Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Expose indra virtual tensors in deeplake #2350

Merged
merged 6 commits into from
May 16, 2023
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 8 additions & 5 deletions deeplake/core/dataset/deeplake_query_dataset.py
Original file line number Diff line number Diff line change
Expand Up @@ -167,13 +167,16 @@ def __getitem__(
def __getattr__(self, key):
try:
return self.__getitem__(key)
except TensorDoesNotExistError as ke:
except:
try:
return getattr(self.deeplake_ds, key)
except AttributeError:
raise AttributeError(
f"'{self.__class__}' object has no attribute '{key}'"
) from ke
except:
try:
return getattr(self.indra_ds, key)
except:
raise AttributeError(
f"'{self.__class__}' object has no attribute '{key}'"
)

def __len__(self):
return len(self.indra_ds)
Expand Down
32 changes: 32 additions & 0 deletions deeplake/core/tests/test_deeplake_indra_dataset.py
Original file line number Diff line number Diff line change
Expand Up @@ -276,3 +276,35 @@ def test_random_split(hub_cloud_ds_generator):
l = val.dataloader().pytorch().shuffle()
for b in l:
pass


@requires_libdeeplake
def test_virtual_tensors(hub_cloud_ds_generator):
deeplake_ds = hub_cloud_ds_generator()
with deeplake_ds:
deeplake_ds.create_tensor("label", htype="generic", dtype=np.int32)
deeplake_ds.create_tensor("embeddings", htype="generic", dtype=np.float32)
for i in range(100):
count = i % 5
deeplake_ds.label.append([int(i % 100)] * count)
deeplake_ds.embeddings.append(
[1.0 / float(i + 1), 0.0, -1.0 / float(i + 1)]
)

deeplake_indra_ds = deeplake_ds.query("SELECT shape(label)[0] as num_labels")
assert len(deeplake_indra_ds) == 100
assert deeplake_indra_ds.num_labels[0].numpy() == [0]
assert deeplake_indra_ds.num_labels[1].numpy() == [1]
assert deeplake_indra_ds.num_labels[2].numpy() == [2]
assert deeplake_indra_ds.num_labels[3].numpy() == [3]
assert deeplake_indra_ds.num_labels[4].numpy() == [4]
assert np.sum(deeplake_indra_ds.num_labels.numpy()) == 200

deeplake_indra_ds = deeplake_ds.query(
"SELECT l2_norm(embeddings - ARRAY[0, 0, 0]) as score order by l2_norm(embeddings - ARRAY[0, 0, 0]) asc"
)
assert len(deeplake_indra_ds) == 100
for i in range(100, 1):
assert deeplake_indra_ds.score[100 - i].numpy() == [
np.sqrt(2.0 / (i + 1) / (i + 1))
]
6 changes: 4 additions & 2 deletions deeplake/enterprise/libdeeplake_query.py
Original file line number Diff line number Diff line change
Expand Up @@ -39,10 +39,12 @@ def query(dataset, query_string: str):
else:
ds = dataset_to_libdeeplake(dataset)
dsv = ds.query(query_string)
try:
from deeplake.enterprise.convert_to_libdeeplake import INDRA_API

if not isinstance(dataset, DeepLakeQueryDataset) and INDRA_API.tql.parse(query_string).is_filter: # type: ignore
indexes = dsv.indexes
return dataset[indexes]
except RuntimeError:
else:
view = DeepLakeQueryDataset(deeplake_ds=dataset, indra_ds=dsv)
view._tql_query = query_string
if hasattr(dataset, "is_actually_cloud"):
Expand Down
1 change: 1 addition & 0 deletions deeplake/requirements/tests.txt
Original file line number Diff line number Diff line change
Expand Up @@ -18,3 +18,4 @@ rich
wandb

pandas; python_version >= '3.11' and sys_platform == 'win32'
datasets; python_version >= '3.11' and sys_platform == 'win32'
2 changes: 1 addition & 1 deletion setup.py
Original file line number Diff line number Diff line change
Expand Up @@ -69,7 +69,7 @@ def libdeeplake_availabe():
extras_require["all"] = [req_map[r] for r in all_extras]

if libdeeplake_availabe():
libdeeplake = "libdeeplake==0.0.52"
libdeeplake = "libdeeplake==0.0.53"
extras_require["enterprise"] = [libdeeplake, "pyjwt"]
extras_require["all"].append(libdeeplake)

Expand Down