--- base_model: BAAI/bge-m3 datasets: [] language: [] library_name: sentence-transformers metrics: - cosine_accuracy - dot_accuracy - manhattan_accuracy - euclidean_accuracy - max_accuracy pipeline_tag: sentence-similarity tags: - sentence-transformers - sentence-similarity - feature-extraction - generated_from_trainer - dataset_size:39836 - loss:MultipleNegativesRankingLoss widget: - source_sentence: Seorang pria bertopi biru dan rompi keselamatan oranye berdiri di persimpangan sambil memegang bendera. sentences: - Sekelompok orang menaiki eskalator, banyak dari mereka memegang payung. - Seseorang berpakaian agar mudah terlihat. - Seorang pria mengenakan topi keras oranye berdiri di persimpangan jalan. - source_sentence: Dua anjing saling memandang di luar. sentences: - Ada dua anjing di luar. - Empat anjing saling memandang di dalam. - Seorang pria di luar gedung bata merah dengan kereta belanja, sepeda, dan lain-lain. - source_sentence: Pria itu berdiri. sentences: - Seorang pria dan wanita duduk bersama di meja. - Orang-orang di pasar petani luar ruangan. - Seorang pria di kota di luar gedung berdiri di tangga. - source_sentence: Seorang pria sedang tidur. sentences: - Seorang pria berselimut sedang tertidur di trotoar. - Manusia ditutupi spons beraneka warna. - Seorang pria tunawisma tertidur di trotoar. - source_sentence: Orang-orang ada di luar. sentences: - Seorang pria berbaju kotak-kotak dan sandal putih sedang tertidur sambil membaca koran. - Orang-orang berjalan di luar dan mengenakan warna gelap. - Sekelompok orang sedang makan di sebuah restoran dengan mural seorang wanita sedang berbelanja di belakang mereka. model-index: - name: SentenceTransformer based on BAAI/bge-m3 results: - task: type: triplet name: Triplet dataset: name: model evaluation type: model-evaluation metrics: - type: cosine_accuracy value: 0.9636322566071832 name: Cosine Accuracy - type: dot_accuracy value: 0.03636774339281681 name: Dot Accuracy - type: manhattan_accuracy value: 0.9625028235825616 name: Manhattan Accuracy - type: euclidean_accuracy value: 0.9636322566071832 name: Euclidean Accuracy - type: max_accuracy value: 0.9636322566071832 name: Max Accuracy --- # SentenceTransformer based on BAAI/bge-m3 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, text classification, clustering, and more. ## Model Details ### Model Description - **Model Type:** Sentence Transformer - **Base model:** [BAAI/bge-m3](https://huggingface.co/BAAI/bge-m3) - **Maximum Sequence Length:** 8192 tokens - **Output Dimensionality:** 1024 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': 8192, 'do_lower_case': False}) with Transformer model: XLMRobertaModel (1): Pooling({'word_embedding_dimension': 1024, '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}) (2): Normalize() ) ``` ## 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("MarcoAland/Indo-bge-m3") # Run inference sentences = [ 'Orang-orang ada di luar.', 'Orang-orang berjalan di luar dan mengenakan warna gelap.', 'Sekelompok orang sedang makan di sebuah restoran dengan mural seorang wanita sedang berbelanja di belakang mereka.', ] embeddings = model.encode(sentences) print(embeddings.shape) # [3, 1024] # Get the similarity scores for the embeddings similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] ``` ## Evaluation ### Metrics #### Triplet * Dataset: `model-evaluation` * Evaluated with [TripletEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.TripletEvaluator) | Metric | Value | |:-------------------|:-----------| | cosine_accuracy | 0.9636 | | dot_accuracy | 0.0364 | | manhattan_accuracy | 0.9625 | | euclidean_accuracy | 0.9636 | | **max_accuracy** | **0.9636** | ## Training Details ### Training Dataset #### Unnamed Dataset * Size: 39,836 training samples * Columns: anchor, positive, and negative * Approximate statistics based on the first 1000 samples: | | anchor | positive | negative | |:--------|:---------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------| | type | string | string | string | | details | | | | * Samples: | anchor | positive | negative | |:-------------------------------------------|:-----------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------------| | Seseorang sedang tidur. | Seorang pemuda tidur siang di jendela sebuah bisnis di pinggir jalan. | Seseorang duduk di kursi yang digantung dengan rantai di taman hiburan. | | Seekor anjing sedang berlari. | Seekor anjing abu-abu berlari di sepanjang rumput hijau. | Seekor anjing coklat sedang menatap anjing coklat dan putih yang sedang tidur. | | Seorang bayi menangis. | Seorang bayi menangis di tempat tidur bayi. | Seorang bayi berbaring telentang dan tersenyum. | * Loss: [MultipleNegativesRankingLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters: ```json { "scale": 20.0, "similarity_fct": "cos_sim" } ``` ### Evaluation Dataset #### Unnamed Dataset * Size: 4,427 evaluation samples * Columns: anchor, positive, and negative * Approximate statistics based on the first 1000 samples: | | anchor | positive | negative | |:--------|:---------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------| | type | string | string | string | | details | | | | * Samples: | anchor | positive | negative | |:-----------------------------------------------|:-----------------------------------------------------------------|:------------------------------------------------------------------------------| | Seorang pria sedang tidur. | Seorang pria tidur di rumput di taman. | Seorang pria membaca koran di samping mobil. | | Seorang pria sedang membaca buku. | Seorang pria tua duduk di luar sambil membaca buku. | Seorang pria berbaju pelangi berhenti untuk melihat kotak surat. | | Anjing coklat melangkah di air. | Anjing coklat berjalan di air saat dia basah kuyup | Anjing coklat sedang tidur di samping air | * Loss: [MultipleNegativesRankingLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters: ```json { "scale": 20.0, "similarity_fct": "cos_sim" } ``` ### Training Hyperparameters #### Non-Default Hyperparameters - `eval_strategy`: steps - `per_device_train_batch_size`: 4 - `per_device_eval_batch_size`: 4 - `max_steps`: 500 - `warmup_ratio`: 0.1 - `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`: 4 - `per_device_eval_batch_size`: 4 - `per_gpu_train_batch_size`: None - `per_gpu_eval_batch_size`: None - `gradient_accumulation_steps`: 1 - `eval_accumulation_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`: 3.0 - `max_steps`: 500 - `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`: 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 - `batch_sampler`: no_duplicates - `multi_dataset_batch_sampler`: proportional
### Training Logs | Epoch | Step | Training Loss | loss | model-evaluation_max_accuracy | |:------:|:----:|:-------------:|:------:|:-----------------------------:| | 0.0100 | 100 | 0.7797 | 0.6925 | - | | 0.0201 | 200 | 0.6337 | 0.6018 | - | | 0.0301 | 300 | 0.6129 | 0.5737 | - | | 0.0402 | 400 | 0.5982 | 0.5116 | - | | 0.0502 | 500 | 0.5504 | 0.4719 | 0.9636 | ### Framework Versions - Python: 3.10.12 - Sentence Transformers: 3.0.1 - Transformers: 4.42.4 - PyTorch: 2.3.1+cu121 - Accelerate: 0.32.1 - Datasets: 2.20.0 - 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", } ``` #### MultipleNegativesRankingLoss ```bibtex @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} } ```