Movatterモバイル変換

[0]ホーム
URL:

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

 Sign up
Appearance settings

Separate embedding kwargs into init kwargs and encode kwargs#1555

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 ourterms of service andprivacy statement. We’ll occasionally send you account related emails.

Already on GitHub?Sign in to your account

Merged
Show file tree
Hide file tree
Changes fromall 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
7 changes: 7 additions & 0 deletionspgml-extension/examples/embedding.sql
View file
Open in desktop
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,7 @@
\timing on

SELECT pgml.embed('Alibaba-NLP/gte-base-en-v1.5', 'hi mom', '{"trust_remote_code": true}');
SELECT pgml.embed('Alibaba-NLP/gte-base-en-v1.5', 'hi mom', '{"device": "cuda", "trust_remote_code": true}');
SELECT pgml.embed('Alibaba-NLP/gte-base-en-v1.5', 'hi mom', '{"device": "cpu", "trust_remote_code": true}');
SELECT pgml.embed('hkunlp/instructor-xl', 'hi mom', '{"instruction": "Encode it with love"}');
SELECT pgml.embed('mixedbread-ai/mxbai-embed-large-v1', 'test', '{"prompt": "test prompt: "}');
2 changes: 1 addition & 1 deletionpgml-extension/examples/image_classification.sql
View file
Open in desktop
Original file line numberDiff line numberDiff line change
Expand Up@@ -66,7 +66,7 @@ SELECT * FROM pgml.train('Handwritten Digits', algorithm => 'xgboost', hyperpara

-- runtimes
SELECT * FROM pgml.train('Handwritten Digits', algorithm => 'linear', runtime => 'python');
SELECT * FROM pgml.train('Handwritten Digits', algorithm => 'linear', runtime => 'rust');
--SELECT * FROM pgml.train('Handwritten Digits', algorithm => 'linear', runtime => 'rust');

--SELECT * FROM pgml.train('Handwritten Digits', algorithm => 'xgboost', runtime => 'python', hyperparams => '{"n_estimators": 10}'); -- too slow
SELECT * FROM pgml.train('Handwritten Digits', algorithm => 'xgboost', runtime => 'rust', hyperparams => '{"n_estimators": 10}');
Expand Down
26 changes: 21 additions & 5 deletionspgml-extension/src/bindings/transformers/transformers.py
View file
Open in desktop
Original file line numberDiff line numberDiff line change
Expand Up@@ -527,8 +527,8 @@ def rank(transformer, query, documents, kwargs):
return rank_using(model, query, documents, kwargs)


def create_embedding(transformer):
return SentenceTransformer(transformer)
def create_embedding(transformer, kwargs):
return SentenceTransformer(transformer, **kwargs)


def embed_using(model, transformer, inputs, kwargs):
Expand All@@ -545,16 +545,32 @@ def embed_using(model, transformer, inputs, kwargs):

def embed(transformer, inputs, kwargs):
kwargs = orjson.loads(kwargs)

ensure_device(kwargs)

init_kwarg_keys = [
"device",
"trust_remote_code",
"revision",
"model_kwargs",
"tokenizer_kwargs",
"config_kwargs",
"truncate_dim",
"token",
]
init_kwargs = {
key: value for key, value in kwargs.items() if key in init_kwarg_keys
}
encode_kwargs = {
key: value for key, value in kwargs.items() if key not in init_kwarg_keys
}

if transformer not in __cache_sentence_transformer_by_name:
__cache_sentence_transformer_by_name[transformer] = create_embedding(
transformer
transformer, init_kwargs
)
model = __cache_sentence_transformer_by_name[transformer]

return embed_using(model, transformer, inputs,kwargs)
return embed_using(model, transformer, inputs,encode_kwargs)


def clear_gpu_cache(memory_usage: None):
Expand Down
1 change: 1 addition & 0 deletionspgml-extension/tests/test.sql
View file
Open in desktop
Original file line numberDiff line numberDiff line change
Expand Up@@ -31,5 +31,6 @@ SELECT pgml.load_dataset('wine');
\i examples/vectors.sql
\i examples/chunking.sql
\i examples/preprocessing.sql
\i examples/embedding.sql
-- transformers are generally too slow to run in the test suite
--\i examples/transformers.sql
[8]ページ先頭
©2009-2025 Movatter.jp