Add tfio.IOTensor.from_hdf5 support #441
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
yongtang
force-pushed
the
hdf5_io_tensor
branch
from
49d7203 to
9015ba6
Compare
HDF5 file is a widely used format. It normally stores data into each named `dataset` which is a block of array with shape. It is not exactly columnar as different `dataset` in HDF5 could have different shapes unrelated to each other. From that standpoint it is more like a storage for collections of tensors (where each `dataset` represent one `tensor`). HDF5 does allow slicing and indexing. In fact, the slicing and indexing in HDF5 are much more powerful than many other formats. This PR adds tfio.IOTensor.from_hdf5. It treats HDF5 as a collection of BaseIOTensor which could be further used for slicing and indexing. Note the `collection` here essentially is just a dictionary of key with BaseIOTensor as the value. It is different from Columnar IOTensor's case like Parquet or Avro. Signed-off-by: Yong Tang <[email protected]>
yongtang
force-pushed
the
hdf5_io_tensor
branch
from
9015ba6 to
ed7e181
Compare
|
Plan to merge this PR soon, as all tests passed now, and this PR is less impactful to other column dataset like parquet/avro. |
i-ony
pushed a commit
to i-ony/io
that referenced
this pull request
Feb 8, 2021
HDF5 file is a widely used format. It normally stores data into each named `dataset` which is a block of array with shape. It is not exactly columnar as different `dataset` in HDF5 could have different shapes unrelated to each other. From that standpoint it is more like a storage for collections of tensors (where each `dataset` represent one `tensor`). HDF5 does allow slicing and indexing. In fact, the slicing and indexing in HDF5 are much more powerful than many other formats. This PR adds tfio.IOTensor.from_hdf5. It treats HDF5 as a collection of BaseIOTensor which could be further used for slicing and indexing. Note the `collection` here essentially is just a dictionary of key with BaseIOTensor as the value. It is different from Columnar IOTensor's case like Parquet or Avro. Signed-off-by: Yong Tang <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
HDF5 file is a widely used format. It normally stores data into each named
datasetwhich is a block of array with shape. It is not exactly columnar as differentdatasetin HDF5 could have different shapes unrelated to each other. From that standpoint it is more like a storage for collections of tensors (where eachdatasetrepresent onetensor).HDF5 does allow slicing and indexing. In fact, the slicing and indexing in HDF5 are much more powerful than many other formats.
This PR adds tfio.IOTensor.from_hdf5. It treats HDF5 as a collection of BaseIOTensor which could be further used for slicing and indexing.
Note the
collectionhere essentially is just a dictionary of key with BaseIOTensor as the value. It is different from Columnar IOTensor's case like Parquet or Avro.Signed-off-by: Yong Tang [email protected]