Open esraaelmligy opened 5 days ago
@esraaelmligy ,
The RNAseq model is trained and released for DeepVariant 1.4.0, please see this documentation on how to run it. Please change deepvariant:latest
to deepvariant:1.4.0
.
Thank you so much, I changed the version and the run finished well. I just have to ask whether using a non-updated version will dramatically affect the results, and if you are planning to do any updates on the RNAseq model soon so it could be suitable for later versions?
Describe the issue: Hello, I am trying to run the deepvariant RNA model on HG005 data locally (not restricted to chr20 like the Github example) but the model keeps generating this error while running the call_variants.py step
WARNING:tensorflow:From /usr/local/lib/python3.8/dist-packages/tensorflow/python/checkpoint/checkpoint.py:1473: NameBasedSaverStatus.init (from tensorflow.python.checkpoint.checkpoint) is deprecated and will be removed in a future version. Instructions for updating: Restoring a name-based tf.train.Saver checkpoint using the object-based restore API. This mode uses global names to match variables, and so is somewhat fragile. It also adds new restore ops to the graph each time it is called when graph building. Prefer re-encoding training checkpoints in the object-based format: run save() on the object-based saver (the same one this message is coming from) and use that checkpoint in the future. W0626 13:39:06.145823 140632388314944 deprecation.py:350] From /usr/local/lib/python3.8/dist-packages/tensorflow/python/checkpoint/checkpoint.py:1473: NameBasedSaverStatus.init (from tensorflow.python.checkpoint.checkpoint) is deprecated and will be removed in a future version. Instructions for updating: Restoring a name-based tf.train.Saver checkpoint using the object-based restore API. This mode uses global names to match variables, and so is somewhat fragile. It also adds new restore ops to the graph each time it is called when graph building. Prefer re-encoding training checkpoints in the object-based format: run save() on the object-based saver (the same one this message is coming from) and use that checkpoint in the future. Traceback (most recent call last): File "/tmp/Bazel.runfileszgkztyv/runfiles/com_google_deepvariant/deepvariant/call_variants.py", line 633, in
app.run(main)
File "/tmp/Bazel.runfiles__zgkztyv/runfiles/absl_py/absl/app.py", line 312, in run
_run_main(main, args)
File "/tmp/Bazel.runfiles zgkztyv/runfiles/absl_py/absl/app.py", line 258, in _run_main
sys.exit(main(argv))
File "/tmp/Bazel.runfileszgkztyv/runfiles/com_google_deepvariant/deepvariant/call_variants.py", line 618, in main
call_variants(
File "/tmp/Bazel.runfileszgkztyv/runfiles/com_google_deepvariant/deepvariant/call_variants.py", line 558, in call_variants
model.load_weights(checkpoint_path).expect_partial()
File "/usr/local/lib/python3.8/dist-packages/keras/utils/traceback_utils.py", line 70, in error_handler
raise e.with_traceback(filtered_tb) from None
File "/usr/local/lib/python3.8/dist-packages/tensorflow/python/checkpoint/checkpoint.py", line 1047, in assert_consumed
raise AssertionError(
AssertionError: Some objects had attributes which were not restored:
<tf.Variable 'conv2d/kernel:0' shape=(3, 3, 7, 32) dtype=float32, numpy=
My knowledge in deep learning models is not the best, so if you could please tell me how to overcome this error, as the RNA model seems to have very promising results for RNA variant calling and i want to use it.
Setup
Steps to reproduce: