select
*
from mldb.ONNXEmbeddings(
on (SELECT CAST(rev_id AS VARCHAR(21)) AS rev_id, rev_text AS txt FROM amazon_reviews_25)
on (select model_id, model from onnx_models where model_id = 'bge-small-en-v1.5') DIMENSION
on (select tokenizer from embeddings_tokenizers where model_id = 'bge-small-en-v1.5') DIMENSION
USING
Accumulate('rev_id')
ModelOutputTensor('sentence_embedding')
OutputFormat('VARCHAR(30000)')
) a
;
*** Query completed. 25 rows found. 2 columns returned.
*** Total elapsed time was 1 second.
The following example is a snippet of the total output:
rev_id sentence_embedding ----------------------------------------------------------------------------------------------------------------------------- A3V1MKC2BVWY48 0.010341135 0.039603148 0.02032054 0.017130554 7.526378E-4 -0.0029828206 0.08408154 -0.06986211 0.0011597961 -0.067057185 -0.042081792 0.0044557634 0.051 A26GKZPS079GFF -0.033367846 0.013266026 0.039863694 0.03130837 -0.0101601565 0.0021604085 0.035781875 -0.0029493966 0.038836565 -0.02592681 0.020648086 -0.010395628 0.0 AUTNO7VDY4H4A -0.011844048 0.022235172 -0.005435251 0.023775024 -0.011273234 -0.0017255824 0.017376512 0.020886978 -0.011113251 -0.07543064 0.010284837 0.052515816 -0. A2S166WSCFIFP5 0.002855719 -0.014396132 -0.016711127 -0.008662777 -0.010505226 0.05228832 0.036535893 -0.0014042043 0.048945963 -0.037907925 -0.007915686 -0.014041564 0 A5E9TSD20U9PR -0.025540393 0.042433415 0.0038648746 0.024865156 0.008279116 0.010526418 0.056619547 0.021985738 -0.025672358 -0.062182724 -0.05664199 0.03468444 0.0134 A27ZH1AQORJ1L -0.0040720943 0.031230347 0.054846305 0.02475443 -0.025144236 -0.024486065 0.042867593 -0.014755931 -0.006141968 -0.022795495 -0.005213292 3.0384486E-4 0 ....