Lucas Brambrink

Is it possible for you to share `/tmp/call_variants_output.tfrecord.gz` (you can send it to [email protected])? I would be very interested in trying to reproduce the issue myself. I have a theory...

Yes please, that would be great!

@LogCrab Thank you for providing the files. We've tried to reproduce this in various ways without success. Could you provide the specs of the machine you ran this on? Thank...

So we were unable to reproduce this specific error. Regardless, we are overhauling how multiprocessing is used in `postprocess_variants` with our next release, which will very likely avoid this type...

Hi, Thanks for pointing this out! We are going to investigate this issue more closely and provide updates as we make progress. Thank you for providing links to the files,...

Hi, Just to make sure I understand the issue: you are able to successfully install DeepVariant 0.7.0 with Python 2.7, but you are unable to install 1.15 or later; the...

Hi, You can provide a [--intermediate_results_dir](https://github.com/google/deepvariant/blob/r1.6.1/scripts/run_deepvariant.py#L117-L125) flag to `run_deepvariant`, which will save all the intermediate outputs (from make examples and call_variants). If you are interested at looking at the PIL...

Great question, always happy to help! Setting `keep_supplementary_alignments=true` shouldn't dramatically change the accuracy. Is it possible for you to share the BAM and VCF/truth set used? One possibility is that...

Hi! Thanks for the kind words. The issue seems to be that `call_variants` expects a JSON file with the checkpoint. ``` model_example_info_json = f'{checkpoint_path}/example_info.json' model_example_shape = dv_utils.get_shape_and_channels_from_json( model_example_info_json ) ```...

Hello @ayeshbond! Could you please add a log statement with the exact error that you're seeing? Additionally, you can always disable the creation of the `vcf_stats_repot` in case it's blocking...