SentenceTransformer based on sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
This is a sentence-transformers model finetuned from sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2. It maps sentences & paragraphs to a 384-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
- Base model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
- Maximum Sequence Length: 128 tokens
- Output Dimensionality: 384 tokens
- Similarity Function: Cosine Similarity
Model Sources
- Documentation: Sentence Transformers Documentation
- Repository: Sentence Transformers on GitHub
- Hugging Face: Sentence Transformers on Hugging Face
Full Model Architecture
SentenceTransformer(
(0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel
(1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, '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:
pip install -U sentence-transformers
Then you can load this model and run inference.
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("annazdr/nace-pl-v2")
# Run inference
sentences = [
'manufacture of glass mirrors',
' protective face shields/visors, of plastics, e',
'manufacture of electroplating machinery',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
Training Details
Training Dataset
Unnamed Dataset
- Size: 12,822 training samples
- Columns:
sentence_0
andlabel
- Approximate statistics based on the first 1000 samples:
sentence_0 label type string int details - min: 2 tokens
- mean: 15.14 tokens
- max: 128 tokens
- 0: ~0.20%
- 1: ~0.10%
- 2: ~0.20%
- 4: ~0.30%
- 5: ~0.10%
- 6: ~0.10%
- 7: ~0.40%
- 9: ~0.10%
- 10: ~0.60%
- 11: ~0.20%
- 12: ~0.30%
- 13: ~0.30%
- 14: ~0.10%
- 15: ~0.10%
- 16: ~0.40%
- 17: ~0.10%
- 18: ~0.40%
- 20: ~0.40%
- 22: ~0.30%
- 23: ~0.30%
- 24: ~0.30%
- 25: ~0.40%
- 27: ~0.20%
- 28: ~0.10%
- 30: ~0.10%
- 32: ~0.10%
- 33: ~0.20%
- 34: ~0.10%
- 35: ~0.30%
- 37: ~0.30%
- 38: ~0.30%
- 39: ~0.30%
- 41: ~0.20%
- 42: ~0.10%
- 43: ~0.20%
- 44: ~0.50%
- 46: ~0.10%
- 48: ~0.20%
- 49: ~0.30%
- 50: ~0.30%
- 51: ~0.20%
- 52: ~0.40%
- 53: ~0.30%
- 54: ~0.20%
- 55: ~0.20%
- 56: ~0.20%
- 58: ~0.20%
- 59: ~0.10%
- 60: ~0.30%
- 61: ~0.20%
- 63: ~0.40%
- 64: ~0.30%
- 65: ~0.10%
- 66: ~0.70%
- 68: ~0.10%
- 69: ~0.20%
- 70: ~0.50%
- 71: ~0.30%
- 72: ~0.10%
- 73: ~0.40%
- 74: ~0.20%
- 75: ~0.30%
- 76: ~0.20%
- 78: ~0.10%
- 79: ~0.10%
- 80: ~0.10%
- 81: ~0.30%
- 82: ~0.30%
- 83: ~0.30%
- 84: ~0.10%
- 85: ~0.20%
- 86: ~0.20%
- 89: ~0.10%
- 90: ~0.10%
- 91: ~0.30%
- 92: ~0.20%
- 93: ~0.10%
- 94: ~0.30%
- 95: ~0.20%
- 96: ~0.20%
- 97: ~0.40%
- 98: ~0.70%
- 99: ~0.20%
- 100: ~0.50%
- 101: ~0.20%
- 102: ~0.10%
- 103: ~0.10%
- 104: ~0.20%
- 106: ~0.10%
- 108: ~0.20%
- 110: ~0.10%
- 111: ~0.10%
- 112: ~0.20%
- 115: ~0.10%
- 116: ~0.10%
- 119: ~0.30%
- 120: ~0.10%
- 121: ~0.20%
- 123: ~0.10%
- 125: ~0.20%
- 126: ~0.10%
- 127: ~0.20%
- 128: ~0.40%
- 130: ~0.20%
- 134: ~0.10%
- 135: ~0.10%
- 136: ~0.10%
- 138: ~0.10%
- 139: ~0.10%
- 140: ~0.20%
- 141: ~0.10%
- 142: ~0.10%
- 143: ~0.40%
- 144: ~0.10%
- 148: ~0.10%
- 149: ~0.10%
- 150: ~0.30%
- 151: ~0.10%
- 152: ~0.30%
- 153: ~0.40%
- 154: ~0.50%
- 156: ~0.10%
- 157: ~0.30%
- 158: ~0.20%
- 159: ~0.30%
- 160: ~0.10%
- 161: ~0.10%
- 162: ~0.10%
- 163: ~0.10%
- 165: ~0.10%
- 166: ~0.20%
- 167: ~0.20%
- 168: ~0.20%
- 170: ~0.10%
- 171: ~0.10%
- 172: ~0.10%
- 173: ~0.10%
- 174: ~0.20%
- 176: ~0.20%
- 178: ~0.10%
- 179: ~0.10%
- 181: ~0.10%
- 182: ~0.30%
- 183: ~0.30%
- 184: ~0.20%
- 185: ~0.30%
- 186: ~0.40%
- 187: ~0.20%
- 188: ~0.40%
- 189: ~0.20%
- 190: ~0.50%
- 191: ~0.30%
- 192: ~0.40%
- 193: ~0.10%
- 196: ~0.20%
- 197: ~0.20%
- 198: ~0.30%
- 199: ~0.60%
- 200: ~0.50%
- 201: ~0.10%
- 202: ~0.10%
- 203: ~0.30%
- 204: ~0.10%
- 205: ~0.30%
- 206: ~0.40%
- 208: ~0.20%
- 210: ~0.20%
- 211: ~0.40%
- 212: ~0.20%
- 214: ~0.30%
- 215: ~0.10%
- 217: ~0.30%
- 218: ~0.20%
- 220: ~0.30%
- 221: ~0.10%
- 222: ~0.20%
- 223: ~0.10%
- 225: ~0.10%
- 226: ~0.10%
- 227: ~0.20%
- 228: ~0.10%
- 230: ~0.10%
- 231: ~0.30%
- 233: ~0.10%
- 234: ~0.10%
- 235: ~0.20%
- 236: ~0.20%
- 237: ~0.20%
- 238: ~0.30%
- 239: ~0.10%
- 240: ~0.10%
- 241: ~0.20%
- 242: ~0.10%
- 243: ~0.40%
- 244: ~0.40%
- 245: ~0.20%
- 246: ~0.20%
- 247: ~0.30%
- 248: ~0.20%
- 249: ~0.20%
- 250: ~0.10%
- 253: ~0.30%
- 254: ~0.50%
- 255: ~0.30%
- 256: ~0.20%
- 257: ~0.20%
- 258: ~0.20%
- 259: ~0.10%
- 260: ~0.60%
- 261: ~0.10%
- 262: ~0.10%
- 264: ~0.30%
- 266: ~0.10%
- 267: ~0.10%
- 269: ~0.20%
- 271: ~0.10%
- 272: ~0.10%
- 273: ~0.10%
- 274: ~0.40%
- 275: ~0.10%
- 276: ~0.30%
- 277: ~0.20%
- 278: ~0.10%
- 279: ~0.20%
- 281: ~0.10%
- 283: ~0.40%
- 284: ~0.10%
- 285: ~0.20%
- 286: ~0.10%
- 287: ~0.20%
- 289: ~0.20%
- 290: ~0.20%
- 291: ~0.20%
- 292: ~0.30%
- 293: ~0.20%
- 294: ~0.20%
- 295: ~0.40%
- 296: ~0.20%
- 297: ~0.20%
- 298: ~0.10%
- 302: ~0.10%
- 303: ~0.10%
- 306: ~0.60%
- 307: ~0.50%
- 310: ~0.40%
- 311: ~0.40%
- 313: ~0.10%
- 314: ~0.40%
- 316: ~0.10%
- 319: ~0.20%
- 320: ~0.10%
- 322: ~0.50%
- 324: ~0.20%
- 325: ~0.30%
- 326: ~0.30%
- 327: ~0.10%
- 328: ~0.10%
- 329: ~0.10%
- 330: ~0.10%
- 331: ~0.10%
- 332: ~0.20%
- 334: ~0.10%
- 336: ~0.30%
- 337: ~0.50%
- 338: ~0.10%
- 341: ~0.10%
- 343: ~0.10%
- 344: ~0.20%
- 347: ~0.20%
- 348: ~0.10%
- 349: ~0.10%
- 350: ~0.50%
- 351: ~0.70%
- 352: ~0.20%
- 353: ~0.10%
- 354: ~0.20%
- 355: ~0.10%
- 356: ~0.10%
- 357: ~0.20%
- 358: ~0.30%
- 359: ~0.10%
- 360: ~0.20%
- 361: ~0.30%
- 362: ~0.10%
- 363: ~0.10%
- 364: ~0.10%
- 365: ~0.30%
- 368: ~0.30%
- 369: ~0.20%
- 372: ~0.30%
- 373: ~0.10%
- 374: ~0.30%
- 375: ~0.70%
- 376: ~0.10%
- 377: ~0.20%
- 378: ~0.20%
- 380: ~0.10%
- 381: ~0.10%
- 382: ~0.20%
- 383: ~0.10%
- 385: ~0.20%
- 393: ~0.10%
- 394: ~0.10%
- 395: ~0.20%
- 396: ~0.30%
- 398: ~0.10%
- 399: ~0.20%
- 401: ~0.20%
- 402: ~0.20%
- 404: ~0.40%
- 405: ~0.10%
- 407: ~0.20%
- 409: ~0.20%
- 410: ~0.10%
- 411: ~0.10%
- 412: ~0.10%
- 413: ~0.20%
- 414: ~0.20%
- 415: ~0.10%
- 416: ~0.10%
- 417: ~0.10%
- 418: ~0.10%
- 419: ~0.20%
- 420: ~0.10%
- 421: ~0.20%
- 423: ~0.30%
- 424: ~0.10%
- 425: ~0.10%
- 427: ~0.20%
- 428: ~0.10%
- 429: ~0.10%
- 430: ~0.10%
- 432: ~0.10%
- 434: ~0.10%
- 435: ~0.40%
- 436: ~0.20%
- 437: ~0.30%
- 438: ~0.20%
- 440: ~0.20%
- 441: ~0.30%
- 442: ~0.20%
- 443: ~0.10%
- 444: ~0.30%
- 445: ~0.20%
- 446: ~0.20%
- 448: ~0.20%
- 449: ~0.30%
- 451: ~0.20%
- 452: ~0.10%
- 454: ~0.20%
- 455: ~0.20%
- 456: ~0.10%
- 458: ~0.30%
- 459: ~0.10%
- 460: ~0.10%
- 462: ~0.10%
- 463: ~0.40%
- 464: ~0.10%
- 465: ~0.20%
- 466: ~0.10%
- 467: ~0.40%
- 468: ~0.10%
- 469: ~0.30%
- 471: ~0.10%
- 475: ~0.30%
- 476: ~0.50%
- 477: ~0.10%
- 479: ~0.40%
- 480: ~0.30%
- 482: ~0.10%
- 483: ~0.30%
- 484: ~0.10%
- 485: ~0.20%
- 486: ~0.10%
- 487: ~0.10%
- 490: ~0.30%
- 491: ~0.40%
- 492: ~0.40%
- 493: ~0.10%
- 494: ~0.10%
- 495: ~0.10%
- 498: ~0.20%
- 499: ~0.40%
- 500: ~0.30%
- 501: ~0.30%
- 502: ~0.30%
- 504: ~0.20%
- 505: ~0.20%
- 506: ~0.10%
- 507: ~0.20%
- 508: ~0.10%
- 511: ~0.10%
- 512: ~0.60%
- 513: ~0.10%
- 515: ~0.10%
- 516: ~0.30%
- 517: ~0.40%
- 519: ~0.30%
- 520: ~0.30%
- 521: ~0.10%
- 522: ~0.20%
- 523: ~0.10%
- 524: ~0.50%
- 525: ~0.60%
- 527: ~0.20%
- 528: ~0.10%
- 530: ~0.10%
- 533: ~0.40%
- 534: ~0.50%
- 535: ~0.40%
- 536: ~0.10%
- 537: ~0.20%
- 538: ~0.40%
- 539: ~0.10%
- 540: ~0.10%
- 542: ~0.30%
- 543: ~0.10%
- 544: ~0.10%
- 545: ~0.20%
- 546: ~0.20%
- 548: ~0.20%
- 549: ~0.20%
- 550: ~0.30%
- 551: ~0.30%
- 552: ~0.10%
- 554: ~0.10%
- 555: ~0.20%
- 557: ~0.20%
- 560: ~0.10%
- 561: ~0.20%
- 562: ~0.10%
- 564: ~0.40%
- 565: ~0.10%
- 566: ~0.10%
- 567: ~0.20%
- 570: ~0.10%
- 572: ~0.30%
- 573: ~0.10%
- 574: ~0.10%
- 575: ~0.10%
- 576: ~0.10%
- 577: ~0.20%
- 578: ~0.50%
- 579: ~0.40%
- 581: ~0.20%
- 585: ~0.40%
- 586: ~0.10%
- 587: ~0.20%
- 588: ~0.20%
- 590: ~0.20%
- 592: ~0.10%
- 595: ~0.10%
- 597: ~0.20%
- 600: ~0.10%
- 601: ~0.10%
- 603: ~0.10%
- 604: ~0.10%
- 608: ~0.10%
- 611: ~0.10%
- 612: ~0.20%
- 613: ~0.10%
- 619: ~0.20%
- 620: ~0.20%
- 622: ~0.10%
- 625: ~0.20%
- 629: ~0.10%
- 631: ~0.20%
- 632: ~0.10%
- 633: ~0.20%
- 634: ~0.10%
- 635: ~0.40%
- 640: ~0.10%
- 643: ~0.10%
- 645: ~0.10%
- 648: ~0.10%
- Samples:
sentence_0 label swimming clubs
475
581
this class includes: mining of ores valued chiefly for iron content
351
- Loss:
BatchAllTripletLoss
Training Hyperparameters
Non-Default Hyperparameters
per_device_train_batch_size
: 256per_device_eval_batch_size
: 256num_train_epochs
: 4multi_dataset_batch_sampler
: round_robin
All Hyperparameters
Click to expand
overwrite_output_dir
: Falsedo_predict
: Falseeval_strategy
: noprediction_loss_only
: Trueper_device_train_batch_size
: 256per_device_eval_batch_size
: 256per_gpu_train_batch_size
: Noneper_gpu_eval_batch_size
: Nonegradient_accumulation_steps
: 1eval_accumulation_steps
: Nonelearning_rate
: 5e-05weight_decay
: 0.0adam_beta1
: 0.9adam_beta2
: 0.999adam_epsilon
: 1e-08max_grad_norm
: 1num_train_epochs
: 4max_steps
: -1lr_scheduler_type
: linearlr_scheduler_kwargs
: {}warmup_ratio
: 0.0warmup_steps
: 0log_level
: passivelog_level_replica
: warninglog_on_each_node
: Truelogging_nan_inf_filter
: Truesave_safetensors
: Truesave_on_each_node
: Falsesave_only_model
: Falserestore_callback_states_from_checkpoint
: Falseno_cuda
: Falseuse_cpu
: Falseuse_mps_device
: Falseseed
: 42data_seed
: Nonejit_mode_eval
: Falseuse_ipex
: Falsebf16
: Falsefp16
: Falsefp16_opt_level
: O1half_precision_backend
: autobf16_full_eval
: Falsefp16_full_eval
: Falsetf32
: Nonelocal_rank
: 0ddp_backend
: Nonetpu_num_cores
: Nonetpu_metrics_debug
: Falsedebug
: []dataloader_drop_last
: Falsedataloader_num_workers
: 0dataloader_prefetch_factor
: Nonepast_index
: -1disable_tqdm
: Falseremove_unused_columns
: Truelabel_names
: Noneload_best_model_at_end
: Falseignore_data_skip
: Falsefsdp
: []fsdp_min_num_params
: 0fsdp_config
: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}fsdp_transformer_layer_cls_to_wrap
: Noneaccelerator_config
: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}deepspeed
: Nonelabel_smoothing_factor
: 0.0optim
: adamw_torchoptim_args
: Noneadafactor
: Falsegroup_by_length
: Falselength_column_name
: lengthddp_find_unused_parameters
: Noneddp_bucket_cap_mb
: Noneddp_broadcast_buffers
: Falsedataloader_pin_memory
: Truedataloader_persistent_workers
: Falseskip_memory_metrics
: Trueuse_legacy_prediction_loop
: Falsepush_to_hub
: Falseresume_from_checkpoint
: Nonehub_model_id
: Nonehub_strategy
: every_savehub_private_repo
: Falsehub_always_push
: Falsegradient_checkpointing
: Falsegradient_checkpointing_kwargs
: Noneinclude_inputs_for_metrics
: Falseeval_do_concat_batches
: Truefp16_backend
: autopush_to_hub_model_id
: Nonepush_to_hub_organization
: Nonemp_parameters
:auto_find_batch_size
: Falsefull_determinism
: Falsetorchdynamo
: Noneray_scope
: lastddp_timeout
: 1800torch_compile
: Falsetorch_compile_backend
: Nonetorch_compile_mode
: Nonedispatch_batches
: Nonesplit_batches
: Noneinclude_tokens_per_second
: Falseinclude_num_input_tokens_seen
: Falseneftune_noise_alpha
: Noneoptim_target_modules
: Nonebatch_eval_metrics
: Falsebatch_sampler
: batch_samplermulti_dataset_batch_sampler
: round_robin
Framework Versions
- Python: 3.10.12
- Sentence Transformers: 3.0.1
- Transformers: 4.41.2
- PyTorch: 2.3.0+cu121
- Accelerate: 0.31.0
- Datasets: 2.20.0
- Tokenizers: 0.19.1
Citation
BibTeX
Sentence Transformers
@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",
}
BatchAllTripletLoss
@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}
}
- Downloads last month
- 81
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Model tree for annazdr/nace-pl-v2
Finetuned
this model