--- library_name: sentence-transformers pipeline_tag: sentence-similarity tags: - sentence-transformers - sentence-similarity - feature-extraction - generated_from_trainer - dataset_size:1830 - loss:TripletLoss widget: - source_sentence: 100% Cotton Throw Blanket for Couch Sofa Bed Outdoors Hypoallergenic 83"x70" Brown sentences: - '80''s Lace Headband Costume Accessories for 80s Theme Party, No Headache Neon Lace Bow Headband, Set of 4 Product Description Read more ' - ' ' - 'Our Family Recipes Journal ' - source_sentence: '"Belkin QODE Ultimate Pro Keyboard Case for iPad Air 2 White"' sentences: - 'Detectorcatty 18 inch Foil Star Shape Balloon Helium Metallic Birthday Summer Outdoor Party Wedding Decor Air Mylar Balloon ' - ' ' - 'iPad Keyboard Case for iPad 2018 (6th Gen) - iPad 2017 (5th Gen) - iPad Pro 9.7 - iPad Air 2 & 1 - Thin & Light - 360 Rotatable - Wireless/BT - Backlit 10 Color - iPad Case with Keyboard (Silver) Product Description Compatibility: iPad Air (Wi-Fi Only) | A1474 : MD785LL/A, MD786LL/A, MD787LL/A, MD898LL/A, MD788LL/A, MD789LL/A, MD790LL/A, ME906LL/A iPad Air (Wi-Fi/Cellular) | A1475 : ME991LL/A, MF003LL/A, MF009LL/A, MF015LL/A, ME997LL/A, MF529LL/A, MF012LL/A, MF018LL/A, MF020LL/A, MF024LL/A, MF026LL/A, MF028LL/A, MF021LL/A, MF025LL/A, MF027LL/A, MF029LL/A, MF496LL/A, MF520LL/A, MF534LL/A, MF558LL/A, MF502LL/A, MF527LL/A, MF539LL/A, MF563LL/A, ME993LL/A, MF004LL/A, MF010LL/A, MF016LL/A, ME999LL/A, MF532LL/A, MF013LL/A, MF019LL/A iPad Air (Wi-Fi/TD-LTE - China) | A1476 : MD785CH/A, MD768CH/A, MD787CH/A, ME898CH/A, MD788CH/A, MD789CH/A, MD790CH/A, ME906CH/A iPad Air 2 (Wi-Fi Only) | A1566 : MGLW2LL/A, MGKM2LL/A, MGTY2LL/A, MH0W2LL/A, MH182LL/A, MH1J2LL/A, MGL12LL/A, MGKL2LL/A, MGTX2LL/A, MNV62LL/A, MNV72LL/A, MNV22LL/A iPad Air 2 (Wi-Fi/Cellular) | A1567 : MH2V2LL/A, MH2N2LL/A, MH322LL/A, MH2W2LL/A, MH2P2LL/A, MH332LL/A, MH2U2LL/A, MH2M2LL/A, MH312LL/A, MNW22LL/A, MNW32LL/A, MNW12LL/A iPad Pro 9.7" (Wi-Fi Only) | A1673 : MLMP2LL/A, MLMW2LL/A, MLN02LL/A, MLMN2LL/A, MLMV2LL/A, MLMY2LL/A, MLMQ2LL/A, MLMX2LL/A, MLN12LL/A, MM172LL/A, MM192LL/A, MM1A2LL/A iPad Pro 9.7" (Wi-Fi/Cellular) | A1674 : MLPX2LL/A, MLQ42LL/A, MLQ72LL/A, MLPW2LL/A, MLQ32LL/A, MLQ62LL/A, MLPY2LL/A, MLQ52LL/A, MLQ82LL/A, MLYJ2LL/A, MLYL2LL/A, MLYM2LL/A iPad 9.7" 5th Gen (Wi-Fi Only) | A1822 : MP2G2LL/A, MP2J2LL/A, MPGT2LL/A, MPGW2LL/A, MP2F2LL/A, MP2H2LL/A iPad 9.7" 5th Gen (Wi-Fi/Cellular) | A1823 : MP252LL/A, MP2E2LL/A, MPGA2LL/A, MPGC2LL/A, MP242LL/A, MP2D2LL/A iPad 9.7" 6th Gen (Wi-Fi Only) | A1893 : MR7G2LL/A, MR7K2LL/A, MRJN2LL/A, MRJP2LL/A, MR7F2LL/A, MR7J2LL/A iPad 9.7" 6th Gen (Wi-Fi/Cellular) | A1954 : MR702LL/A, MR7D2LL/A, MRM52LL/A, MRM82LL/A, MR6Y2LL/A, MR7C2LL/A ' - source_sentence: onlypuff Pocket Shirts for Women Casual sentences: - 'Womens Hoodies Zip Up Sweatshirt Lightweight Loose Jackets with Pockets Rose Red XL ' - 'WOODWORKING FOR BEGINNERS: DIY Project Plans, Step-by-Step Guide to Learn the Best Techniques, Tools, Safety Precautions and Tips to Start Your First Projects with Illustrations and Much More! ' - 'Roorsily Transparent Face Covering Unisex Face_Shields Integrated Protective Glasses Goggles, Male and Female Transparent Face Covering, Kitchen Anti-Sputum, Sneeze, and Oil Splash Protection Panels ' - source_sentence: '"XR Extinction Rebellion Rebel For Life T-Shirt"' sentences: - 'Raw Paws Soft-Tip Pet Grooming Gloves for Dogs & Cats - Cat Deshedding Glove - Cat Gloves for Grooming - Horse Grooming Gloves - Dog Deshedding Glove - Cat Grooming Glove, Dog Brush Glove for Shedding Product Description Raw Paws dog and cat glove brush for shedding and grooming is the perfect solution for your pet hair needs! Regular brushing reduces shedding and increases blood circulation, which supports healthy skin and a shiny coat. Our glove brush for dogs is gentle enough to feel like a massage to your pet, but effective enough to remove excess hair and debris. This is a product you and your furry friends will both appreciate! Our pet mitt makes a great pet hair removal tool for pets who may not love traditional brushes. You can also use them as grooming gloves for rabbits, ferrets and horses, in addition to cats and dogs. Simply put on the hair remover mitten, tighten the wrist strap to fit your hand, and pet your animal as you normally would. Once the pet brush glove is full of fur, easily peel the accumulation of hair from the mitt in one heap and throw it away. Spend quality time with your pets while keeping them and your house clean with Raw Paws grooming mitts! Raw Paws Pet Food is a family-owned business that believes the best chance of having a happy pet is through high quality nutrition and pet supplies. That''s why we use only pet-safe materials to make our cat gloves for grooming. When you shop with Raw Paws Pet Food you can have peace of mind knowing that you''re using high quality pet shedding grooming gloves on your pet! Pet Shedding Grooming Gloves FROM OUR FAMILY TO YOURS: We only use responsible and ethical sources, so you get the highest-quality products! We hand inspect and ship these pet shedding gloves from our own Indianapolis warehouse! EXCELLENT BATH TIME TOOLS: : These mitts make great dog bath brush gloves! Use a pet grooming mitt prior to bath time to get rid of loose and dirt. During their bath, relax your pet while getting them squeaky clean with our dog brushing gloves! PERFECT FOR ALL YOUR PETS: Raw Paws animal grooming gloves will work on almost any furry friend in your home or on your farm. Use as a ferret brush, rabbit grooming glove or grooming mitt for horses. Great for pets with long hair, short hair, curly locks or coarse strands! Reduce Shedding Great for Bathing Adjustable Wrist Strap 259 Silicone Tips Gently Massage Your Pet Get Hard-to-Reach Places Support Healthy Skin & Coat For Dogs, Cats, Rabbits, Horses Read more We are a pet-rescuing and pet-loving organization that is committed to green shipping practices as well as maintaining an office and warehouse environment that is friendly to the earth. We are a small, family owned American company. We support shelters, charities and non-profits. We love your pets too! WHY RAW PAWS PET FOOD? I started Raw Paws Pet Food to make it practical, affordable and accessible for pet parents to provide their dogs and cats with healthy pet food, treats, chews, whole food toppers and supplements. At Raw Paws Pet Food, we believe that the best chance of having a happy, healthy pet is through high quality nutrition. That''s why we use only fresh, all-natural ingredients sourced from responsible and ethical farms. ~Shelli McDonald , Raw Paws Pet Food Owner When you shop with Raw Paws Pet Food, you can have peace of mind knowing that you''re giving your pet only the best. We strive to ensure that you and your pet are truly happy. We stand behind our brand, and value your business. Meeting your expectations is our #1 priority. Read more ' - 'Mens Floral Chakras V-dye T-Shirt, 3XL Vee Rainbow ' - ' ' - source_sentence: 100% Cotton Throw Blanket for Couch Sofa Bed Outdoors Hypoallergenic 83"x70" Brown sentences: - 'Chapstick Key Chain Holder with Clip Portable Lip Balm Cloth Holder Case ' - 'Hetao 100% Cotton Handmade Crochet Round Tablecloth Doilies Lace Table Covers,Beige, 27 Inch ' - 'Life Clothing Co. Womens Tops Jade Tie Dye Hoodie (XL) ' --- # SentenceTransformer This is a [sentence-transformers](https://www.SBERT.net) model trained. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more. ## Model Details ### Model Description - **Model Type:** Sentence Transformer - **Maximum Sequence Length:** 512 tokens - **Output Dimensionality:** 768 tokens - **Similarity Function:** Cosine Similarity ### Model Sources - **Documentation:** [Sentence Transformers Documentation](https://sbert.net) - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers) - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers) ### Full Model Architecture ``` SentenceTransformer( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: DistilBertModel (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True}) ) ``` ## Usage ### Direct Usage (Sentence Transformers) First install the Sentence Transformers library: ```bash pip install -U sentence-transformers ``` Then you can load this model and run inference. ```python from sentence_transformers import SentenceTransformer # Download from the 🤗 Hub model = SentenceTransformer("sentence_transformers_model_id") # Run inference sentences = [ '100% Cotton Throw Blanket for Couch Sofa Bed Outdoors Hypoallergenic 83"x70" Brown', 'Hetao 100% Cotton Handmade Crochet Round Tablecloth Doilies Lace Table Covers,Beige, 27 Inch\n', 'Life Clothing Co. Womens Tops Jade Tie Dye Hoodie (XL)\n', ] embeddings = model.encode(sentences) print(embeddings.shape) # [3, 768] # Get the similarity scores for the embeddings similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] ``` ## Training Details ### Training Dataset #### Unnamed Dataset * Size: 1,830 training samples * Columns: sentence_0, sentence_1, and sentence_2 * Approximate statistics based on the first 1000 samples: | | sentence_0 | sentence_1 | sentence_2 | |:--------|:----------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------| | type | string | string | string | | details | | | | * Samples: | sentence_0 | sentence_1 | sentence_2 | |:----------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| | "acrylic lollipop holder cake pop stand with sticks, bags, and twist ties" | Zealax 15pcs Treat Bags Gold Polka Dot Print Drawstring Plastic Party Favors for Cookie Roasting Treat Candy Buffet Gift Wrapping Goodies Package, 4.6 inches x 6.7 inches
 | Twinkle Star Solid Brass Heavy Duty Adjustable Twist Hose Nozzle Jet Sweeper Nozzle, TWIS3231
Product Description Shut Off Valve Shut Off Valve Adjustable Hose Nozzle Adjustable Twist Hose Nozzle Jet Sweeper Screw Threads 3/4" 3/4" 3/4" 3/4" 3/4" Shut-Off Valve YES YES YES YES NO Material Brass Brass Brass Brass Brass Package Includes 2 Pack 1 Pack 1 Pack 2 Pack 2 Pack Garden Hose Quick Connect Set Garden Hose Quick Connect Set Hose Caps Double Female Swivel Connectors Double Male Quick Connectors Screw Threads 3/4" 3/4" 3/4" 3/4" 3/4" Material Aluminum Brass Brass Brass Brass Package Includes 4 Sets 4 Sets 4 Pack 2 Pack 2 Pack Specifications: Body Material: Brass Package Includes: 1 x adjustable nozzle, 1 x jet sweeper nozzle. Twinkle Star Solid Brass Heavy Duty Adjustable Twist Hose Nozzle Jet Sweeper Nozzle Heavy-duty solid brass construction. With 4 holes at the tip for maximum pressure and water flow, fitted with O-ring seals at the back and front to prevent any leaks. Twisting barrel to adjusts water from a fine mist to a powerful jet stream. Fits standard garden hose, great for watering gardens, car washing, deck, siding & driveway cleaning and more. Notes: 1. Please choose the correct hose size. 2. To prevent leakage, make sure the jet has rubber ring. 3. If water leaks after a long period of use, please replace with a new washer. Read more Adjustable jet rotates from a light stream to powerful stream. Heavy duty brass 3/4”female thread. Solid brass integral inner core, anti - damage, anti - rust, anti - leakage, durable. Read more | | NFL Women's OTS Fleece Hoodie | Ultra Game NFL womens Fleece Hoodie Pullover Sweatshirt Tie Neck
 | Womens Antler Evolution Whitetail Tee Short Sleeve
 | | "Belkin QODE Ultimate Pro Keyboard Case for iPad Air 2 White" | iPad Pro Guide
 | Under Armour Boys' Prototype Short
From the manufacturer Read more Read more Read more | * Loss: [TripletLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#tripletloss) with these parameters: ```json { "distance_metric": "TripletDistanceMetric.EUCLIDEAN", "triplet_margin": 25.0 } ``` ### Training Hyperparameters #### Non-Default Hyperparameters - `per_device_train_batch_size`: 54 - `per_device_eval_batch_size`: 54 - `num_train_epochs`: 5 - `multi_dataset_batch_sampler`: round_robin #### All Hyperparameters
Click to expand - `overwrite_output_dir`: False - `do_predict`: False - `eval_strategy`: no - `prediction_loss_only`: True - `per_device_train_batch_size`: 54 - `per_device_eval_batch_size`: 54 - `per_gpu_train_batch_size`: None - `per_gpu_eval_batch_size`: None - `gradient_accumulation_steps`: 1 - `eval_accumulation_steps`: None - `torch_empty_cache_steps`: None - `learning_rate`: 5e-05 - `weight_decay`: 0.0 - `adam_beta1`: 0.9 - `adam_beta2`: 0.999 - `adam_epsilon`: 1e-08 - `max_grad_norm`: 1 - `num_train_epochs`: 5 - `max_steps`: -1 - `lr_scheduler_type`: linear - `lr_scheduler_kwargs`: {} - `warmup_ratio`: 0.0 - `warmup_steps`: 0 - `log_level`: passive - `log_level_replica`: warning - `log_on_each_node`: True - `logging_nan_inf_filter`: True - `save_safetensors`: True - `save_on_each_node`: False - `save_only_model`: False - `restore_callback_states_from_checkpoint`: False - `no_cuda`: False - `use_cpu`: False - `use_mps_device`: False - `seed`: 42 - `data_seed`: None - `jit_mode_eval`: False - `use_ipex`: False - `bf16`: False - `fp16`: False - `fp16_opt_level`: O1 - `half_precision_backend`: auto - `bf16_full_eval`: False - `fp16_full_eval`: False - `tf32`: None - `local_rank`: 0 - `ddp_backend`: None - `tpu_num_cores`: None - `tpu_metrics_debug`: False - `debug`: [] - `dataloader_drop_last`: False - `dataloader_num_workers`: 0 - `dataloader_prefetch_factor`: None - `past_index`: -1 - `disable_tqdm`: False - `remove_unused_columns`: True - `label_names`: None - `load_best_model_at_end`: False - `ignore_data_skip`: False - `fsdp`: [] - `fsdp_min_num_params`: 0 - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False} - `fsdp_transformer_layer_cls_to_wrap`: None - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None} - `deepspeed`: None - `label_smoothing_factor`: 0.0 - `optim`: adamw_torch - `optim_args`: None - `adafactor`: False - `group_by_length`: False - `length_column_name`: length - `ddp_find_unused_parameters`: None - `ddp_bucket_cap_mb`: None - `ddp_broadcast_buffers`: False - `dataloader_pin_memory`: True - `dataloader_persistent_workers`: False - `skip_memory_metrics`: True - `use_legacy_prediction_loop`: False - `push_to_hub`: False - `resume_from_checkpoint`: None - `hub_model_id`: None - `hub_strategy`: every_save - `hub_private_repo`: False - `hub_always_push`: False - `gradient_checkpointing`: False - `gradient_checkpointing_kwargs`: None - `include_inputs_for_metrics`: False - `eval_do_concat_batches`: True - `fp16_backend`: auto - `push_to_hub_model_id`: None - `push_to_hub_organization`: None - `mp_parameters`: - `auto_find_batch_size`: False - `full_determinism`: False - `torchdynamo`: None - `ray_scope`: last - `ddp_timeout`: 1800 - `torch_compile`: False - `torch_compile_backend`: None - `torch_compile_mode`: None - `dispatch_batches`: None - `split_batches`: None - `include_tokens_per_second`: False - `include_num_input_tokens_seen`: False - `neftune_noise_alpha`: None - `optim_target_modules`: None - `batch_eval_metrics`: False - `eval_on_start`: False - `eval_use_gather_object`: False - `batch_sampler`: batch_sampler - `multi_dataset_batch_sampler`: round_robin
### Framework Versions - Python: 3.10.12 - Sentence Transformers: 3.1.1 - Transformers: 4.44.2 - PyTorch: 2.4.1+cu121 - Accelerate: 0.34.2 - Datasets: 3.0.0 - Tokenizers: 0.19.1 ## Citation ### BibTeX #### Sentence Transformers ```bibtex @inproceedings{reimers-2019-sentence-bert, title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks", author = "Reimers, Nils and Gurevych, Iryna", booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing", month = "11", year = "2019", publisher = "Association for Computational Linguistics", url = "https://arxiv.org/abs/1908.10084", } ``` #### TripletLoss ```bibtex @misc{hermans2017defense, title={In Defense of the Triplet Loss for Person Re-Identification}, author={Alexander Hermans and Lucas Beyer and Bastian Leibe}, year={2017}, eprint={1703.07737}, archivePrefix={arXiv}, primaryClass={cs.CV} } ```