.

@edwardpwtsoi you are running out of memory and the OOM killer is kicking in. I can successfully decode your `posterior.json` on a machine with enough system memory. Try increasing `beam_cut_threshold`:...

No a batch dimension is not handled.

It looks like the new [GRU](https://www.tensorflow.org/versions/r2.0/api_docs/python/tf/keras/layers/GRU) layer is CuDNN compatible and supports `reset_after` so I think it should just be a case of swapping `tf.contrib.cudnn_rnn.CudnnCompatibleGRUCell` for `tf.keras.layers.GRU`. I will have...