--- tags: - sentence-transformers - sentence-similarity - feature-extraction - generated_from_trainer - dataset_size:11113 - loss:CosineSimilarityLoss base_model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 widget: - source_sentence: Эр цэргийн баярыг тохиолдуулан тоглолт болно sentences: - Тоглолт эрчүүдэд зориулагдсан. - Бие даасан тоглолт болно. - Уг тоглолтыг Атлантад төвтэй хүмүүнлэгийн байгууллагаас зохион байгуулж байгаа ажээ. - source_sentence: Энэ ном нь МУИС-ийн оюутны бүтээл юм sentences: - Тоглолтонд дэлхийн алдартай хийлч оролцоно. - Дөрвөн жилийн өмнө Нью-Йоркийн Филармонийн уран бүтээлчид Пхеньян хотод тоглолт хийж байсан. - Энэ ном нь утга зохиолын шүүмжлэлийн салбарт хувь нэмэр оруулжээ. - source_sentence: “Paprika” кино нь Японы анимэйшн, шинжлэх ухаан, танин мэдэхүйн бүтээл юм sentences: - “Paprika” киноны найруулагчаар Satoshi Kon ажиллажээ - Элчин сайд Ш.Алтангэрэл арга хэмжээг зохион байгуулсан хүмүүст талархал илэрхийллээ. - Уран бүтээлчдийн санаачилгаар. - source_sentence: Пэрэнлэйн Нямлхагва бол Д.Нацагдоржийн шагналт яруу найрагч юм sentences: - ',“Ньюанс” хамтлагийн тоглолт ОХУ-ын олон хотод болсон.' - Зохиол нь өгүүлэмж сайтай, утгын найруулгын хувьд сонгодог - Пэрэнлэйн Нямлхагвагийн тоглолт удахгүй болно - source_sentence: Тэд “Зэрлэг цэцэгсийн хүлэмж” кино хийсэн sentences: - Наадам "Давалгаа" тайзан дээр болно. - Францын агуу дуучин эмэгтэй Эдит Пиаф. - Тэдний хоёр дахь бүтээл “Зүрхээр наадагч” кино юм. pipeline_tag: sentence-similarity library_name: sentence-transformers metrics: - pearson_cosine - spearman_cosine model-index: - name: SentenceTransformer based on sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 results: - task: type: semantic-similarity name: Semantic Similarity dataset: name: sts dev type: sts-dev metrics: - type: pearson_cosine value: 0.5175384750107511 name: Pearson Cosine - type: spearman_cosine value: 0.48776566583188496 name: Spearman Cosine --- # SentenceTransformer based on sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2) on the csv dataset. 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](https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2) - **Maximum Sequence Length:** 128 tokens - **Output Dimensionality:** 384 dimensions - **Similarity Function:** Cosine Similarity - **Training Dataset:** - csv ### 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': 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: ```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("gmunkhtur/paraphrase-multilingual-minilm-l12-v3-mn") # Run inference sentences = [ 'Тэд “Зэрлэг цэцэгсийн хүлэмж” кино хийсэн', 'Тэдний хоёр дахь бүтээл “Зүрхээр наадагч” кино юм.', 'Францын агуу дуучин эмэгтэй Эдит Пиаф.', ] 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] ``` ## Evaluation ### Metrics #### Semantic Similarity * Dataset: `sts-dev` * Evaluated with [EmbeddingSimilarityEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator) | Metric | Value | |:--------------------|:-----------| | pearson_cosine | 0.5175 | | **spearman_cosine** | **0.4878** | ## Training Details ### Training Dataset #### csv * Dataset: csv * Size: 11,113 training samples * Columns: sentence1, sentence2, and score * Approximate statistics based on the first 1000 samples: | | sentence1 | sentence2 | score | |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------| | type | string | string | float | | details | | | | * Samples: | sentence1 | sentence2 | score | |:-------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------|:--------------------------------| | Тэднээс 6 нь л ажиллагаатай юм | ,Sothеbey’s-ийн зарсан төхөөрөмж ажилладаг гэж Sothеbey’s-ийн сайтан дээрх барааны танилцуулгад бичсэн байна. | 0.0607873052358627 | | Түүний хит дуу нь "Миллион алых розe" | Тэрбээр "Миллион алых розe" дууг олон хэлээр дуулсан. | 0.7139256596565247 | | “Нисванис” хамтлаг олон арга хэмжээ зохион байгуулдаг | Хамтлагийн тоглолт нь рок цугларалт, шоу юм | 0.454399824142456 | * Loss: [CosineSimilarityLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosinesimilarityloss) with these parameters: ```json { "loss_fct": "torch.nn.modules.loss.MSELoss" } ``` ### Evaluation Dataset #### csv * Dataset: csv * Size: 11,113 evaluation samples * Columns: sentence1, sentence2, and score * Approximate statistics based on the first 1000 samples: | | sentence1 | sentence2 | score | |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:------------------------------------------------------------------| | type | string | string | float | | details | | | | * Samples: | sentence1 | sentence2 | score | |:--------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------|:--------------------------------| | Америк хөгжмийн зохиолуудыг бэлтгэл хийж байж тоглодог | Монгол, Америк хоёр улсын хөгжмийн нотны тэмдэглэгээ өөр. | 0.5598467588424683 | | Энэхүү хэсэг нь дөрвөн ангитай байна. | Зохиолч Б.Мөнхболд энэхүү хөгжмийг бичихдээ Галдан бошготийн түүхийг таван жил судласан. | 0.2354833334684372 | | Д.Шинэцог удахгүй тоглолт хийхээр төлөвлөж байна | Д.Шинэцог Монголын шилдэг хөгжмийн зохиолчдын бүтээлийг тоглоно | 0.3461045920848846 | * Loss: [CosineSimilarityLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosinesimilarityloss) with these parameters: ```json { "loss_fct": "torch.nn.modules.loss.MSELoss" } ``` ### Training Hyperparameters #### Non-Default Hyperparameters - `eval_strategy`: steps - `per_device_train_batch_size`: 16 - `per_device_eval_batch_size`: 16 - `num_train_epochs`: 1 - `warmup_ratio`: 0.1 - `fp16`: True - `batch_sampler`: no_duplicates #### 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`: 1.0 - `num_train_epochs`: 1 - `max_steps`: -1 - `lr_scheduler_type`: linear - `lr_scheduler_kwargs`: {} - `warmup_ratio`: 0.1 - `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`: None - `hub_always_push`: False - `gradient_checkpointing`: False - `gradient_checkpointing_kwargs`: None - `include_inputs_for_metrics`: False - `include_for_metrics`: [] - `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 - `use_liger_kernel`: False - `eval_use_gather_object`: False - `average_tokens_across_devices`: False - `prompts`: None - `batch_sampler`: no_duplicates - `multi_dataset_batch_sampler`: proportional
### Training Logs | Epoch | Step | Training Loss | Validation Loss | sts-dev_spearman_cosine | |:------:|:----:|:-------------:|:---------------:|:-----------------------:| | 0 | 0 | - | - | 0.2082 | | 0.1799 | 100 | 0.0433 | - | - | | 0.3597 | 200 | 0.0367 | - | - | | 0.5396 | 300 | 0.033 | - | - | | 0.7194 | 400 | 0.0351 | - | - | | 0.8993 | 500 | 0.0312 | 0.0329 | 0.4878 | ### Framework Versions - Python: 3.10.12 - Sentence Transformers: 3.3.1 - Transformers: 4.47.1 - PyTorch: 2.5.1+cu121 - Accelerate: 1.2.1 - Datasets: 3.2.0 - Tokenizers: 0.21.0 ## 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", } ```