update tensors from generic types to hub schema representations

[haiyangdeperci]

This PR reflects the changes to the Objectron datasets' schema, i.e. the substitution of generic tensor types with hub schema representations.

[github-actions]

Locust summary

Git references

Initial: 46e6cec
Terminal: 0621b9e

---

hub/api/integrations.py

Changes:

• Name: _to_tensorflow
  Type: function
  Changed lines: 13
  Total lines: 110
  Changes:
  • Name: _to_tensorflow._get_active_item
    Type: function
    Changed lines: 1
    Total lines: 13
  • Name: _to_tensorflow.dict_to_tf
    Type: function
    Changed lines: 2
    Total lines: 5
  • Name: _to_tensorflow.dtype_to_tf
    Type: function
    Changed lines: 2
    Total lines: 11
  • Name: _to_tensorflow.get_output_shapes
    Type: function
    Changed lines: 2
    Total lines: 7
  • Name: _to_tensorflow.output_shapes_from_dict
    Type: function
    Changed lines: 2
    Total lines: 5

hub/api/dataset.py

Changes:

• Name: Dataset
  Type: class
  Changed lines: 26
  Total lines: 1087
  Changes:
  • Name: Dataset._get_dynamic_tensor_dtype
    Type: function
    Changed lines: 7
    Total lines: 7
  • Name: Dataset._get_compressor
    Type: function
    Changed lines: 13
    Total lines: 13
  • Name: Dataset._generate_storage_tensors
    Type: function
    Changed lines: 2
    Total lines: 25
  • Name: Dataset.to_tensorflow
    Type: function
    Changed lines: 2
    Total lines: 14

hub/store/dynamic_tensor.py

Changes:

• Name: DynamicTensor
  Type: class
  Changed lines: 17
  Total lines: 484
  Changes:
  • Name: DynamicTensor.__getitem__
    Type: function
    Changed lines: 14
    Total lines: 24
  • Name: DynamicTensor.get_shape
    Type: function
    Changed lines: 3
    Total lines: 16

hub/api/datasetview.py

Changes:

• Name: DatasetView
  Type: class
  Changed lines: 2
  Total lines: 320
  Changes:
  • Name: DatasetView.to_tensorflow
    Type: function
    Changed lines: 2
    Total lines: 13

hub/api/tensorview.py

Changes:

• Name: TensorView
  Type: class
  Changed lines: 7
  Total lines: 335
  Changes:
  • Name: TensorView.numpy
    Type: function
    Changed lines: 7
    Total lines: 56

hub/compute/transform.py

Changes:

• Name: Transform
  Type: class
  Changed lines: 10
  Total lines: 399
  Changes:
  • Name: Transform.upload
    Type: function
    Changed lines: 10
    Total lines: 56
    Changes:

hub/store/metastore.py

Changes:

• Name: MetaStorage
  Type: class
  Changed lines: 1
  Total lines: 133
  Changes:
  • Name: MetaStorage.__setitem__
    Type: function
    Changed lines: 1
    Total lines: 17

hub/api/tests/test_converters.py

Changes:

• Name: test_to_tensorflow
  Type: function
  Changed lines: 1
  Total lines: 10

hub/api/tests/test_dataset.py

Changes:

• Name: test_tensorview_slicing
  Type: function
  Changed lines: 2
  Total lines: 9
• Name: test_dataset_copy_s3_local
  Type: function
  Changed lines: 2
  Total lines: 14
• Name: test_dataset_copy_gcs_local
  Type: function
  Changed lines: 2
  Total lines: 14
• Name: test_dataset_copy_azure_local
  Type: function
  Changed lines: 5
  Total lines: 21
• Name: test_dataset_copy_hub_local
  Type: function
  Changed lines: 2
  Total lines: 14
• Name: test_dataset_copy_gcs_s3
  Type: function
  Changed lines: 2
  Total lines: 14
• Name: test_dataset_copy_exception
  Type: function
  Changed lines: 2
  Total lines: 12
• Name: test_dataset_filter_3
  Type: function
  Changed lines: 2
  Total lines: 19

hub/cli/auth.py

Changes:

• Name: reporting
  Type: function
  Changed lines: 1
  Total lines: 11

hub/store/tests/test_dynamic_tensor.py

Changes:

• Name: test_dynamic_tensor_2
  Type: function
  Changed lines: 1
  Total lines: 15

hub/utils.py

Changes:

• Name: batchify
  Type: function
  Changed lines: 2
  Total lines: 9

[codecov]

Codecov Report

Merging #705 (aadef32) into master (68d851e) will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master     #705   +/-   ##
=======================================
  Coverage   88.95%   88.95%           
=======================================
  Files          58       58           
  Lines        4266     4266           
=======================================
  Hits         3795     3795           
  Misses        471      471           

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 68d851e...aadef32. Read the comment docs.

[mynameisvinn]

This looks great @haiyangdeperci!

1. For get_bbox, can we pass element as an argument and/or keep it in the local scope?
2. Mind clearing cell outputs when you're done? (You can find it in the cell tab.)

[haiyangdeperci]

@mynameisvinn What do you think of clearing out just the installation output? That way those who browse the notebooks will be able to see what we do here.
I've changed the function to accept element as the argument. Let me know if that's sufficient.

[mynameisvinn]

Yep good idea, clearing pip activity will make things more clear.

[imshashank]

Everything looks good.
Just one recommendation, can we change the element in the example for shoeing bounding boxes. The given bike doesn't have a very good bounding box.

Can we use some good example instead. Anything else will do. Rest all looks good to me.

[haiyangdeperci]

I swapped the example, I think it looks good. Ready to merge after you re-review it.

[imshashank]

LGTM