def eval_request():
num_docs = 10
req = jina_pb2.RequestProto()
for idx in range(num_docs):
doc = Document(req.index.docs.add())
gt = Document(req.index.groundtruths.add())
doc.update_id()
gt.update_id()
chunk_doc = doc.add_chunk()
chunk_gt = gt.add_chunk()
chunk_doc.embedding = np.array([1, 1])
chunk_gt.embedding = np.array([2, 2])
return Request(req)
def doc_with_multimodal_chunks(embeddings):
doc = Document()
chunk1 = Document()
chunk2 = Document()
chunk3 = Document()
chunk1.modality = 'visual1'
chunk2.modality = 'visual2'
chunk3.modality = 'textual'
chunk1.embedding = embeddings[0]
chunk2.embedding = embeddings[1]
chunk3.embedding = embeddings[2]
chunk1.update_id()
chunk2.update_id()
chunk3.update_id()
doc.update_id()
doc.add_chunk(chunk1)
doc.add_chunk(chunk2)
doc.add_chunk(chunk3)
return doc
def create_document_to_search():
# 1-D embedding
# doc: 0
# - chunk: 1
# - chunk: 2
# - chunk: 3
# - chunk: 4
# - chunk: 5 - will be missing from KV indexer
doc = Document()
doc.id = '0' * 16
for c in range(5):
chunk = doc.add_chunk()
chunk.id = str(c + 1) * 16
return doc
def create_document_to_search_with_matches_on_chunks():
# 1-D embedding
# doc: 0
# - chunk: 1
# - match: 2
# - match: 3
# - match: 4
# - match: 5 - will be missing from KV indexer
# - match: 6 - will be missing from KV indexer
doc = Document()
doc.id = '0' * 16
chunk = doc.add_chunk()
chunk.id = '1' * 16
for m in range(5):
match = chunk.add_match(doc_id=str(m + 2) * 16, score_value=1.)
return doc
