suhwan3/mpnet_step1

SentenceTransformer based on sentence-transformers/all-mpnet-base-v2

This is a sentence-transformers model finetuned from sentence-transformers/all-mpnet-base-v2 on the stage1-triplet-dataset dataset. 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: sentence-transformers/all-mpnet-base-v2
• Maximum Sequence Length: 384 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • stage1-triplet-dataset

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': 384, 'do_lower_case': False}) with Transformer model: MPNetModel 
  (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})
  (2): Normalize()
)

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 = [
    'The Sprott Gold Miners ETF (SGDM) seeks to track the performance of the Solactive Gold Miners Custom Factors Total Return Index. This index focuses on gold mining companies based in the U.S. and Canada whose shares trade on the Toronto Stock Exchange, New York Stock Exchange, or NASDAQ. The index employs a weighting methodology that begins with market capitalization and then adjusts based on three fundamental factors: higher revenue growth, lower debt-to-equity, and higher free cash flow yield. The fund is non-diversified and normally invests at least 90% of its net assets in securities included in this index.',
    'The KraneShares Global Carbon Offset Strategy ETF (KSET) was the first US-listed ETF providing exposure to the global voluntary carbon market. It achieved this by investing primarily in liquid carbon offset credit futures, including CME-traded Global Emissions Offsets (GEOs) and Nature-Based Global Emission Offsets (N-GEOs), which are designed to help businesses meet greenhouse gas reduction goals. Tracking an index that weighted eligible futures based on liquidity, the fund sought exposure to the same carbon offset credit futures, typically those maturing within two years. The ETF was considered non-diversified and utilized a Cayman Island subsidiary. However, the fund was delisted, with its last day of trading on an exchange being March 14, 2024.',
    "The VanEck Biotech ETF (BBH) seeks to replicate the performance of the MVIS® US Listed Biotech 25 Index, which provides exposure to approximately 25 of the largest or leading U.S.-listed companies in the biotechnology industry. The fund normally invests at least 80% of its assets in securities comprising this market-cap-weighted index. The underlying index includes common stocks and depositary receipts of firms involved in the research, development, production, marketing, and sale of drugs based on genetic analysis and diagnostic equipment. While focusing on U.S.-listed companies, it may include foreign firms listed domestically, and medium-capitalization companies can be included. Reflecting the index's concentration, the fund is non-diversified and may have a top-heavy portfolio. The index is reviewed semi-annually.",
]
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

stage1-triplet-dataset

• Dataset: stage1-triplet-dataset at a0fb998
• Size: 23,175 training samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 80 tokens mean: 148.35 tokens max: 211 tokens	min: 80 tokens mean: 153.81 tokens max: 238 tokens	min: 82 tokens mean: 150.74 tokens max: 208 tokens

• Samples:

  anchor	positive	negative
  The Invesco Financial Preferred ETF (PGF) seeks to track the ICE Exchange-Listed Fixed Rate Financial Preferred Securities Index, primarily by investing at least 90% of its total assets in the securities comprising the index. The underlying index is market capitalization weighted and designed to track the performance of exchange-listed, fixed rate, U.S. dollar denominated preferred securities, including functionally equivalent instruments, issued by U.S. financial companies. PGF provides a concentrated portfolio exclusively focused on financial-sector preferred securities and is considered non-diversified, holding both investment- and non-investment-grade securities within this focus.	The FlexShares ESG & Climate Investment Grade Corporate Core Index Fund (FEIG) is a passively managed ETF designed to provide broad-market, core exposure to USD-denominated investment-grade corporate bonds. It seeks to track the performance of the Northern Trust ESG & Climate Investment Grade U.S. Corporate Core IndexSM, which selects bonds from a universe of USD-denominated, investment-grade corporate debt with maturities of at least one year. The index employs an optimization process to increase the aggregate ESG score and reduce aggregate climate-related risk among constituent companies, involving ranking firms on material ESG metrics, governance, and carbon risks, while excluding controversial companies and international initiative violators. Weights are also optimized to minimize systematic risk, and the index is rebalanced monthly. Under normal circumstances, the fund invests at least 80% of its assets in the index's securities.	The Pacer Nasdaq-100 Top 50 Cash Cows Growth Leaders ETF (QQQG) seeks to track the Pacer Nasdaq 100 Top 50 Cash Cows Growth Leaders Index, which draws its universe from the Nasdaq-100 Index. Following a rules-based strategy, the fund screens these companies based on average projected free cash flows and earnings over the next two fiscal years, excluding financials, real estate, and those with negative projections. It then ranks identified stocks by their trailing twelve-month free cash flow margins and selects the top 50 names, weighted by price momentum. The portfolio is reconstituted and rebalanced quarterly. Aiming to identify quality growth leaders with strong cash flow generation, the fund seeks to invest at least 80% of assets in growth securities and is non-diversified.
  The Invesco Financial Preferred ETF (PGF) seeks to track the ICE Exchange-Listed Fixed Rate Financial Preferred Securities Index, primarily by investing at least 90% of its total assets in the securities comprising the index. The underlying index is market capitalization weighted and designed to track the performance of exchange-listed, fixed rate, U.S. dollar denominated preferred securities, including functionally equivalent instruments, issued by U.S. financial companies. PGF provides a concentrated portfolio exclusively focused on financial-sector preferred securities and is considered non-diversified, holding both investment- and non-investment-grade securities within this focus.	The FlexShares ESG & Climate Investment Grade Corporate Core Index Fund (FEIG) is a passively managed ETF designed to provide broad-market, core exposure to USD-denominated investment-grade corporate bonds. It seeks to track the performance of the Northern Trust ESG & Climate Investment Grade U.S. Corporate Core IndexSM, which selects bonds from a universe of USD-denominated, investment-grade corporate debt with maturities of at least one year. The index employs an optimization process to increase the aggregate ESG score and reduce aggregate climate-related risk among constituent companies, involving ranking firms on material ESG metrics, governance, and carbon risks, while excluding controversial companies and international initiative violators. Weights are also optimized to minimize systematic risk, and the index is rebalanced monthly. Under normal circumstances, the fund invests at least 80% of its assets in the index's securities.	The Nuveen Global Net Zero Transition ETF (NTZG) was an actively managed fund that sought capital appreciation by investing in global equity securities. The fund focused on companies positioned to contribute to the transition to a net zero carbon economy through their current or planned efforts to reduce global greenhouse gas emissions. Utilizing bottom-up, fundamental analysis, NTZG invested in a range of companies, including climate leaders, firms with disruptive climate mitigation technologies, and high carbon emitters working towards real-world emissions decline. The fund aimed to align with the Paris Climate Agreement by seeking to lower portfolio carbon intensity annually towards a 2050 net zero goal and engaging with portfolio companies, while excluding companies involved in weapons and firearms and investing globally across market capitalizations with allocations to non-US and emerging markets. **Please note: The security has been delisted, and the last day of trading on an exc...
  The Invesco Financial Preferred ETF (PGF) seeks to track the ICE Exchange-Listed Fixed Rate Financial Preferred Securities Index, primarily by investing at least 90% of its total assets in the securities comprising the index. The underlying index is market capitalization weighted and designed to track the performance of exchange-listed, fixed rate, U.S. dollar denominated preferred securities, including functionally equivalent instruments, issued by U.S. financial companies. PGF provides a concentrated portfolio exclusively focused on financial-sector preferred securities and is considered non-diversified, holding both investment- and non-investment-grade securities within this focus.	The FlexShares ESG & Climate Investment Grade Corporate Core Index Fund (FEIG) is a passively managed ETF designed to provide broad-market, core exposure to USD-denominated investment-grade corporate bonds. It seeks to track the performance of the Northern Trust ESG & Climate Investment Grade U.S. Corporate Core IndexSM, which selects bonds from a universe of USD-denominated, investment-grade corporate debt with maturities of at least one year. The index employs an optimization process to increase the aggregate ESG score and reduce aggregate climate-related risk among constituent companies, involving ranking firms on material ESG metrics, governance, and carbon risks, while excluding controversial companies and international initiative violators. Weights are also optimized to minimize systematic risk, and the index is rebalanced monthly. Under normal circumstances, the fund invests at least 80% of its assets in the index's securities.	The First Trust Expanded Technology ETF (XPND) is an actively managed fund seeking long-term capital appreciation by investing primarily in US stocks identified as "Expanded Technology Companies." Defined as companies whose operations are principally derived from or dependent upon technology, these include traditional information technology firms as well as tech-dependent companies in other sectors, such as communication services and consumer discretionary (like internet and direct marketing retail). The fund invests at least 80% of its net assets in common stocks of these companies. While concentrated in the information technology sector and considered non-diversified, XPND aims for expanded exposure through a portfolio of around 50 companies selected using a quantitative model based on factors like return on equity, momentum, and free cash flow growth. Portfolio weights are generally market-cap-based within set ranges, and the fund is reconstituted and rebalanced quarterly.

• Loss: TripletLoss with these parameters:

  {
      "distance_metric": "TripletDistanceMetric.COSINE",
      "triplet_margin": 0.05
  }
  

Evaluation Dataset

stage1-triplet-dataset

• Dataset: stage1-triplet-dataset at a0fb998
• Size: 3,010 evaluation samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 84 tokens mean: 152.57 tokens max: 214 tokens	min: 70 tokens mean: 154.43 tokens max: 224 tokens	min: 70 tokens mean: 150.04 tokens max: 204 tokens

• Samples:

  anchor	positive	negative
  The Global X S&P 500 Risk Managed Income ETF seeks to track the Cboe S&P 500 Risk Managed Income Index by investing at least 80% of its assets in index securities. The index's strategy involves holding the underlying stocks of the S&P 500 Index while applying an options collar, specifically selling at-the-money covered call options and buying monthly 5% out-of-the-money put options corresponding to the portfolio's value. This approach aims to generate income, ideally resulting in a net credit from the options premiums, and provide risk management, though selling at-the-money calls inherently caps the fund's potential for upside participation.	The U.S. Global Technology and Aerospace & Defense ETF is an actively managed ETF seeking capital appreciation by investing in equity securities of companies expected to benefit from national defense efforts. These efforts include technological innovations and the development of products and services related to aerospace, physical, and cybersecurity defense, often in preparation for or in response to domestic, regional, or global conflicts. The fund is non-diversified.	The BlackRock Future Climate and Sustainable Economy ETF (BECO) is an actively managed equity fund focused on the transition to a lower carbon economy and future climate themes. It seeks a relatively concentrated, non-diversified portfolio of globally-listed companies of any market capitalization, investing across multiple subthemes such as sustainable energy, resource efficiency, future transport, sustainable nutrition, and biodiversity. The fund utilizes proprietary environmental criteria, including carbon metrics, and aims to align with the Paris Climate Agreement goals for net-zero emissions by 2050, while excluding certain high-emission industries and companies violating the UN Global Compact. It also attempts to achieve a better aggregate environmental and ESG score than its benchmark, the MSCI ACWI Multiple Industries Select Index. Note that BECO is being delisted, with its last day of trading on an exchange scheduled for August 12, 2024.
  The Global X S&P 500 Risk Managed Income ETF seeks to track the Cboe S&P 500 Risk Managed Income Index by investing at least 80% of its assets in index securities. The index's strategy involves holding the underlying stocks of the S&P 500 Index while applying an options collar, specifically selling at-the-money covered call options and buying monthly 5% out-of-the-money put options corresponding to the portfolio's value. This approach aims to generate income, ideally resulting in a net credit from the options premiums, and provide risk management, though selling at-the-money calls inherently caps the fund's potential for upside participation.	The U.S. Global Technology and Aerospace & Defense ETF is an actively managed ETF seeking capital appreciation by investing in equity securities of companies expected to benefit from national defense efforts. These efforts include technological innovations and the development of products and services related to aerospace, physical, and cybersecurity defense, often in preparation for or in response to domestic, regional, or global conflicts. The fund is non-diversified.	The iShares Energy Storage & Materials ETF (IBAT) seeks to track the STOXX Global Energy Storage and Materials Index, which measures the performance of equity securities of global companies involved in energy storage solutions, including hydrogen, fuel cells, and batteries, aiming to support the transition to a low carbon economy. Determined by STOXX Ltd., the index selects companies based on their exposure to the theme through revenue analysis and patent assessment, while also applying exclusionary ESG screens. The index is price-weighted, based on market capitalization with capping rules. The fund generally invests at least 90% of its assets in the component securities of its underlying index or substantially identical investments and is considered non-diversified.
  The Global X S&P 500 Risk Managed Income ETF seeks to track the Cboe S&P 500 Risk Managed Income Index by investing at least 80% of its assets in index securities. The index's strategy involves holding the underlying stocks of the S&P 500 Index while applying an options collar, specifically selling at-the-money covered call options and buying monthly 5% out-of-the-money put options corresponding to the portfolio's value. This approach aims to generate income, ideally resulting in a net credit from the options premiums, and provide risk management, though selling at-the-money calls inherently caps the fund's potential for upside participation.	The U.S. Global Technology and Aerospace & Defense ETF is an actively managed ETF seeking capital appreciation by investing in equity securities of companies expected to benefit from national defense efforts. These efforts include technological innovations and the development of products and services related to aerospace, physical, and cybersecurity defense, often in preparation for or in response to domestic, regional, or global conflicts. The fund is non-diversified.	The Sprott Gold Miners ETF (SGDM) seeks to track the performance of the Solactive Gold Miners Custom Factors Total Return Index. This index focuses on gold mining companies based in the U.S. and Canada whose shares trade on the Toronto Stock Exchange, New York Stock Exchange, or NASDAQ. The index employs a weighting methodology that begins with market capitalization and then adjusts based on three fundamental factors: higher revenue growth, lower debt-to-equity, and higher free cash flow yield. The fund is non-diversified and normally invests at least 90% of its net assets in securities included in this index.

• Loss: TripletLoss with these parameters:

  {
      "distance_metric": "TripletDistanceMetric.COSINE",
      "triplet_margin": 0.05
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• learning_rate: 3e-05
• num_train_epochs: 1
• warmup_ratio: 0.1
• bf16: True
• dataloader_drop_last: True
• load_best_model_at_end: 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: 3e-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: True
• 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: True
• 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}
• tp_size: 0
• 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
• 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

Click to expand

Epoch	Step	Training Loss	Validation Loss
0.0069	10	0.0448	-
0.0138	20	0.0354	-
0.0207	30	0.0293	-
0.0276	40	0.0381	-
0.0345	50	0.0228	-
0.0414	60	0.0238	-
0.0483	70	0.0229	-
0.0552	80	0.0148	-
0.0622	90	0.0175	-
0.0691	100	0.0161	-
0.0760	110	0.0124	-
0.0829	120	0.0111	-
0.0898	130	0.0165	-
0.0967	140	0.0162	-
0.1036	150	0.0141	-
0.1105	160	0.0116	-
0.1174	170	0.01	-
0.1243	180	0.0134	-
0.1312	190	0.0117	-
0.1381	200	0.0127	0.0131
0.1450	210	0.0083	-
0.1519	220	0.0116	-
0.1588	230	0.0099	-
0.1657	240	0.0086	-
0.1727	250	0.0099	-
0.1796	260	0.0047	-
0.1865	270	0.0052	-
0.1934	280	0.0086	-
0.2003	290	0.0084	-
0.2072	300	0.0068	-
0.2141	310	0.005	-
0.2210	320	0.0077	-
0.2279	330	0.0044	-
0.2348	340	0.0039	-
0.2417	350	0.0058	-
0.2486	360	0.0045	-
0.2555	370	0.0045	-
0.2624	380	0.0064	-
0.2693	390	0.0037	-
0.2762	400	0.0083	0.013
0.2831	410	0.0057	-
0.2901	420	0.0043	-
0.2970	430	0.0028	-
0.3039	440	0.0036	-
0.3108	450	0.0031	-
0.3177	460	0.0072	-
0.3246	470	0.0025	-
0.3315	480	0.0041	-
0.3384	490	0.0049	-
0.3453	500	0.0035	-
0.3522	510	0.0023	-
0.3591	520	0.0043	-
0.3660	530	0.0032	-
0.3729	540	0.0031	-
0.3798	550	0.0039	-
0.3867	560	0.0042	-
0.3936	570	0.0055	-
0.4006	580	0.0041	-
0.4075	590	0.0026	-
0.4144	600	0.002	0.0133
0.4213	610	0.0027	-
0.4282	620	0.0032	-
0.4351	630	0.0025	-
0.4420	640	0.0042	-
0.4489	650	0.0046	-
0.4558	660	0.0011	-
0.4627	670	0.0004	-
0.4696	680	0.0019	-
0.4765	690	0.0034	-
0.4834	700	0.0032	-
0.4903	710	0.0029	-
0.4972	720	0.0038	-
0.5041	730	0.0021	-
0.5110	740	0.0008	-
0.5180	750	0.0015	-
0.5249	760	0.0018	-
0.5318	770	0.0022	-
0.5387	780	0.0006	-
0.5456	790	0.0022	-
0.5525	800	0.0006	0.0160
0.5594	810	0.0021	-
0.5663	820	0.0013	-
0.5732	830	0.0019	-
0.5801	840	0.0017	-
0.5870	850	0.0008	-
0.5939	860	0.0012	-
0.6008	870	0.0003	-
0.6077	880	0.0009	-
0.6146	890	0.001	-
0.6215	900	0.0011	-
0.6285	910	0.0019	-
0.6354	920	0.0009	-
0.6423	930	0.0003	-
0.6492	940	0.0001	-
0.6561	950	0.0019	-
0.6630	960	0.0006	-
0.6699	970	0.0003	-
0.6768	980	0.0005	-
0.6837	990	0.0025	-
0.6906	1000	0.001	0.0154
0.6975	1010	0.0009	-
0.7044	1020	0.0004	-
0.7113	1030	0.0008	-
0.7182	1040	0.001	-
0.7251	1050	0.0018	-
0.7320	1060	0.002	-
0.7390	1070	0.0	-
0.7459	1080	0.0	-
0.7528	1090	0.0003	-
0.7597	1100	0.0002	-
0.7666	1110	0.0004	-
0.7735	1120	0.0004	-
0.7804	1130	0.0001	-
0.7873	1140	0.0002	-
0.7942	1150	0.001	-
0.8011	1160	0.0003	-
0.8080	1170	0.0003	-
0.8149	1180	0.0002	-
0.8218	1190	0.0002	-
0.8287	1200	0.0	0.0179
0.8356	1210	0.0006	-
0.8425	1220	0.0005	-
0.8494	1230	0.0015	-
0.8564	1240	0.0009	-
0.8633	1250	0.0007	-
0.8702	1260	0.0003	-
0.8771	1270	0.0003	-
0.8840	1280	0.0	-
0.8909	1290	0.0	-
0.8978	1300	0.0009	-
0.9047	1310	0.0011	-
0.9116	1320	0.0003	-
0.9185	1330	0.0	-
0.9254	1340	0.0002	-
0.9323	1350	0.0004	-
0.9392	1360	0.0004	-
0.9461	1370	0.0007	-
0.9530	1380	0.0006	-
0.9599	1390	0.0006	-
0.9669	1400	0.0005	0.0167

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.10.12
• Sentence Transformers: 4.1.0
• Transformers: 4.51.3
• PyTorch: 2.1.0+cu118
• Accelerate: 1.6.0
• Datasets: 3.5.0
• Tokenizers: 0.21.1

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",
}

TripletLoss

@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}
}