Skip to content

update ingestion to include image embeddings #248

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.

Sign up for GitHub

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

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

update ingestion to include image embeddings #248

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
33 changes: 29 additions & 4 deletions apps/dataset-ingestion-movies/app/movie_record.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,6 +28,7 @@
HAS_SPOKEN_LANGUAGE_CONNECTION_LABEL = "HasSpokenLanguage"
HAS_IMAGE_CONNECTION_LABEL = "HasImage"
HAS_TAGLINE_EMBEDDING_CONNECTION_LABEL = "HasTaglineEmbedding"
HAS_IMAGE_EMBEDDING_CONNECTION_LABEL = "HasImageEmbedding"

def make_movie_with_all_connections(j: dict, embedder: Embedder, ingest_posters: bool = False, embed_tagline: bool = False) -> List[dict]:
"""
Expand Down Expand Up @@ -218,17 +219,41 @@ def make_movie_with_all_connections(j: dict, embedder: Embedder, ingest_posters:
blobs.append(image_data)
index += 1

image_descriptor = embedder.embed_image(image_data)
image_descriptor_blob = image_descriptor.tobytes()
image_descriptor_id = f"image_embedding_{movie_parameters['properties']['id']}"
image_descriptor_command = QueryBuilder.add_command(DESCRIPTOR_LABEL, dict(
_ref=index,
set=embedder.descriptor_set,
properties=dict(
source="image",
dataset_name=DATASET_NAME,
id=image_descriptor_id,
), if_not_found=dict(id=["==", image_descriptor_id])
))
transaction.append(image_descriptor_command)
blobs.append(image_descriptor_blob)
connection_parameters = dict(src=index - 1, dst=index, properties=dict(
name=HAS_IMAGE_EMBEDDING_CONNECTION_LABEL,
uniqueid=HAS_IMAGE_EMBEDDING_CONNECTION_LABEL
))
connection_parameters["class"] = HAS_IMAGE_EMBEDDING_CONNECTION_LABEL
connection = QueryBuilder.add_command(CONNECTION_LABEL, connection_parameters)
transaction.append(connection)
index += 1

if embed_tagline:
tagline_embedding = embedder.embed_text(movie_parameters['properties']['tagline'])
tagline_blob = tagline_embedding.tobytes()
tagline_descriptor_id = f"tagline_embedding_{movie_parameters['properties']['id']}"
tagline_command = QueryBuilder.add_command(DESCRIPTOR_LABEL, dict(
_ref=index,
set=embedder.descriptor_set,
properties=dict(
source_property="tagline",
id=movie_parameters['properties']['id'],
source="tagline",
id=tagline_descriptor_id,
dataset_name=DATASET_NAME
)
), if_not_found=dict(id=["==", tagline_descriptor_id])
))
transaction.append(tagline_command)
blobs.append(tagline_blob)
Expand Down Expand Up @@ -262,4 +287,4 @@ def create_indexes(utils: Utils):
# Demo has embeddings
utils.create_entity_index(DESCRIPTOR_LABEL, "_create_txn")
utils.create_entity_index(DESCRIPTOR_LABEL, "id")
utils.create_entity_index(DESCRIPTOR_SET_LABEL, "_name")
utils.create_entity_index(DESCRIPTOR_SET_LABEL, "_name")
Loading