This is a sentence-transformers model trained. It maps sentences & paragraphs to a 512-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

Maximum Sequence Length: 512 tokens
Output Dimensionality: 512 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': 512, '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.

python

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("pankajrajdeo/328500_bioformer_8L")
# Run inference
sentences = [
    'vägtrafikolyckor',
    'accidente vial',
    'trimeresurus andersoni',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 512]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Training Details

Training Dataset

Unnamed Dataset

Size: 9,358,675 training samples
Columns: anchor, positive, and negative
Approximate statistics based on the first 1000 samples:
anchor positive negative
type string string string
details
min: 6 tokens
mean: 12.84 tokens
max: 23 tokens
min: 3 tokens
mean: 15.45 tokens
max: 187 tokens
min: 3 tokens
mean: 14.75 tokens
max: 91 tokens

Samples:

anchor	positive	negative
`(131)i-makroaggregerat albumin`	`macroagrégats d'albumine humaine marquée à l'iode 131`	`1-acylglycerophosphorylinositol`
`(131)i-makroaggregerat albumin`	`albumin, radio-iodinated serum`	`allo-aromadendrane-10alpha,14-diol`
`(131)i-makroaggregerat albumin`	`serum albumin, radio iodinated`	`acquired zygomatic hyperplasia`

Loss: TripletLoss with these parameters:
json
1{ 2 "distance_metric": "TripletDistanceMetric.EUCLIDEAN", 3 "triplet_margin": 5 4}

Evaluation Dataset

Unnamed Dataset

Size: 820,102 evaluation samples
Columns: anchor, positive, and negative
Approximate statistics based on the first 1000 samples:
anchor positive negative
type string string string
details
min: 3 tokens
mean: 10.54 tokens
max: 20 tokens
min: 3 tokens
mean: 13.21 tokens
max: 183 tokens
min: 3 tokens
mean: 14.98 tokens
max: 322 tokens

Samples:

anchor	positive	negative
`15-ketosteryloleathydrolase`	`steroid esterase, lipoidal`	`glutamic acid-lysine-tyrosine terpolymer`
`15-ketosteryloleathydrolase`	`hydrolase, cholesterol ester`	`unionicola parvipora`
`15-ketosteryloleathydrolase`	`acylhydrolase, sterol ester`	`mayamaea fossalis var. fossalis`

Loss: TripletLoss with these parameters:
json
1{ 2 "distance_metric": "TripletDistanceMetric.EUCLIDEAN", 3 "triplet_margin": 5 4}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 128
learning_rate: 2e-05
num_train_epochs: 10
warmup_ratio: 0.1
fp16: True
load_best_model_at_end: True

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: steps
prediction_loss_only: True
per_device_train_batch_size: 128
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: 2e-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: 10
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: True
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
use_liger_kernel: False
eval_use_gather_object: False
batch_sampler: batch_sampler
multi_dataset_batch_sampler: proportional

Training Logs

Click to expand

Epoch	Step	Training Loss	loss
0.0137	1000	2.6865	-
0.0274	2000	1.4053	-
0.0410	3000	0.9222	-
0.0547	4000	0.7162	-
0.0684	5000	0.6036	-
0.0821	6000	0.5245	-
0.0957	7000	0.4665	-
0.1094	8000	0.4215	-
0.1231	9000	0.3931	-
0.1368	10000	0.3661	-
0.1504	11000	0.348	-
0.1641	12000	0.3241	-
0.1778	13000	0.3108	-
0.1915	14000	0.2943	-
0.2052	15000	0.2817	-
0.2188	16000	0.2653	-
0.2325	17000	0.2562	-
0.2462	18000	0.2529	-
0.2599	19000	0.2438	-
0.2735	20000	0.2359	-
0.2872	21000	0.2237	-
0.3009	22000	0.2207	-
0.3146	23000	0.2143	-
0.3283	24000	0.2141	-
0.3419	25000	0.2024	-
0.3556	26000	0.196	-
0.3693	27000	0.1951	-
0.3830	28000	0.19	-
0.3966	29000	0.1864	-
0.4103	30000	0.1866	-
0.4240	31000	0.1797	-
0.4377	32000	0.1805	-
0.4513	33000	0.1681	-
0.4650	34000	0.1712	-
0.4787	35000	0.1698	-
0.4924	36000	0.1619	-
0.4992	36500	-	0.1407
0.5061	37000	0.1652	-
0.5197	38000	0.1622	-
0.5334	39000	0.1603	-
0.5471	40000	0.1518	-
0.5608	41000	0.1488	-
0.5744	42000	0.1531	-
0.5881	43000	0.1472	-
0.6018	44000	0.1454	-
0.6155	45000	0.1473	-
0.6291	46000	0.1411	-
0.6428	47000	0.1389	-
0.6565	48000	0.1375	-
0.6702	49000	0.1393	-
0.6839	50000	0.1366	-
0.6975	51000	0.134	-
0.7112	52000	0.1331	-
0.7249	53000	0.1323	-
0.7386	54000	0.1309	-
0.7522	55000	0.1254	-
0.7659	56000	0.1298	-
0.7796	57000	0.1244	-
0.7933	58000	0.1254	-
0.8069	59000	0.1205	-
0.8206	60000	0.1213	-
0.8343	61000	0.1226	-
0.8480	62000	0.1187	-
0.8617	63000	0.1158	-
0.8753	64000	0.1171	-
0.8890	65000	0.1137	-
0.9027	66000	0.1172	-
0.9164	67000	0.1169	-
0.9300	68000	0.1137	-
0.9437	69000	0.1145	-
0.9574	70000	0.1127	-
0.9711	71000	0.1126	-
0.9848	72000	0.1126	-
0.9984	73000	0.1078	0.0997
1.0121	74000	0.0999	-
1.0258	75000	0.1001	-
1.0395	76000	0.0962	-
1.0531	77000	0.0984	-
1.0668	78000	0.0982	-
1.0805	79000	0.098	-
1.0942	80000	0.0964	-
1.1078	81000	0.0964	-
1.1215	82000	0.0949	-
1.1352	83000	0.0929	-
1.1489	84000	0.0914	-
1.1626	85000	0.0918	-
1.1762	86000	0.0916	-
1.1899	87000	0.0891	-
1.2036	88000	0.0921	-
1.2173	89000	0.0925	-
1.2309	90000	0.091	-
1.2446	91000	0.0875	-
1.2583	92000	0.0898	-
1.2720	93000	0.0856	-
1.2856	94000	0.0866	-
1.2993	95000	0.0843	-
1.3130	96000	0.0848	-
1.3267	97000	0.0872	-
1.3404	98000	0.0853	-
1.3540	99000	0.0898	-
1.3677	100000	0.0831	-
1.3814	101000	0.0819	-
1.3951	102000	0.0842	-
1.4087	103000	0.083	-
1.4224	104000	0.0824	-
1.4361	105000	0.0802	-
1.4498	106000	0.0834	-
1.4634	107000	0.0833	-
1.4771	108000	0.0815	-
1.4908	109000	0.079	-
1.4976	109500	-	0.0820
1.5045	110000	0.0809	-
1.5182	111000	0.0784	-
1.5318	112000	0.0767	-
1.5455	113000	0.0782	-
1.5592	114000	0.0799	-
1.5729	115000	0.0787	-
1.5865	116000	0.0798	-
1.6002	117000	0.0821	-
1.6139	118000	0.0771	-
1.6276	119000	0.0758	-
1.6413	120000	0.0789	-
1.6549	121000	0.0777	-
1.6686	122000	0.0755	-
1.6823	123000	0.0774	-
1.6960	124000	0.0748	-
1.7096	125000	0.077	-
1.7233	126000	0.0755	-
1.7370	127000	0.0749	-
1.7507	128000	0.0718	-
1.7643	129000	0.0753	-
1.7780	130000	0.0728	-
1.7917	131000	0.0704	-
1.8054	132000	0.0719	-
1.8191	133000	0.0711	-
1.8327	134000	0.0713	-
1.8464	135000	0.0695	-
1.8601	136000	0.0716	-
1.8738	137000	0.0691	-
1.8874	138000	0.0692	-
1.9011	139000	0.0744	-
1.9148	140000	0.0726	-
1.9285	141000	0.0682	-
1.9421	142000	0.0695	-
1.9558	143000	0.0723	-
1.9695	144000	0.0711	-
1.9832	145000	0.0692	-
1.9969	146000	0.0694	0.0704
2.0105	147000	0.0572	-
2.0242	148000	0.0545	-
2.0379	149000	0.0549	-
2.0516	150000	0.0552	-
2.0652	151000	0.0551	-
2.0789	152000	0.0559	-
2.0926	153000	0.0582	-
2.1063	154000	0.0587	-
2.1199	155000	0.0529	-
2.1336	156000	0.059	-
2.1473	157000	0.0534	-
2.1610	158000	0.0547	-
2.1747	159000	0.0543	-
2.1883	160000	0.0558	-
2.2020	161000	0.0548	-
2.2157	162000	0.0534	-
2.2294	163000	0.0548	-
2.2430	164000	0.0546	-
2.2567	165000	0.053	-
2.2704	166000	0.0557	-
2.2841	167000	0.0541	-
2.2978	168000	0.0527	-
2.3114	169000	0.0542	-
2.3251	170000	0.0529	-
2.3388	171000	0.0554	-
2.3525	172000	0.054	-
2.3661	173000	0.0506	-
2.3798	174000	0.054	-
2.3935	175000	0.0525	-
2.4072	176000	0.0542	-
2.4208	177000	0.0546	-
2.4345	178000	0.0516	-
2.4482	179000	0.053	-
2.4619	180000	0.0542	-
2.4756	181000	0.0538	-
2.4892	182000	0.0536	-
2.4961	182500	-	0.0655
2.5029	183000	0.0489	-
2.5166	184000	0.0543	-
2.5303	185000	0.0509	-
2.5439	186000	0.0525	-
2.5576	187000	0.0525	-
2.5713	188000	0.0535	-
2.5850	189000	0.052	-
2.5986	190000	0.0524	-
2.6123	191000	0.0537	-
2.6260	192000	0.055	-
2.6397	193000	0.0501	-
2.6534	194000	0.0507	-
2.6670	195000	0.0525	-
2.6807	196000	0.0525	-
2.6944	197000	0.0524	-
2.7081	198000	0.0516	-
2.7217	199000	0.051	-
2.7354	200000	0.0533	-
2.7491	201000	0.0515	-
2.7628	202000	0.0505	-
2.7764	203000	0.0492	-
2.7901	204000	0.0536	-
2.8038	205000	0.0488	-
2.8175	206000	0.0473	-
2.8312	207000	0.0536	-
2.8448	208000	0.0499	-
2.8585	209000	0.0507	-
2.8722	210000	0.0511	-
2.8859	211000	0.0494	-
2.8995	212000	0.0508	-
2.9132	213000	0.0494	-
2.9269	214000	0.048	-
2.9406	215000	0.0494	-
2.9543	216000	0.0511	-
2.9679	217000	0.0498	-
2.9816	218000	0.0487	-
2.9953	219000	0.0486	0.0622
3.0090	220000	0.0403	-
3.0226	221000	0.0353	-
3.0363	222000	0.037	-
3.0500	223000	0.0361	-
3.0637	224000	0.0386	-
3.0773	225000	0.0366	-
3.0910	226000	0.0383	-
3.1047	227000	0.0374	-
3.1184	228000	0.0389	-
3.1321	229000	0.0365	-
3.1457	230000	0.0385	-
3.1594	231000	0.037	-
3.1731	232000	0.0382	-
3.1868	233000	0.0391	-
3.2004	234000	0.0361	-
3.2141	235000	0.0369	-
3.2278	236000	0.0387	-
3.2415	237000	0.0402	-
3.2551	238000	0.0394	-
3.2688	239000	0.0377	-
3.2825	240000	0.0361	-
3.2962	241000	0.0375	-
3.3099	242000	0.0385	-
3.3235	243000	0.0384	-
3.3372	244000	0.0389	-
3.3509	245000	0.0376	-
3.3646	246000	0.0367	-
3.3782	247000	0.036	-
3.3919	248000	0.0409	-
3.4056	249000	0.0358	-
3.4193	250000	0.0384	-
3.4329	251000	0.0361	-
3.4466	252000	0.0371	-
3.4603	253000	0.0397	-
3.4740	254000	0.0393	-
3.4877	255000	0.0379	-
3.4945	255500	-	0.0609
3.5013	256000	0.0375	-
3.5150	257000	0.0394	-
3.5287	258000	0.0392	-
3.5424	259000	0.0397	-
3.5560	260000	0.036	-
3.5697	261000	0.0368	-
3.5834	262000	0.0388	-
3.5971	263000	0.0395	-
3.6108	264000	0.0392	-
3.6244	265000	0.038	-
3.6381	266000	0.0359	-
3.6518	267000	0.0382	-
3.6655	268000	0.0371	-
3.6791	269000	0.0383	-
3.6928	270000	0.0398	-
3.7065	271000	0.0372	-
3.7202	272000	0.0382	-
3.7338	273000	0.0382	-
3.7475	274000	0.0346	-
3.7612	275000	0.0396	-
3.7749	276000	0.0366	-
3.7886	277000	0.0398	-
3.8022	278000	0.037	-
3.8159	279000	0.0376	-
3.8296	280000	0.0379	-
3.8433	281000	0.0378	-
3.8569	282000	0.0372	-
3.8706	283000	0.0394	-
3.8843	284000	0.0385	-
3.8980	285000	0.0366	-
3.9116	286000	0.0368	-
3.9253	287000	0.0365	-
3.9390	288000	0.0378	-
3.9527	289000	0.0365	-
3.9664	290000	0.0384	-
3.9800	291000	0.0376	-
3.9937	292000	0.0376	0.0578
4.0074	293000	0.0314	-
4.0211	294000	0.0274	-
4.0347	295000	0.0281	-
4.0484	296000	0.0262	-
4.0621	297000	0.027	-
4.0758	298000	0.0275	-
4.0894	299000	0.0273	-
4.1031	300000	0.0265	-
4.1168	301000	0.0293	-
4.1305	302000	0.0291	-
4.1442	303000	0.0274	-
4.1578	304000	0.0268	-
4.1715	305000	0.0267	-
4.1852	306000	0.0264	-
4.1989	307000	0.0288	-
4.2125	308000	0.0284	-
4.2262	309000	0.0279	-
4.2399	310000	0.0283	-
4.2536	311000	0.0283	-
4.2673	312000	0.0308	-
4.2809	313000	0.0284	-
4.2946	314000	0.0266	-
4.3083	315000	0.0268	-
4.3220	316000	0.0272	-
4.3356	317000	0.0293	-
4.3493	318000	0.0258	-
4.3630	319000	0.0302	-
4.3767	320000	0.0272	-
4.3903	321000	0.0287	-
4.4040	322000	0.0283	-
4.4177	323000	0.0279	-
4.4314	324000	0.0284	-
4.4451	325000	0.0272	-
4.4587	326000	0.0285	-
4.4724	327000	0.028	-
4.4861	328000	0.0265	-
4.4929	328500	-	0.0567

Framework Versions

Python: 3.9.16
Sentence Transformers: 3.1.1
Transformers: 4.45.2
PyTorch: 2.4.1+cu121
Accelerate: 1.0.0
Datasets: 3.0.1
Tokenizers: 0.20.0

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",
}

TripletLoss

bibtex

@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}
}

	anchor	positive	negative
type	string	string	string
details	min: 6 tokens mean: 12.84 tokens max: 23 tokens	min: 3 tokens mean: 15.45 tokens max: 187 tokens	min: 3 tokens mean: 14.75 tokens max: 91 tokens

pankajrajdeo/UMLS-ED-Bioformer-8L-V-1

SentenceTransformer

Model Details

Model Description

Model Sources

Full Model Architecture

Usage

Direct Usage (Sentence Transformers)

Training Details

Training Dataset

Unnamed Dataset

Evaluation Dataset

Unnamed Dataset

Training Hyperparameters

Non-Default Hyperparameters

All Hyperparameters

Training Logs

Framework Versions

Citation

BibTeX

Sentence Transformers

TripletLoss

pankajrajdeo/
UMLS-ED-Bioformer-8L-V-1