---
base_model: TaylorAI/bge-micro-v2
datasets: []
language: []
library_name: sentence-transformers
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:160000
- loss:TripletLoss
widget:
- source_sentence: Corpus luteum cyst of left ovary
sentences:
- Hematoma of broad ligament
- Corpus luteum cyst of ovary, unspecified side
- Drug/chem diab with mod nonp rtnop with macular edema, bi Drug or
- source_sentence: Sltr-haris Type IV physl fx low end r tibia, 7thK Salter-Harris
Type IV
sentences:
- Sltr-haris Type IV physl fx low end r tibia, 7thG Salter-Harris Type IV physeal
- Contusion of unspecified ear, sequela
- Sltr-haris Type III physl fx low end l tibia, 7thP Salter-Harris Type III
- source_sentence: Torus fracture of upper end of left tibia, init for clos fx
sentences:
- Torus fracture of upper end of left tibia, sequela
- Torus fx upper end of unsp tibia, subs for fx w malunion Torus
- Congenital absence, atresia and stenosis of jejunum
- source_sentence: Pre-existing essential htn comp pregnancy, second trimester
sentences:
- Wear of artic bearing surface of int prosth l knee jt, subs
- Unsp pre-existing hypertension compl preg/chldbrth Unspecified pre-existing
- Pre-existing essential hypertension complicating childbirth
- source_sentence: Carbuncle, unspecified
sentences:
- Poisoning by appetite depressants, self-harm, subs Poisoning
- Cutaneous abscess, furuncle and carbuncle, unspecified
- Furuncle of neck
---
# SentenceTransformer based on TaylorAI/bge-micro-v2
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [TaylorAI/bge-micro-v2](https://huggingface.co/TaylorAI/bge-micro-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:** [TaylorAI/bge-micro-v2](https://huggingface.co/TaylorAI/bge-micro-v2)
- **Maximum Sequence Length:** 512 tokens
- **Output Dimensionality:** 384 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: 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("training")
# Run inference
sentences = [
'Carbuncle, unspecified',
'Cutaneous abscess, furuncle and carbuncle, unspecified',
'Furuncle of neck',
]
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: 160,000 training samples
* Columns: anchor, positive, and negative
* Approximate statistics based on the first 1000 samples:
| | anchor | positive | negative |
|:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
| type | string | string | string |
| details | - min: 4 tokens
- mean: 15.92 tokens
- max: 47 tokens
| - min: 4 tokens
- mean: 15.81 tokens
- max: 41 tokens
| - min: 3 tokens
- mean: 15.75 tokens
- max: 45 tokens
|
* Samples:
| anchor | positive | negative |
|:-------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------|:------------------------------------------------------------------------|
| Sudden visual loss, right eye | Sudden visual loss | Visual distortions of shape and size |
| Drug/chem diab with mild nonp rtnop without mclr edema, unsp Drug or chemical | Drug/chem diab with mod nonp rtnop with macular edema, bi Drug or | Hypostatic pneumonia, unspecified organism |
| Bronchiectasis with (acute) exacerbation | Bronchiectasis | Gestatnl htn w/o significant proteinuria, second trimester |
* Loss: [TripletLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#tripletloss) with these parameters:
```json
{
"distance_metric": "TripletDistanceMetric.EUCLIDEAN",
"triplet_margin": 5
}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `per_device_train_batch_size`: 16
- `max_steps`: 10000
#### All Hyperparameters
Click to expand
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: no
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 16
- `per_device_eval_batch_size`: 8
- `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`: 3.0
- `max_steps`: 10000
- `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`: proportional
Click to expand
| Epoch | Step | Training Loss |
|:-----:|:-----:|:-------------:|
| 0.005 | 50 | 3.9819 |
| 0.01 | 100 | 3.8181 |
| 0.015 | 150 | 3.7244 |
| 0.02 | 200 | 3.6362 |
| 0.025 | 250 | 3.5459 |
| 0.03 | 300 | 3.4653 |
| 0.035 | 350 | 3.4066 |
| 0.04 | 400 | 3.3441 |
| 0.045 | 450 | 3.3497 |
| 0.05 | 500 | 3.2625 |
| 0.055 | 550 | 3.1359 |
| 0.06 | 600 | 3.1542 |
| 0.065 | 650 | 3.1528 |
| 0.07 | 700 | 3.1634 |
| 0.075 | 750 | 3.0737 |
| 0.08 | 800 | 3.1022 |
| 0.085 | 850 | 3.0288 |
| 0.09 | 900 | 2.9434 |
| 0.095 | 950 | 2.9014 |
| 0.1 | 1000 | 3.0412 |
| 0.105 | 1050 | 2.9844 |
| 0.11 | 1100 | 2.845 |
| 0.115 | 1150 | 2.9053 |
| 0.12 | 1200 | 2.8447 |
| 0.125 | 1250 | 2.8222 |
| 0.13 | 1300 | 2.8545 |
| 0.135 | 1350 | 2.7114 |
| 0.14 | 1400 | 2.7586 |
| 0.145 | 1450 | 2.6997 |
| 0.15 | 1500 | 2.5484 |
| 0.155 | 1550 | 2.7853 |
| 0.16 | 1600 | 2.6711 |
| 0.165 | 1650 | 2.7364 |
| 0.17 | 1700 | 2.8237 |
| 0.175 | 1750 | 2.737 |
| 0.18 | 1800 | 2.7059 |
| 0.185 | 1850 | 2.6577 |
| 0.19 | 1900 | 2.777 |
| 0.195 | 1950 | 2.7369 |
| 0.2 | 2000 | 2.6317 |
| 0.205 | 2050 | 2.6678 |
| 0.21 | 2100 | 2.6889 |
| 0.215 | 2150 | 2.5734 |
| 0.22 | 2200 | 2.7214 |
| 0.225 | 2250 | 2.5059 |
| 0.23 | 2300 | 2.623 |
| 0.235 | 2350 | 2.6761 |
| 0.24 | 2400 | 2.5663 |
| 0.245 | 2450 | 2.6678 |
| 0.25 | 2500 | 2.5856 |
| 0.255 | 2550 | 2.5436 |
| 0.26 | 2600 | 2.6359 |
| 0.265 | 2650 | 2.6266 |
| 0.27 | 2700 | 2.5698 |
| 0.275 | 2750 | 2.5611 |
| 0.28 | 2800 | 2.6306 |
| 0.285 | 2850 | 2.658 |
| 0.29 | 2900 | 2.5878 |
| 0.295 | 2950 | 2.553 |
| 0.3 | 3000 | 2.5295 |
| 0.305 | 3050 | 2.5211 |
| 0.31 | 3100 | 2.6489 |
| 0.315 | 3150 | 2.6131 |
| 0.32 | 3200 | 2.7298 |
| 0.325 | 3250 | 2.5931 |
| 0.33 | 3300 | 2.5927 |
| 0.335 | 3350 | 2.5403 |
| 0.34 | 3400 | 2.4497 |
| 0.345 | 3450 | 2.6764 |
| 0.35 | 3500 | 2.5673 |
| 0.355 | 3550 | 2.6134 |
| 0.36 | 3600 | 2.6298 |
| 0.365 | 3650 | 2.5747 |
| 0.37 | 3700 | 2.6245 |
| 0.375 | 3750 | 2.5275 |
| 0.38 | 3800 | 2.5541 |
| 0.385 | 3850 | 2.5469 |
| 0.39 | 3900 | 2.452 |
| 0.395 | 3950 | 2.483 |
| 0.4 | 4000 | 2.5592 |
| 0.405 | 4050 | 2.4209 |
| 0.41 | 4100 | 2.6014 |
| 0.415 | 4150 | 2.3952 |
| 0.42 | 4200 | 2.5131 |
| 0.425 | 4250 | 2.4455 |
| 0.43 | 4300 | 2.5441 |
| 0.435 | 4350 | 2.5412 |
| 0.44 | 4400 | 2.3887 |
| 0.445 | 4450 | 2.5183 |
| 0.45 | 4500 | 2.4578 |
| 0.455 | 4550 | 2.5733 |
| 0.46 | 4600 | 2.6645 |
| 0.465 | 4650 | 2.5156 |
| 0.47 | 4700 | 2.4689 |
| 0.475 | 4750 | 2.4995 |
| 0.48 | 4800 | 2.6219 |
| 0.485 | 4850 | 2.605 |
| 0.49 | 4900 | 2.4358 |
| 0.495 | 4950 | 2.6028 |
| 0.5 | 5000 | 2.5858 |
| 0.505 | 5050 | 2.3894 |
| 0.51 | 5100 | 2.6398 |
| 0.515 | 5150 | 2.4805 |
| 0.52 | 5200 | 2.5322 |
| 0.525 | 5250 | 2.4 |
| 0.53 | 5300 | 2.4541 |
| 0.535 | 5350 | 2.5067 |
| 0.54 | 5400 | 2.5244 |
| 0.545 | 5450 | 2.5514 |
| 0.55 | 5500 | 2.4608 |
| 0.555 | 5550 | 2.5884 |
| 0.56 | 5600 | 2.4291 |
| 0.565 | 5650 | 2.6395 |
| 0.57 | 5700 | 2.3873 |
| 0.575 | 5750 | 2.652 |
| 0.58 | 5800 | 2.5328 |
| 0.585 | 5850 | 2.5713 |
| 0.59 | 5900 | 2.4961 |
| 0.595 | 5950 | 2.4438 |
| 0.6 | 6000 | 2.5537 |
| 0.605 | 6050 | 2.6323 |
| 0.61 | 6100 | 2.6427 |
| 0.615 | 6150 | 2.5648 |
| 0.62 | 6200 | 2.4444 |
| 0.625 | 6250 | 2.6298 |
| 0.63 | 6300 | 2.583 |
| 0.635 | 6350 | 2.6873 |
| 0.64 | 6400 | 2.5556 |
| 0.645 | 6450 | 2.5652 |
| 0.65 | 6500 | 2.618 |
| 0.655 | 6550 | 2.4977 |
| 0.66 | 6600 | 2.5805 |
| 0.665 | 6650 | 2.4989 |
| 0.67 | 6700 | 2.5527 |
| 0.675 | 6750 | 2.5616 |
| 0.68 | 6800 | 2.5378 |
| 0.685 | 6850 | 2.5159 |
| 0.69 | 6900 | 2.6366 |
| 0.695 | 6950 | 2.5066 |
| 0.7 | 7000 | 2.498 |
| 0.705 | 7050 | 2.5416 |
| 0.71 | 7100 | 2.5362 |
| 0.715 | 7150 | 2.5541 |
| 0.72 | 7200 | 2.5598 |
| 0.725 | 7250 | 2.4584 |
| 0.73 | 7300 | 2.6006 |
| 0.735 | 7350 | 2.5072 |
| 0.74 | 7400 | 2.4681 |
| 0.745 | 7450 | 2.4808 |
| 0.75 | 7500 | 2.5695 |
| 0.755 | 7550 | 2.5131 |
| 0.76 | 7600 | 2.5227 |
| 0.765 | 7650 | 2.5553 |
| 0.77 | 7700 | 2.4966 |
| 0.775 | 7750 | 2.4811 |
| 0.78 | 7800 | 2.5081 |
| 0.785 | 7850 | 2.5916 |
| 0.79 | 7900 | 2.4911 |
| 0.795 | 7950 | 2.5778 |
| 0.8 | 8000 | 2.5111 |
| 0.805 | 8050 | 2.5094 |
| 0.81 | 8100 | 2.5456 |
| 0.815 | 8150 | 2.5445 |
| 0.82 | 8200 | 2.5531 |
| 0.825 | 8250 | 2.6358 |
| 0.83 | 8300 | 2.5247 |
| 0.835 | 8350 | 2.4117 |
| 0.84 | 8400 | 2.5442 |
| 0.845 | 8450 | 2.537 |
| 0.85 | 8500 | 2.4553 |
| 0.855 | 8550 | 2.6114 |
| 0.86 | 8600 | 2.4397 |
| 0.865 | 8650 | 2.5667 |
| 0.87 | 8700 | 2.5281 |
| 0.875 | 8750 | 2.4894 |
| 0.88 | 8800 | 2.5723 |
| 0.885 | 8850 | 2.5952 |
| 0.89 | 8900 | 2.4053 |
| 0.895 | 8950 | 2.4827 |
| 0.9 | 9000 | 2.5784 |
| 0.905 | 9050 | 2.4545 |
| 0.91 | 9100 | 2.527 |
| 0.915 | 9150 | 2.5998 |
| 0.92 | 9200 | 2.4528 |
| 0.925 | 9250 | 2.5195 |
| 0.93 | 9300 | 2.5508 |
| 0.935 | 9350 | 2.5952 |
| 0.94 | 9400 | 2.607 |
| 0.945 | 9450 | 2.5086 |
| 0.95 | 9500 | 2.4972 |
| 0.955 | 9550 | 2.4919 |
| 0.96 | 9600 | 2.5147 |
| 0.965 | 9650 | 2.4523 |
| 0.97 | 9700 | 2.6027 |
| 0.975 | 9750 | 2.4286 |
| 0.98 | 9800 | 2.5617 |
| 0.985 | 9850 | 2.4994 |
| 0.99 | 9900 | 2.6527 |
| 0.995 | 9950 | 2.538 |
| 1.0 | 10000 | 2.4506 |