#758

Committed 04 Sep 2023 08:37PM UTC coverage: 0.0% (-78.3%) from 78.333%

Build # #758

Build Type

push

circle-ci

Committed by

hershd23

Commit Message

Increased underline length in at line 75 in text_summarization.rst
	modified:   docs/source/benchmarks/text_summarization.rst

Run Details

0 of 11303 relevant lines covered (0.0%)

0.0 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

0.0

/evadb/udfs/ndarray/similarity.py

# coding=utf-8
# Copyright 2018-2023 EvaDB
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import pandas as pd

from evadb.udfs.abstract.abstract_udf import AbstractUDF
from evadb.utils.generic_utils import try_to_import_faiss


class Similarity(AbstractUDF):
    def _get_distance(self, numpy_distance):
        return numpy_distance[0][0]

    def setup(self):
        try_to_import_faiss()
        pass

    @property
    def name(self):
        return "Similarity"

    def forward(self, df: pd.DataFrame) -> pd.DataFrame:
        """
        Get similarity score between two feature vectors: 1. feature vector of an opened image;
        and 2. feature vector from base table.
        """

        def _similarity(row: pd.Series) -> float:
            open_feat_np, base_feat_np = (
                row.iloc[0],
                row.iloc[1],
            )

            # TODO: currently system takes care of feature vector shape
            # transformation. Improve this later on.
            # Transform to 2D.
            open_feat_np = open_feat_np.reshape(1, -1)
            base_feat_np = base_feat_np.reshape(1, -1)
            import faiss

            distance_np = faiss.pairwise_distances(open_feat_np, base_feat_np)

            return self._get_distance(distance_np)

        ret = pd.DataFrame()
        ret["distance"] = df.apply(_similarity, axis=1)
        return ret

1	# coding=utf-8
2	# Copyright 2018-2023 EvaDB
3	#
4	# Licensed under the Apache License, Version 2.0 (the "License");
5	# you may not use this file except in compliance with the License.
6	# You may obtain a copy of the License at
7	#
8	# http://www.apache.org/licenses/LICENSE-2.0
9	#
10	# Unless required by applicable law or agreed to in writing, software
11	# distributed under the License is distributed on an "AS IS" BASIS,
12	# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13	# See the License for the specific language governing permissions and
14	# limitations under the License.
15	import pandas as pd	×
16
17	from evadb.udfs.abstract.abstract_udf import AbstractUDF	×
18	from evadb.utils.generic_utils import try_to_import_faiss	×
19
20
21	class Similarity(AbstractUDF):	×
22	def _get_distance(self, numpy_distance):	×
23	return numpy_distance[0][0]	×
24
25	def setup(self):	×
26	try_to_import_faiss()	×
27	pass	×
28
29	@property	×
30	def name(self):	×
31	return "Similarity"	×
32
33	def forward(self, df: pd.DataFrame) -> pd.DataFrame:	×
34	"""
35	Get similarity score between two feature vectors: 1. feature vector of an opened image;
36	and 2. feature vector from base table.
37	"""
38
39	def _similarity(row: pd.Series) -> float:	×
40	open_feat_np, base_feat_np = (	×
41	row.iloc[0],
42	row.iloc[1],
43	)
44
45	# TODO: currently system takes care of feature vector shape
46	# transformation. Improve this later on.
47	# Transform to 2D.
48	open_feat_np = open_feat_np.reshape(1, -1)	×
49	base_feat_np = base_feat_np.reshape(1, -1)	×
50	import faiss	×
51
52	distance_np = faiss.pairwise_distances(open_feat_np, base_feat_np)	×
53
54	return self._get_distance(distance_np)	×
55
56	ret = pd.DataFrame()	×
57	ret["distance"] = df.apply(_similarity, axis=1)	×
58	return ret	×

georgia-tech-db / eva / #758

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous