Sentence Similarity
sentence-transformers
Safetensors
xlm-roberta
feature-extraction
dense
Generated from Trainer
dataset_size:284573
loss:MultipleNegativesRankingLoss
text-embeddings-inference
Instructions to use EmiliaLee/k_model with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use EmiliaLee/k_model with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("EmiliaLee/k_model") sentences = [ "durable plastic shed for garden supplies", "stoney resin outdoor 4 ft. 4 in . w x 2 ft. 8 in . d plastic horizontal storage shed Sheds this handy horizontal storage shed can handle any and all of your outdoor organizational needs . ideal for storing garbage cans , lawn and garden products , snow blowers , bicycles , and more , this stylish storage solution is essential in every season . additional features include durable , multi-w doorstyle : double door|primarymaterial : plastic|countryoforigin : united states|shedtype : storage shed|roofstyle : pent|estimatedtimetosetup:30|countryoforigin-additionaldetails : made in usa|overallproductweight:76|interiorheight-toptobottom:43.5|color : gray|doorwidth-sidetoside:46.25|overallwi", "outdoor 4 ft. w x 2 ft. d plastic vertical tool shed Deck Boxes|Sheds this large vertical shed is designed to supply generous storage space for outdoor gear , tools , equipment , firewood , sporting goods , and more . it 's fashioned with multi-wall resin and is outfitted with adjoining door handles to attach a padlock ( sold separately ) . to augment this unit with s coveragearea:11.23|estimatedtimetosetup:180|approximatedepth-fronttoback:3|windrating:65|overallwidth-sidetoside:53|interiorheight-toptobottom:69.5|overallheight-toptobottom:71.5|primarymaterial : plastic|shedtype : tool shed|overalldepth-fronttoback:32.5|color : mustang brown|countryoforigin-additi", "suncast golf organizer in black Bike And Sport Racks when it comes to the great outdoors and engaging in such sporting activities as golf , an organizer is just right to keep those accessories in place . it is sure to become your best friend when out playing on the links . producttype : sports rack|finish : black|compatiblebike : |overallheight-toptobottom:37|overallwidth-sidetoside:32|overallproductweight:38.15|mounttype : wall mounted|overalldepth-fronttoback:16|holdingcapacity:2|hooksincluded : no|multi-use : no|installationrequired : yes|productwarranty : yes|adju" ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [4, 4] - Notebooks
- Google Colab
- Kaggle
Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- 1_Pooling/config.json +5 -0
- README.md +541 -0
- config.json +30 -0
- config_sentence_transformers.json +14 -0
- model.safetensors +3 -0
- modules.json +20 -0
- sentence_bert_config.json +10 -0
- tokenizer.json +3 -0
- tokenizer_config.json +17 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
1_Pooling/config.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"embedding_dimension": 1024,
|
| 3 |
+
"pooling_mode": "cls",
|
| 4 |
+
"include_prompt": true
|
| 5 |
+
}
|
README.md
ADDED
|
@@ -0,0 +1,541 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- sentence-transformers
|
| 4 |
+
- sentence-similarity
|
| 5 |
+
- feature-extraction
|
| 6 |
+
- dense
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
- dataset_size:284573
|
| 9 |
+
- loss:MultipleNegativesRankingLoss
|
| 10 |
+
base_model: BAAI/bge-m3
|
| 11 |
+
widget:
|
| 12 |
+
- source_sentence: durable plastic shed for garden supplies
|
| 13 |
+
sentences:
|
| 14 |
+
- 'stoney resin outdoor 4 ft. 4 in . w x 2 ft. 8 in . d plastic horizontal storage
|
| 15 |
+
shed Sheds this handy horizontal storage shed can handle any and all of your outdoor
|
| 16 |
+
organizational needs . ideal for storing garbage cans , lawn and garden products
|
| 17 |
+
, snow blowers , bicycles , and more , this stylish storage solution is essential
|
| 18 |
+
in every season . additional features include durable , multi-w doorstyle : double
|
| 19 |
+
door|primarymaterial : plastic|countryoforigin : united states|shedtype : storage
|
| 20 |
+
shed|roofstyle : pent|estimatedtimetosetup:30|countryoforigin-additionaldetails
|
| 21 |
+
: made in usa|overallproductweight:76|interiorheight-toptobottom:43.5|color :
|
| 22 |
+
gray|doorwidth-sidetoside:46.25|overallwi'
|
| 23 |
+
- 'outdoor 4 ft. w x 2 ft. d plastic vertical tool shed Deck Boxes|Sheds this large
|
| 24 |
+
vertical shed is designed to supply generous storage space for outdoor gear ,
|
| 25 |
+
tools , equipment , firewood , sporting goods , and more . it ''s fashioned with
|
| 26 |
+
multi-wall resin and is outfitted with adjoining door handles to attach a padlock
|
| 27 |
+
( sold separately ) . to augment this unit with s coveragearea:11.23|estimatedtimetosetup:180|approximatedepth-fronttoback:3|windrating:65|overallwidth-sidetoside:53|interiorheight-toptobottom:69.5|overallheight-toptobottom:71.5|primarymaterial
|
| 28 |
+
: plastic|shedtype : tool shed|overalldepth-fronttoback:32.5|color : mustang brown|countryoforigin-additi'
|
| 29 |
+
- 'suncast golf organizer in black Bike And Sport Racks when it comes to the great
|
| 30 |
+
outdoors and engaging in such sporting activities as golf , an organizer is just
|
| 31 |
+
right to keep those accessories in place . it is sure to become your best friend
|
| 32 |
+
when out playing on the links . producttype : sports rack|finish : black|compatiblebike
|
| 33 |
+
: |overallheight-toptobottom:37|overallwidth-sidetoside:32|overallproductweight:38.15|mounttype
|
| 34 |
+
: wall mounted|overalldepth-fronttoback:16|holdingcapacity:2|hooksincluded : no|multi-use
|
| 35 |
+
: no|installationrequired : yes|productwarranty : yes|adju'
|
| 36 |
+
- source_sentence: large standing mirror for living room
|
| 37 |
+
sentences:
|
| 38 |
+
- 'bar serving cart Patio Bar Sets time to grill but do n''t have enough space to
|
| 39 |
+
set everything nearby ? forget running back and forth into the house and simply
|
| 40 |
+
open up the portable outdoor patio serving station table and cabinet . great for
|
| 41 |
+
entertaining and dining al fresco . this suncast serving station patio cabinet
|
| 42 |
+
has a lot of s overallproductweight:20|interiorcabinetdepth-fronttoback:19|outerframematerial
|
| 43 |
+
: plastic|producttype : bar cart|weightcapacity:30|cabinetinteriorheight-toptobottom:11|countertopmaterialdetails
|
| 44 |
+
: polypropelene|countryoforigin : united states|overallheight-toptobottom:34|overalldepth-fronttoback:22|du'
|
| 45 |
+
- 'kori industrial metal wall mirror Wall & Accent Mirrors mimicking the look of
|
| 46 |
+
exposed pipes , this mirror brings industrial style to your living room , bedroom
|
| 47 |
+
, or guest room . it ''s crafted from iron and awash in a slate gray finish that
|
| 48 |
+
complements any color palette . this rugged piece showcases slight distressed
|
| 49 |
+
detailing for a weathered look , and a countryoforigin-additionaldetails : made
|
| 50 |
+
in china|mirrorheight-toptobottom:29|shape : rectangle|additionaltoolsrequirednotincluded
|
| 51 |
+
: hammer|installationtypedetails : keyhole|mirrorfinishes : beveled|surfaceshape
|
| 52 |
+
: flat|orientation : horizontal and vertical|framefinish : distressed slate gray|lsbinve'
|
| 53 |
+
- 'wyrick on the door full length mirror Wall & Accent Mirrors this high-quality
|
| 54 |
+
mirror complements your decor without the need for unsightly `` over the door
|
| 55 |
+
'''' brackets . viewing area , it gives you a full image either on the back of
|
| 56 |
+
a door or on the wall as a standard full-length mirror . glassthickness:0.141732|installationtypedetails
|
| 57 |
+
: d rings|silvercontent : silver|framematerialdetails : polystyrene|dampdryorwetlocationlisted
|
| 58 |
+
: dry|countryoforigin : united states|mirrorwidth-sidetoside:14|frameconstruction
|
| 59 |
+
: closed corner frame|mirrormaterial : glass|holidayoccasion : no holiday|o'
|
| 60 |
+
- source_sentence: sleek contemporary coffee table for tidy living space
|
| 61 |
+
sentences:
|
| 62 |
+
- 'kalmus coffee table marrying modern and glamorous elements , this coffee table
|
| 63 |
+
sets a sleek and chic foundation in your living room layout . crafted with a metal
|
| 64 |
+
frame with a metallic and glossy finish , this piece features a mirrored glass
|
| 65 |
+
top that provides a place to set down a spread of snacks , a stack of magazines
|
| 66 |
+
basematerial : nickel|basetype:4 legs|dsmetallic : nickel|topcolor : black|producttype
|
| 67 |
+
: coffee table|topshape : round|weightcapacity:30|dssecondaryproductstyle : contemporary
|
| 68 |
+
industrial|overallheight-toptobottom:17|topmaterial : glass|basecolor : black|overallproductweight:22|topdesign
|
| 69 |
+
: glass|over'
|
| 70 |
+
- 'pangburn haven all-weather outdoor recliner patio chair with cushions and ottoman
|
| 71 |
+
Patio Lounge Chairs this gorgeous set of two outdoor recliners with matching ottomans
|
| 72 |
+
mean you and your pal can sit back and relax in style on your porch or patio .
|
| 73 |
+
all-weather wicker and weather-resistant cushions will provide comfort and a fresh
|
| 74 |
+
update to your outdoor living space . simply pull on the lever and the s cushionfillmaterialdetails
|
| 75 |
+
: polyurethane foam and polyester fiber batting|overallheight-toptobottom:43.75|design
|
| 76 |
+
: armchair|cushioncolor : dark gray|overalldepth-fronttoback:26.75|cushioncovermaterial
|
| 77 |
+
: polyester/polyester blend|levelofassembly : full assembly needed|ottomanheight-toptobottom:17|fr'
|
| 78 |
+
- 'plascencia coffee table with storage Coffee & Cocktail Tables bring mid-century
|
| 79 |
+
modern design to your home with the storage coffee table . this rectangle coffee
|
| 80 |
+
table is the answer to keeping communal space tidy , without sacrificing style
|
| 81 |
+
. the hidden storage space is perfect for stowing away books , or mail during
|
| 82 |
+
your next get together . the wood coffee tab dssecondaryproductstyle : mid-century
|
| 83 |
+
modern|warrantylength:1 year|basemetalfinishapplication : powder coated|overallproductweight:40.35|basecolor
|
| 84 |
+
: walnut|overalllength-endtoend:39.76|dsprimaryproductstyle : modern|producttype
|
| 85 |
+
: coffee table|weightcapacity:100|basematerialdetails : mdf and melamine'
|
| 86 |
+
- source_sentence: durable plastic storage bench for garden
|
| 87 |
+
sentences:
|
| 88 |
+
- 'wicker 22 gallon resin/plastic storage bench Deck Boxes this wicker resin storage
|
| 89 |
+
container from suncast is an attractive and highly versatile solution for use
|
| 90 |
+
throughout your home and yard . the chest design and hinged lid allow for easy
|
| 91 |
+
access to your tools and accessories stored inside . with a holding capacity of
|
| 92 |
+
22 gallons ( 17 '''' d x 22 '''' w 20.5 '' storagecapacity:22|estimatedtimetosetup:20|additionaltoolsrequirednotincluded
|
| 93 |
+
: screw driver|topstyle : flat| : waterproof|topweightcapacity:150|levelofassembly
|
| 94 |
+
: full assembly needed|overallproductweight:9.8|countryoforigin : united states|color
|
| 95 |
+
: java|primarymaterial : resin|countryoforigin-additi'
|
| 96 |
+
- 'galaz console table - rustic black a demi lune shape and scalloped edge make
|
| 97 |
+
this console table a stunning addition to a foyer or entryway . it comes with
|
| 98 |
+
tapered legs and a bold rustic black finish that goes with anything . made of
|
| 99 |
+
iron , it will withstand the test of time , and its low profile aesthetic will
|
| 100 |
+
ensure it fits in with overallwidth-sidetoside:35.4|overallproductweight:26.8|basecolor
|
| 101 |
+
: rustic black|basematerialdetails:100 % iron|topshape : rectangle|overallheight-toptobottom:29.5|levelofassembly
|
| 102 |
+
: none|topcolor : rustic black|overalldepth-fronttoback:15.7|basetype : pedestal|glossfinish
|
| 103 |
+
: no|cablemanagement : no|ca'
|
| 104 |
+
- 'plastic deck storage box Deck Boxes elegant and functional , the lifetime 116
|
| 105 |
+
gallon plastic deck storage box is an excellent addition to your patio or deck
|
| 106 |
+
. made from high-quality plastic , this storage box is strong and durable . it
|
| 107 |
+
is enameled with a brown finish . it has a storage capacity of 116 gallons and
|
| 108 |
+
gives enough space fo levelofassembly : partial assembly|warrantylength:10 years|topweightcapacity:600|dswoodtone
|
| 109 |
+
: gray wood|additionaltoolsrequirednotincluded : ratchet|interiorwidth-sidetoside:47|primarymaterial
|
| 110 |
+
: plastic|overallheight-toptobottom:26|overallwidth-sidetoside:50.3|storagecapacity:116|interiorheight-topt'
|
| 111 |
+
- source_sentence: compact garage storage shed under 3 feet wide
|
| 112 |
+
sentences:
|
| 113 |
+
- 'garden plastic/metal storage shed Sheds easily add storage space and value to
|
| 114 |
+
your property with an outdoor garden shed . the metal shed has distinct advantages
|
| 115 |
+
over traditional wooden models that make it a much more attractive choice for
|
| 116 |
+
purchase . a metal shed is more affordable and being pre-fabricated significantly
|
| 117 |
+
cuts down on expens dsmetallic : steel|shedtype : storage shed|color : grey|construction
|
| 118 |
+
: traditional|overallheight-toptobottom:75.5|countryoforigin-additionaldetails
|
| 119 |
+
: made in china|warrantylength:1 year|levelofassembly : full assembly needed|overallwidth-sidetoside:126.75|overalldepth-fronttoback:109|overallproductw'
|
| 120 |
+
- 'outdoor 2 ft. 9 in . w x 2 ft. d plastic vertical tool shed Sheds with its tall
|
| 121 |
+
profile , this 22 cubic foot vertical shed is great for storing ladders , long-handled
|
| 122 |
+
tools , garden accessories , and more when space is at a premium . the durable
|
| 123 |
+
, all-weather resin construction provides water resistance and uv protection ,
|
| 124 |
+
while the multi-wall panels are engineere interiorwidth-sidetoside:27|interiorheight-toptobottom:69|compatibleshelfpartnumber
|
| 125 |
+
: xa1182|windrating:65|overallwidth-sidetoside:33|countryoforigin : united states|interiordepth-fronttoback:20.25|levelofassembly
|
| 126 |
+
: full assembly needed|countryoforigin-additionaldetails : made in usa|doorwidth-sidet'
|
| 127 |
+
- 'foulds 11.5 '''' wall clock Wall Clocks large , glow-in-the-dark numbers make
|
| 128 |
+
this clock easy to read , even at night ! with attractive styling , the wall clock
|
| 129 |
+
features a round face with luminous numbers that glow automatically when the lights
|
| 130 |
+
are off . overallheight-toptobottom:11.5|overallwidth-sidetoside:11.5|operatingmechanism
|
| 131 |
+
: mechanical|dsprimaryproductstyle : modern|shape : round|dssecondaryproductstyle
|
| 132 |
+
: transitional modern|theme : no theme|timedisplay : analog : numerical|specialfeatures
|
| 133 |
+
: none|overalldepth-fronttoback:1.75|supplierintend'
|
| 134 |
+
pipeline_tag: sentence-similarity
|
| 135 |
+
library_name: sentence-transformers
|
| 136 |
+
---
|
| 137 |
+
|
| 138 |
+
# SentenceTransformer based on BAAI/bge-m3
|
| 139 |
+
|
| 140 |
+
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [BAAI/bge-m3](https://huggingface.co/BAAI/bge-m3). It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, classification, clustering, and more.
|
| 141 |
+
|
| 142 |
+
## Model Details
|
| 143 |
+
|
| 144 |
+
### Model Description
|
| 145 |
+
- **Model Type:** Sentence Transformer
|
| 146 |
+
- **Base model:** [BAAI/bge-m3](https://huggingface.co/BAAI/bge-m3) <!-- at revision 5617a9f61b028005a4858fdac845db406aefb181 -->
|
| 147 |
+
- **Maximum Sequence Length:** 8192 tokens
|
| 148 |
+
- **Output Dimensionality:** 1024 dimensions
|
| 149 |
+
- **Similarity Function:** Cosine Similarity
|
| 150 |
+
- **Supported Modality:** Text
|
| 151 |
+
<!-- - **Training Dataset:** Unknown -->
|
| 152 |
+
<!-- - **Language:** Unknown -->
|
| 153 |
+
<!-- - **License:** Unknown -->
|
| 154 |
+
|
| 155 |
+
### Model Sources
|
| 156 |
+
|
| 157 |
+
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
|
| 158 |
+
- **Repository:** [Sentence Transformers on GitHub](https://github.com/huggingface/sentence-transformers)
|
| 159 |
+
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
|
| 160 |
+
|
| 161 |
+
### Full Model Architecture
|
| 162 |
+
|
| 163 |
+
```
|
| 164 |
+
SentenceTransformer(
|
| 165 |
+
(0): Transformer({'transformer_task': 'feature-extraction', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'last_hidden_state'}}, 'module_output_name': 'token_embeddings', 'architecture': 'XLMRobertaModel'})
|
| 166 |
+
(1): Pooling({'embedding_dimension': 1024, 'pooling_mode': 'cls', 'include_prompt': True})
|
| 167 |
+
(2): Normalize({})
|
| 168 |
+
)
|
| 169 |
+
```
|
| 170 |
+
|
| 171 |
+
## Usage
|
| 172 |
+
|
| 173 |
+
### Direct Usage (Sentence Transformers)
|
| 174 |
+
|
| 175 |
+
First install the Sentence Transformers library:
|
| 176 |
+
|
| 177 |
+
```bash
|
| 178 |
+
pip install -U sentence-transformers
|
| 179 |
+
```
|
| 180 |
+
Then you can load this model and run inference.
|
| 181 |
+
```python
|
| 182 |
+
from sentence_transformers import SentenceTransformer
|
| 183 |
+
|
| 184 |
+
# Download from the 🤗 Hub
|
| 185 |
+
model = SentenceTransformer("sentence_transformers_model_id")
|
| 186 |
+
# Run inference
|
| 187 |
+
sentences = [
|
| 188 |
+
'compact garage storage shed under 3 feet wide',
|
| 189 |
+
'outdoor 2 ft. 9 in . w x 2 ft. d plastic vertical tool shed Sheds with its tall profile , this 22 cubic foot vertical shed is great for storing ladders , long-handled tools , garden accessories , and more when space is at a premium . the durable , all-weather resin construction provides water resistance and uv protection , while the multi-wall panels are engineere interiorwidth-sidetoside:27|interiorheight-toptobottom:69|compatibleshelfpartnumber : xa1182|windrating:65|overallwidth-sidetoside:33|countryoforigin : united states|interiordepth-fronttoback:20.25|levelofassembly : full assembly needed|countryoforigin-additionaldetails : made in usa|doorwidth-sidet',
|
| 190 |
+
'garden plastic/metal storage shed Sheds easily add storage space and value to your property with an outdoor garden shed . the metal shed has distinct advantages over traditional wooden models that make it a much more attractive choice for purchase . a metal shed is more affordable and being pre-fabricated significantly cuts down on expens dsmetallic : steel|shedtype : storage shed|color : grey|construction : traditional|overallheight-toptobottom:75.5|countryoforigin-additionaldetails : made in china|warrantylength:1 year|levelofassembly : full assembly needed|overallwidth-sidetoside:126.75|overalldepth-fronttoback:109|overallproductw',
|
| 191 |
+
]
|
| 192 |
+
embeddings = model.encode(sentences)
|
| 193 |
+
print(embeddings.shape)
|
| 194 |
+
# [3, 1024]
|
| 195 |
+
|
| 196 |
+
# Get the similarity scores for the embeddings
|
| 197 |
+
similarities = model.similarity(embeddings, embeddings)
|
| 198 |
+
print(similarities)
|
| 199 |
+
# tensor([[1.0000, 0.5560, 0.5587],
|
| 200 |
+
# [0.5560, 1.0000, 0.5424],
|
| 201 |
+
# [0.5587, 0.5424, 1.0000]])
|
| 202 |
+
```
|
| 203 |
+
<!--
|
| 204 |
+
### Direct Usage (Transformers)
|
| 205 |
+
|
| 206 |
+
<details><summary>Click to see the direct usage in Transformers</summary>
|
| 207 |
+
|
| 208 |
+
</details>
|
| 209 |
+
-->
|
| 210 |
+
|
| 211 |
+
<!--
|
| 212 |
+
### Downstream Usage (Sentence Transformers)
|
| 213 |
+
|
| 214 |
+
You can finetune this model on your own dataset.
|
| 215 |
+
|
| 216 |
+
<details><summary>Click to expand</summary>
|
| 217 |
+
|
| 218 |
+
</details>
|
| 219 |
+
-->
|
| 220 |
+
|
| 221 |
+
<!--
|
| 222 |
+
### Out-of-Scope Use
|
| 223 |
+
|
| 224 |
+
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
|
| 225 |
+
-->
|
| 226 |
+
|
| 227 |
+
<!--
|
| 228 |
+
## Bias, Risks and Limitations
|
| 229 |
+
|
| 230 |
+
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
|
| 231 |
+
-->
|
| 232 |
+
|
| 233 |
+
<!--
|
| 234 |
+
### Recommendations
|
| 235 |
+
|
| 236 |
+
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
|
| 237 |
+
-->
|
| 238 |
+
|
| 239 |
+
## Training Details
|
| 240 |
+
|
| 241 |
+
### Training Dataset
|
| 242 |
+
|
| 243 |
+
#### Unnamed Dataset
|
| 244 |
+
|
| 245 |
+
* Size: 284,573 training samples
|
| 246 |
+
* Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
|
| 247 |
+
* Approximate statistics based on the first 100 samples:
|
| 248 |
+
| | anchor | positive | negative |
|
| 249 |
+
|:---------|:---------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|
|
| 250 |
+
| type | string | string | string |
|
| 251 |
+
| modality | text | text | text |
|
| 252 |
+
| details | <ul><li>min: 7 tokens</li><li>mean: 10.2 tokens</li><li>max: 16 tokens</li></ul> | <ul><li>min: 115 tokens</li><li>mean: 173.42 tokens</li><li>max: 196 tokens</li></ul> | <ul><li>min: 97 tokens</li><li>mean: 168.43 tokens</li><li>max: 205 tokens</li></ul> |
|
| 253 |
+
* Samples:
|
| 254 |
+
| anchor | positive | negative |
|
| 255 |
+
|:------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
|
| 256 |
+
| <code>modern 10 inch semi-flush mount light for entryway</code> | <code>bucholz 1 -light 10 '' semi flush mount Flush Mount Lighting this 1-light semi-flush mount has a sleek , streamlined shape that gives a modern look to your hallway , entryway , or over your dining room table . it has a metal frame and downrod , and an egg-shaped glass shade with an opening at the bottom for light to shine through . the downrod retracts and ro shadematerial : glass\|dry , damporwetlocationlisted : dry\|whatisdrydamporwetlocationlisted : this indicates whether the fixture is safe to use in dry locations , damp locations ( moist environments ) , or wet locations ( direct exposure to water ) .\|estimatedtimetosetup:5\|bodyheight-toptobottom:10\|w</code> | <code>weatherford 1 - light 10 '' semi flush mount Flush Mount Lighting marrying modern minimalism and factory flair , this understated one-light flush mount brings a splash of industrial style as it shines a light down on your space . crafted of metal , this fixture features a round canopy and a conical shade with its upper portion left open to provide a peek at the so canopywidth-sidetoside:4.875\|recommendedbulbshapecode : a19\|warrantylength:1 year\|canopydepth-fronttoback:0.75\|powersource : hardwired\|dssecondaryproductstyle : nautical\|canopyheight-toptobottom:4.875\|recommendedbulbshape : standard\|producttype : flush mount\|whatisullisted : the underwriters laborat</code> |
|
| 257 |
+
| <code>sleek glass pendant light for dining room</code> | <code>bucholz 1 -light 10 '' semi flush mount Flush Mount Lighting this 1-light semi-flush mount has a sleek , streamlined shape that gives a modern look to your hallway , entryway , or over your dining room table . it has a metal frame and downrod , and an egg-shaped glass shade with an opening at the bottom for light to shine through . the downrod retracts and ro shadematerial : glass\|dry , damporwetlocationlisted : dry\|whatisdrydamporwetlocationlisted : this indicates whether the fixture is safe to use in dry locations , damp locations ( moist environments ) , or wet locations ( direct exposure to water ) .\|estimatedtimetosetup:5\|bodyheight-toptobottom:10\|w</code> | <code>6 '' h glass sphere pendant shade ( screw on ) in smoke color : smoke\|shape : sphere\|overallproductweight:0.67\|overalldepth-fronttoback:6\|theme : no theme\|attachmenttype : screw on\|style : modern & contemporary\|overallwidth-sidetoside:6\|fittersize:3.25\|producttype : pendant shade\|primarymaterial : glass\|overallheight-toptobottom:6\| : no\|fitterincluded : </code> |
|
| 258 |
+
| <code>small metal frame flush mount light with glass shade</code> | <code>bucholz 1 -light 10 '' semi flush mount Flush Mount Lighting this 1-light semi-flush mount has a sleek , streamlined shape that gives a modern look to your hallway , entryway , or over your dining room table . it has a metal frame and downrod , and an egg-shaped glass shade with an opening at the bottom for light to shine through . the downrod retracts and ro shadematerial : glass\|dry , damporwetlocationlisted : dry\|whatisdrydamporwetlocationlisted : this indicates whether the fixture is safe to use in dry locations , damp locations ( moist environments ) , or wet locations ( direct exposure to water ) .\|estimatedtimetosetup:5\|bodyheight-toptobottom:10\|w</code> | <code>3 - light semi flush mount Flush Mount Lighting\|Commercial Ceiling Lighting it is perfect to transform your dining room or entryway with this charming transitional flush mount ceiling light . bodydepth-fronttoback:16.1\|producttype : semi flush mount\|recommendedbulbshape : candle\|bodydepth-fronttoback:12.2\|lightdirection : up\|warrantylength:2 years\|dssecondaryproductstyle : classic glam\|dsprimaryproductstyle : glam\|fixturedesign : unique/statement\|craftsmanshiptype : handcrafted / hand-fo</code> |
|
| 259 |
+
* Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
|
| 260 |
+
```json
|
| 261 |
+
{
|
| 262 |
+
"scale": 20.0,
|
| 263 |
+
"similarity_fct": "cos_sim",
|
| 264 |
+
"gather_across_devices": false,
|
| 265 |
+
"directions": [
|
| 266 |
+
"query_to_doc"
|
| 267 |
+
],
|
| 268 |
+
"partition_mode": "joint",
|
| 269 |
+
"hardness_mode": null,
|
| 270 |
+
"hardness_strength": 0.0
|
| 271 |
+
}
|
| 272 |
+
```
|
| 273 |
+
|
| 274 |
+
### Evaluation Dataset
|
| 275 |
+
|
| 276 |
+
#### Unnamed Dataset
|
| 277 |
+
|
| 278 |
+
* Size: 14,978 evaluation samples
|
| 279 |
+
* Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
|
| 280 |
+
* Approximate statistics based on the first 100 samples:
|
| 281 |
+
| | anchor | positive | negative |
|
| 282 |
+
|:---------|:----------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|
|
| 283 |
+
| type | string | string | string |
|
| 284 |
+
| modality | text | text | text |
|
| 285 |
+
| details | <ul><li>min: 8 tokens</li><li>mean: 11.03 tokens</li><li>max: 18 tokens</li></ul> | <ul><li>min: 110 tokens</li><li>mean: 159.21 tokens</li><li>max: 193 tokens</li></ul> | <ul><li>min: 90 tokens</li><li>mean: 168.18 tokens</li><li>max: 201 tokens</li></ul> |
|
| 286 |
+
* Samples:
|
| 287 |
+
| anchor | positive | negative |
|
| 288 |
+
|:--------------------------------------------------------------|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
|
| 289 |
+
| <code>quirky rooster shelf sitters</code> | <code>4 piece rooster shelf sitter figurine set Decorative Objects add these colorful rooster sitters to shelves , windowsills , and more throughout your home . the set includes four hand-painted roosters with delightful floral accents . each cute sitter has dangling legs and plenty of charm . overallwidth-sidetoside:1.63\|geometricshapes : no geometric & shapes\|producttype : figurine\|subject : animals\|nature : no nature\|spiritualreligious : no spiritual & religious\|structuresbuildings : no structures & buildings\|wordstext : no words & text\|overallproductweight:0.25\|quantity:4\|entertainmen</code> | <code>dunellen farmhouse rooster 2 piece figurine set Decorative Objects be the early bird in grabbing these farmhouse silver metal and wood rooster sculpture by the head and complete your farmhouse set up now . this 2-piece set features two small and medium standing rooster sculptures with a gray finish on the body , and silver on the rest . this will look great as a ce overallwidth-sidetoside:13\|color : brown , silver\|purposefuldistressingtype : distressed metal\|style : farmhouse / country\|overallheight-toptobottom:5.1\|fantasysci-fi : no fantasy & sci-fi\|overallheight-toptobottom:16\|nature : no nature\|geometricshapes : no geometric & shapes\|spiritualreligious : no</code> |
|
| 290 |
+
| <code>decorative rooster figurines with floral accents</code> | <code>4 piece rooster shelf sitter figurine set Decorative Objects add these colorful rooster sitters to shelves , windowsills , and more throughout your home . the set includes four hand-painted roosters with delightful floral accents . each cute sitter has dangling legs and plenty of charm . overallwidth-sidetoside:1.63\|geometricshapes : no geometric & shapes\|producttype : figurine\|subject : animals\|nature : no nature\|spiritualreligious : no spiritual & religious\|structuresbuildings : no structures & buildings\|wordstext : no words & text\|overallproductweight:0.25\|quantity:4\|entertainmen</code> | <code>2 piece metal figurine set Decorative Objects the antique style of these two metal rooster decors will bring a fascinating appeal to any rustic , vintage , or antique accent table . with its iron construction , these rooster sculptures will continue to enhance the look of your home interior for years to come . featuring a decorative rustic bron geometricshapes : no geometric & shapes\|spiritualreligious : no spiritual & religious\|overallwidth-sidetoside:20.47\|producttype : figurine\|nature : no nature\|foodbeverage : no food & beverage\|fantasysci-fi : no fantasy & sci-fi\|people : no people\|agegroup : adult\|animals : rooster\|overallproductweig</code> |
|
| 291 |
+
| <code>comfortable anti-fatigue mat for kitchen counter</code> | <code>catrisha soothing foam comfort anti-fatigue mat Kitchen Mats this anti-fatigue mat provides a full inch of soothing foam comfort for your tired feet . perfect in front of a kitchen counter and has a skid-resistant backing . color : red\|pattern : solid color\|backingmaterial : polyester\|color : brown\|shape : rectangle\|supplierintendedandapproveduse : non residential use\|productcare : machine washable\|construction : machine made\|color : blue\|dssecondaryproductstyle : transitional modern\|overalllength-endtoend:30\|color : g</code> | <code>grano hello standing anti-fatigue mat Kitchen Mats use this mat in any part of your home . whether it be the kitchen , laundry room , bathroom , or doorway this elegant mat will bring luscious texture , enduring material , and comfortable steps for your convenience . this perfectly crafted pvc mat is joined with a rubber backing for long-lasting usa pattern : no pattern and not solid color\|overalllength-endtoend:30\|overalllength-endtoend:47\|overallproductweight:2.09\|holidayoccasion : no holiday\|dssecondaryproductstyle : modern farmhouse\|overallheight-toptobottom:0.47\|material : synthetics\|theme : text\|materialdetails:100 % pvc\|overallwidth-side</code> |
|
| 292 |
+
* Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
|
| 293 |
+
```json
|
| 294 |
+
{
|
| 295 |
+
"scale": 20.0,
|
| 296 |
+
"similarity_fct": "cos_sim",
|
| 297 |
+
"gather_across_devices": false,
|
| 298 |
+
"directions": [
|
| 299 |
+
"query_to_doc"
|
| 300 |
+
],
|
| 301 |
+
"partition_mode": "joint",
|
| 302 |
+
"hardness_mode": null,
|
| 303 |
+
"hardness_strength": 0.0
|
| 304 |
+
}
|
| 305 |
+
```
|
| 306 |
+
|
| 307 |
+
### Training Hyperparameters
|
| 308 |
+
#### Non-Default Hyperparameters
|
| 309 |
+
|
| 310 |
+
- `per_device_train_batch_size`: 64
|
| 311 |
+
- `num_train_epochs`: 1
|
| 312 |
+
- `learning_rate`: 2e-05
|
| 313 |
+
- `warmup_steps`: 0.1
|
| 314 |
+
- `bf16`: True
|
| 315 |
+
- `per_device_eval_batch_size`: 64
|
| 316 |
+
- `dataloader_num_workers`: 4
|
| 317 |
+
|
| 318 |
+
#### All Hyperparameters
|
| 319 |
+
<details><summary>Click to expand</summary>
|
| 320 |
+
|
| 321 |
+
- `per_device_train_batch_size`: 64
|
| 322 |
+
- `num_train_epochs`: 1
|
| 323 |
+
- `max_steps`: -1
|
| 324 |
+
- `learning_rate`: 2e-05
|
| 325 |
+
- `lr_scheduler_type`: linear
|
| 326 |
+
- `lr_scheduler_kwargs`: None
|
| 327 |
+
- `warmup_steps`: 0.1
|
| 328 |
+
- `optim`: adamw_torch_fused
|
| 329 |
+
- `optim_args`: None
|
| 330 |
+
- `weight_decay`: 0.0
|
| 331 |
+
- `adam_beta1`: 0.9
|
| 332 |
+
- `adam_beta2`: 0.999
|
| 333 |
+
- `adam_epsilon`: 1e-08
|
| 334 |
+
- `optim_target_modules`: None
|
| 335 |
+
- `gradient_accumulation_steps`: 1
|
| 336 |
+
- `average_tokens_across_devices`: True
|
| 337 |
+
- `max_grad_norm`: 1.0
|
| 338 |
+
- `label_smoothing_factor`: 0.0
|
| 339 |
+
- `bf16`: True
|
| 340 |
+
- `fp16`: False
|
| 341 |
+
- `bf16_full_eval`: False
|
| 342 |
+
- `fp16_full_eval`: False
|
| 343 |
+
- `tf32`: None
|
| 344 |
+
- `gradient_checkpointing`: False
|
| 345 |
+
- `gradient_checkpointing_kwargs`: None
|
| 346 |
+
- `torch_compile`: False
|
| 347 |
+
- `torch_compile_backend`: None
|
| 348 |
+
- `torch_compile_mode`: None
|
| 349 |
+
- `use_liger_kernel`: False
|
| 350 |
+
- `liger_kernel_config`: None
|
| 351 |
+
- `use_cache`: False
|
| 352 |
+
- `neftune_noise_alpha`: None
|
| 353 |
+
- `torch_empty_cache_steps`: None
|
| 354 |
+
- `auto_find_batch_size`: False
|
| 355 |
+
- `log_on_each_node`: True
|
| 356 |
+
- `logging_nan_inf_filter`: True
|
| 357 |
+
- `include_num_input_tokens_seen`: no
|
| 358 |
+
- `log_level`: passive
|
| 359 |
+
- `log_level_replica`: warning
|
| 360 |
+
- `disable_tqdm`: False
|
| 361 |
+
- `project`: huggingface
|
| 362 |
+
- `trackio_space_id`: None
|
| 363 |
+
- `trackio_bucket_id`: None
|
| 364 |
+
- `trackio_static_space_id`: None
|
| 365 |
+
- `per_device_eval_batch_size`: 64
|
| 366 |
+
- `prediction_loss_only`: True
|
| 367 |
+
- `eval_on_start`: False
|
| 368 |
+
- `eval_do_concat_batches`: True
|
| 369 |
+
- `eval_use_gather_object`: False
|
| 370 |
+
- `eval_accumulation_steps`: None
|
| 371 |
+
- `include_for_metrics`: []
|
| 372 |
+
- `batch_eval_metrics`: False
|
| 373 |
+
- `save_only_model`: False
|
| 374 |
+
- `save_on_each_node`: False
|
| 375 |
+
- `enable_jit_checkpoint`: False
|
| 376 |
+
- `push_to_hub`: False
|
| 377 |
+
- `hub_private_repo`: None
|
| 378 |
+
- `hub_model_id`: None
|
| 379 |
+
- `hub_strategy`: every_save
|
| 380 |
+
- `hub_always_push`: False
|
| 381 |
+
- `hub_revision`: None
|
| 382 |
+
- `load_best_model_at_end`: False
|
| 383 |
+
- `ignore_data_skip`: False
|
| 384 |
+
- `restore_callback_states_from_checkpoint`: False
|
| 385 |
+
- `full_determinism`: False
|
| 386 |
+
- `seed`: 42
|
| 387 |
+
- `data_seed`: None
|
| 388 |
+
- `use_cpu`: False
|
| 389 |
+
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
|
| 390 |
+
- `parallelism_config`: None
|
| 391 |
+
- `dataloader_drop_last`: False
|
| 392 |
+
- `dataloader_num_workers`: 4
|
| 393 |
+
- `dataloader_pin_memory`: True
|
| 394 |
+
- `dataloader_persistent_workers`: False
|
| 395 |
+
- `dataloader_prefetch_factor`: None
|
| 396 |
+
- `remove_unused_columns`: True
|
| 397 |
+
- `label_names`: None
|
| 398 |
+
- `train_sampling_strategy`: random
|
| 399 |
+
- `length_column_name`: length
|
| 400 |
+
- `ddp_find_unused_parameters`: None
|
| 401 |
+
- `ddp_bucket_cap_mb`: None
|
| 402 |
+
- `ddp_broadcast_buffers`: False
|
| 403 |
+
- `ddp_static_graph`: None
|
| 404 |
+
- `ddp_backend`: None
|
| 405 |
+
- `ddp_timeout`: 1800
|
| 406 |
+
- `fsdp`: None
|
| 407 |
+
- `fsdp_config`: None
|
| 408 |
+
- `deepspeed`: None
|
| 409 |
+
- `debug`: []
|
| 410 |
+
- `skip_memory_metrics`: True
|
| 411 |
+
- `do_predict`: False
|
| 412 |
+
- `resume_from_checkpoint`: None
|
| 413 |
+
- `warmup_ratio`: None
|
| 414 |
+
- `local_rank`: -1
|
| 415 |
+
- `prompts`: None
|
| 416 |
+
- `batch_sampler`: batch_sampler
|
| 417 |
+
- `multi_dataset_batch_sampler`: proportional
|
| 418 |
+
- `router_mapping`: {}
|
| 419 |
+
- `learning_rate_mapping`: {}
|
| 420 |
+
|
| 421 |
+
</details>
|
| 422 |
+
|
| 423 |
+
### Training Logs
|
| 424 |
+
| Epoch | Step | Training Loss | Validation Loss |
|
| 425 |
+
|:------:|:----:|:-------------:|:---------------:|
|
| 426 |
+
| 0.0225 | 100 | 1.1596 | - |
|
| 427 |
+
| 0.0450 | 200 | 0.6243 | - |
|
| 428 |
+
| 0.0675 | 300 | 0.5484 | - |
|
| 429 |
+
| 0.0899 | 400 | 0.4963 | - |
|
| 430 |
+
| 0.1124 | 500 | 0.4665 | 0.4796 |
|
| 431 |
+
| 0.1349 | 600 | 0.4426 | - |
|
| 432 |
+
| 0.1574 | 700 | 0.4353 | - |
|
| 433 |
+
| 0.1799 | 800 | 0.4284 | - |
|
| 434 |
+
| 0.2024 | 900 | 0.4111 | - |
|
| 435 |
+
| 0.2249 | 1000 | 0.3946 | 0.4274 |
|
| 436 |
+
| 0.2474 | 1100 | 0.3883 | - |
|
| 437 |
+
| 0.2698 | 1200 | 0.3806 | - |
|
| 438 |
+
| 0.2923 | 1300 | 0.3590 | - |
|
| 439 |
+
| 0.3148 | 1400 | 0.3661 | - |
|
| 440 |
+
| 0.3373 | 1500 | 0.3632 | 0.4086 |
|
| 441 |
+
| 0.3598 | 1600 | 0.3717 | - |
|
| 442 |
+
| 0.3823 | 1700 | 0.3517 | - |
|
| 443 |
+
| 0.4048 | 1800 | 0.3428 | - |
|
| 444 |
+
| 0.4273 | 1900 | 0.3574 | - |
|
| 445 |
+
| 0.4497 | 2000 | 0.3412 | 0.3882 |
|
| 446 |
+
| 0.4722 | 2100 | 0.3340 | - |
|
| 447 |
+
| 0.4947 | 2200 | 0.3356 | - |
|
| 448 |
+
| 0.5172 | 2300 | 0.3261 | - |
|
| 449 |
+
| 0.5397 | 2400 | 0.3358 | - |
|
| 450 |
+
| 0.5622 | 2500 | 0.3208 | 0.3833 |
|
| 451 |
+
| 0.5847 | 2600 | 0.3206 | - |
|
| 452 |
+
| 0.6072 | 2700 | 0.3254 | - |
|
| 453 |
+
| 0.6296 | 2800 | 0.3215 | - |
|
| 454 |
+
| 0.6521 | 2900 | 0.3082 | - |
|
| 455 |
+
| 0.6746 | 3000 | 0.3099 | 0.3684 |
|
| 456 |
+
| 0.6971 | 3100 | 0.3060 | - |
|
| 457 |
+
| 0.7196 | 3200 | 0.3009 | - |
|
| 458 |
+
| 0.7421 | 3300 | 0.3021 | - |
|
| 459 |
+
| 0.7646 | 3400 | 0.3163 | - |
|
| 460 |
+
| 0.7870 | 3500 | 0.3020 | 0.3652 |
|
| 461 |
+
| 0.8095 | 3600 | 0.3012 | - |
|
| 462 |
+
| 0.8320 | 3700 | 0.2927 | - |
|
| 463 |
+
| 0.8545 | 3800 | 0.2878 | - |
|
| 464 |
+
| 0.8770 | 3900 | 0.2974 | - |
|
| 465 |
+
| 0.8995 | 4000 | 0.3125 | 0.3584 |
|
| 466 |
+
| 0.9220 | 4100 | 0.2938 | - |
|
| 467 |
+
| 0.9445 | 4200 | 0.2997 | - |
|
| 468 |
+
| 0.9669 | 4300 | 0.2915 | - |
|
| 469 |
+
| 0.9894 | 4400 | 0.2923 | - |
|
| 470 |
+
| 1.0 | 4447 | - | 0.3572 |
|
| 471 |
+
|
| 472 |
+
|
| 473 |
+
### Training Time
|
| 474 |
+
- **Training**: 26.3 minutes
|
| 475 |
+
- **Evaluation**: 3.2 minutes
|
| 476 |
+
- **Total**: 29.5 minutes
|
| 477 |
+
|
| 478 |
+
### Framework Versions
|
| 479 |
+
- Python: 3.12.13
|
| 480 |
+
- Sentence Transformers: 5.7.0
|
| 481 |
+
- Transformers: 5.14.1
|
| 482 |
+
- PyTorch: 2.13.0+cu130
|
| 483 |
+
- Accelerate: 1.14.0
|
| 484 |
+
- Datasets: 5.0.1
|
| 485 |
+
- Tokenizers: 0.22.2
|
| 486 |
+
|
| 487 |
+
## Additional Resources
|
| 488 |
+
|
| 489 |
+
- [Training and Finetuning Embedding Models with Sentence Transformers](https://huggingface.co/blog/train-sentence-transformers): the end-to-end guide for training or finetuning Sentence Transformer models.
|
| 490 |
+
- [Introduction to Matryoshka Embedding Models](https://huggingface.co/blog/matryoshka): variable-size embeddings that can be truncated with minimal quality loss.
|
| 491 |
+
- [Binary and Scalar Embedding Quantization for Significantly Faster & Cheaper Retrieval](https://huggingface.co/blog/embedding-quantization): post-training compression of embedding vectors.
|
| 492 |
+
- [Multimodal Embedding & Reranker Models with Sentence Transformers](https://huggingface.co/blog/multimodal-sentence-transformers): use text, image, audio, and video models through the same API.
|
| 493 |
+
- [Training and Finetuning Multimodal Embedding & Reranker Models with Sentence Transformers](https://huggingface.co/blog/train-multimodal-sentence-transformers): train multimodal embedding models, with a Visual Document Retrieval walkthrough.
|
| 494 |
+
|
| 495 |
+
## Citation
|
| 496 |
+
|
| 497 |
+
### BibTeX
|
| 498 |
+
|
| 499 |
+
#### Sentence Transformers
|
| 500 |
+
```bibtex
|
| 501 |
+
@inproceedings{reimers-2019-sentence-bert,
|
| 502 |
+
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
|
| 503 |
+
author = "Reimers, Nils and Gurevych, Iryna",
|
| 504 |
+
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
|
| 505 |
+
month = "11",
|
| 506 |
+
year = "2019",
|
| 507 |
+
publisher = "Association for Computational Linguistics",
|
| 508 |
+
url = "https://arxiv.org/abs/1908.10084",
|
| 509 |
+
}
|
| 510 |
+
```
|
| 511 |
+
|
| 512 |
+
#### MultipleNegativesRankingLoss
|
| 513 |
+
```bibtex
|
| 514 |
+
@misc{oord2019representationlearningcontrastivepredictive,
|
| 515 |
+
title={Representation Learning with Contrastive Predictive Coding},
|
| 516 |
+
author={Aaron van den Oord and Yazhe Li and Oriol Vinyals},
|
| 517 |
+
year={2019},
|
| 518 |
+
eprint={1807.03748},
|
| 519 |
+
archivePrefix={arXiv},
|
| 520 |
+
primaryClass={cs.LG},
|
| 521 |
+
url={https://arxiv.org/abs/1807.03748},
|
| 522 |
+
}
|
| 523 |
+
```
|
| 524 |
+
|
| 525 |
+
<!--
|
| 526 |
+
## Glossary
|
| 527 |
+
|
| 528 |
+
*Clearly define terms in order to be accessible across audiences.*
|
| 529 |
+
-->
|
| 530 |
+
|
| 531 |
+
<!--
|
| 532 |
+
## Model Card Authors
|
| 533 |
+
|
| 534 |
+
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
|
| 535 |
+
-->
|
| 536 |
+
|
| 537 |
+
<!--
|
| 538 |
+
## Model Card Contact
|
| 539 |
+
|
| 540 |
+
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
|
| 541 |
+
-->
|
config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_cross_attention": false,
|
| 3 |
+
"architectures": [
|
| 4 |
+
"XLMRobertaModel"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"dtype": "float32",
|
| 10 |
+
"eos_token_id": 2,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.1,
|
| 13 |
+
"hidden_size": 1024,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 4096,
|
| 16 |
+
"is_decoder": false,
|
| 17 |
+
"layer_norm_eps": 1e-05,
|
| 18 |
+
"max_position_embeddings": 8194,
|
| 19 |
+
"model_type": "xlm-roberta",
|
| 20 |
+
"num_attention_heads": 16,
|
| 21 |
+
"num_hidden_layers": 24,
|
| 22 |
+
"output_past": true,
|
| 23 |
+
"pad_token_id": 1,
|
| 24 |
+
"position_embedding_type": "absolute",
|
| 25 |
+
"tie_word_embeddings": true,
|
| 26 |
+
"transformers_version": "5.14.1",
|
| 27 |
+
"type_vocab_size": 1,
|
| 28 |
+
"use_cache": false,
|
| 29 |
+
"vocab_size": 250002
|
| 30 |
+
}
|
config_sentence_transformers.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"__version__": {
|
| 3 |
+
"pytorch": "2.13.0+cu130",
|
| 4 |
+
"sentence_transformers": "5.7.0",
|
| 5 |
+
"transformers": "5.14.1"
|
| 6 |
+
},
|
| 7 |
+
"default_prompt_name": null,
|
| 8 |
+
"model_type": "SentenceTransformer",
|
| 9 |
+
"prompts": {
|
| 10 |
+
"document": "",
|
| 11 |
+
"query": ""
|
| 12 |
+
},
|
| 13 |
+
"similarity_fn_name": "cosine"
|
| 14 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d70ae60d470094fbae7b06b78f72a04098811adeb278db5d96ec37ce23e2f986
|
| 3 |
+
size 2271064456
|
modules.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"idx": 0,
|
| 4 |
+
"name": "0",
|
| 5 |
+
"path": "",
|
| 6 |
+
"type": "sentence_transformers.base.modules.transformer.Transformer"
|
| 7 |
+
},
|
| 8 |
+
{
|
| 9 |
+
"idx": 1,
|
| 10 |
+
"name": "1",
|
| 11 |
+
"path": "1_Pooling",
|
| 12 |
+
"type": "sentence_transformers.sentence_transformer.modules.pooling.Pooling"
|
| 13 |
+
},
|
| 14 |
+
{
|
| 15 |
+
"idx": 2,
|
| 16 |
+
"name": "2",
|
| 17 |
+
"path": "2_Normalize",
|
| 18 |
+
"type": "sentence_transformers.sentence_transformer.modules.normalize.Normalize"
|
| 19 |
+
}
|
| 20 |
+
]
|
sentence_bert_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"transformer_task": "feature-extraction",
|
| 3 |
+
"modality_config": {
|
| 4 |
+
"text": {
|
| 5 |
+
"method": "forward",
|
| 6 |
+
"method_output_name": "last_hidden_state"
|
| 7 |
+
}
|
| 8 |
+
},
|
| 9 |
+
"module_output_name": "token_embeddings"
|
| 10 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:326c392f5799a1a73fb5118bac1e78bfc02ad3ca7f04c002cda9915721218c7d
|
| 3 |
+
size 17098339
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": true,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<s>",
|
| 5 |
+
"clean_up_tokenization_spaces": true,
|
| 6 |
+
"cls_token": "<s>",
|
| 7 |
+
"eos_token": "</s>",
|
| 8 |
+
"is_local": false,
|
| 9 |
+
"local_files_only": false,
|
| 10 |
+
"mask_token": "<mask>",
|
| 11 |
+
"model_max_length": 8192,
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"sp_model_kwargs": {},
|
| 15 |
+
"tokenizer_class": "XLMRobertaTokenizer",
|
| 16 |
+
"unk_token": "<unk>"
|
| 17 |
+
}
|