LilNomto/labse_oi_bo

SentenceTransformer based on buddhist-nlp/buddhist-sentence-similarity

This is a sentence-transformers model finetuned from buddhist-nlp/buddhist-sentence-similarity. 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
• Base model: buddhist-nlp/buddhist-sentence-similarity
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 768 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': 512, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 768, '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("sentence_transformers_model_id")
# Run inference
sentences = [
    'ན་མོ་དེ་དག་ལ་སོགས་པ་ཕྱོགས་བཅུ′ཨི་′ཇིག་རྟེན་གྱི་ཁམས་ཐམས་ཅད་ན༑་དེ་བཞིན་གཤེགས་པ་དགྲ་བཅོམ་པ་ཡང་དག་པར་རྫོགས་པ′ཨི་སངས་རྒྱས་བཅོམ་ལྡན་′དས་གང་ཇི་སྙེད་ཅིག་བཞུགས་ཏེ་′ཚོ་ཞིང་གཞེས་པ′ཨི་སངས་རྒྱས་བཅོམ་ལྡན་′དས་དེ་དག་ཐམས་ཅད་བདག་ལ་དགོངས་སུ་གསོལ༑',
    'tede terigüüten arban zügiyin xamuq yertüncüyin oron-du ilaγun tögüsün üleqsen tögünčilen boluqsan dayini durun sayitur dousuqsan burxad ali kedüi soun-yin tālaxui xamuq ilaγun tögüsüq/sen burxad namai ayiladun soyirxo:',
    'subudi ali bodhi-sadv-nar eyin kemēn: bi oroni zoҟōl-noγoudi bütēmüi kemēn ögüülekülē: töüni basa tögünčilen ögüülen bü üyiled:',
]
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: 3,684 training samples
• Columns: sentence_0 and sentence_1
• Approximate statistics based on the first 1000 samples:

  	sentence_0	sentence_1
  type	string	string
  details	min: 3 tokens mean: 77.68 tokens max: 512 tokens	min: 7 tokens mean: 60.33 tokens max: 418 tokens

• Samples:

  sentence_0	sentence_1
  དེ་ལྟ་བུ་དེས༑་ཐོ་རངས༑་སྔ་དྲོ༑་ཕྱི་དྲོ༑་སྲོད་དེ་ཐུན་བཞི་ལ་བསྒོམ་པར་བྱ་སྟེ༑་དེ་ཡང་དང་པོར་ཡུན་རིངས་ན་བྱིང་རྒོད་ཀྱི་དབང་དུ་འགྲོ་སླ་ཞིང༑་དེ་ལ་གོམས་ན་བློ་འཆོས་དཀའ་བས༑་ཐུན་ཆུང་ལ་གྲངས་མང་བར་བྱ༑	tere metü töügēr örlȫbür erte oroi üdeši bür tere dörbön xübidü bišilγan üyiled tere čü urida öni odxulā čibkü doqšir xoxuyin erkēr odxu kilbar bolun: töün-dü bišilxulā oyou-bēn yasaxu kerke yin tula mün: xubi oxor {olo} üyiled:
  འཛམ་བུའི་གླིང་ན་ངེད་ཀྱི་བུ༑༑་འདི་ལྟར་ཆོས་ལྡན་རྒྱལ་པོ་ཡིན༑༑་ཆོས་ཀྱི་ཡུལ་འཁོར་དེར་བསྟན་ནས༑༑་ལེགས་པར་བྱེད་ལ་སྐྱེ་བོ་བཀོད༑༑	zambutib/-tu mani küböün: ene metü nom/toi xān mün: nomiyin oron orčin tende üzüülēd: sayitur üyiledküi-dü tӧrӧlkitӧni zoҟōxu::
  ཌེ་བཞིན་གཤེགས་པ་བརྒྱད་ཀྱི་རིག་སྔགས་′དི་བརྗོད་པ་ཙམ་གྱིས་′གྲོ་བ་རིགས་དྲུག་གི་སེམས་ཅན་རྣམས་སོ་སོ′ཨི་སྡུག་བསྔལ་རྣམས་ཞི་ནས་བདེ་ལེགས་སུ་གྱུར་ཅིག༑	nayiman bodhi sadv-yin zarliq tögünčilen ireqsen züreken öü/ni ögüüleqseni tödüi-gēr zur/γān züyil xamuq amitani tus-buri/yin zobolong amurlīd: amuγuu/lang xotolo tögüskü boltuγai::

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• num_train_epochs: 40
• fp16: True
• multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• 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: 40
• 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: True
• 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

Training Logs

Epoch	Step	Training Loss
1.0	231	-
1.2987	300	-
2.0	462	-
2.1645	500	0.6161
2.5974	600	-
3.0	693	-
3.8961	900	-
4.0	924	-
4.3290	1000	0.0751
5.0	1155	-
5.1948	1200	-
6.0	1386	-
6.4935	1500	0.0292
7.0	1617	-
7.7922	1800	-
8.0	1848	-
8.6580	2000	0.0158
9.0	2079	-
9.0909	2100	-
10.0	2310	-
10.3896	2400	-
10.8225	2500	0.011
11.0	2541	-
11.6883	2700	-
12.0	2772	-
12.9870	3000	0.0102
13.0	3003	-
14.0	3234	-
14.2857	3300	-
15.0	3465	-
15.1515	3500	0.0121
15.5844	3600	-
16.0	3696	-
16.8831	3900	-
17.0	3927	-
17.3160	4000	0.0087
18.0	4158	-
18.1818	4200	-
19.0	4389	-
19.4805	4500	0.0078
20.0	4620	-
20.7792	4800	-
21.0	4851	-
21.6450	5000	0.0083
22.0	5082	-
22.0779	5100	-
23.0	5313	-
23.3766	5400	-
23.8095	5500	0.0083
24.0	5544	-
24.6753	5700	-
25.0	5775	-
25.9740	6000	0.0065
26.0	6006	-
27.0	6237	-
27.2727	6300	-
28.0	6468	-
28.1385	6500	0.0059
28.5714	6600	-
29.0	6699	-
29.8701	6900	-
30.0	6930	-
30.3030	7000	0.007
31.0	7161	-
31.1688	7200	-
32.0	7392	-
32.4675	7500	0.0058
33.0	7623	-
33.7662	7800	-
34.0	7854	-
34.6320	8000	0.0043
35.0	8085	-
35.0649	8100	-
36.0	8316	-
36.3636	8400	-
36.7965	8500	0.0044
37.0	8547	-
37.6623	8700	-
38.0	8778	-
38.9610	9000	0.0059
39.0	9009	-
40.0	9240	-

Framework Versions

• Python: 3.12.1
• Sentence Transformers: 3.1.1
• Transformers: 4.44.2
• PyTorch: 2.6.0.dev20240923+cu121
• Accelerate: 1.12.0
• Datasets: 4.4.2
• 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",
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply},
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}