SentenceTransformer based on BAAI/bge-small-en-v1.5

This is a sentence-transformers model finetuned from BAAI/bge-small-en-v1.5. 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: BAAI/bge-small-en-v1.5
Maximum Sequence Length: 512 tokens
Output Dimensionality: 384 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': True}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 384, '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:

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("Snivellus789/router-embedding-tuned")
# Run inference
sentences = [
    'In Swift, what function can I use to shorten the sentence "I\'m feeling kind of tired after having worked all day" while maintaining the same meaning and tone? Can you provide an example of the shortened sentence using the function? ',
    'Convert the given XML code to JSON code. <root>\n    <data>\n        <item id="1">\n            <name>Sample data</name>\n            <type>Text</type>\n            <value>123</value>\n        </item>\n    </data>\n</root>',
    'How can I create a C# program that generates a travel itinerary based on user preferences and available destinations? The program should take into account factors such as budget, time of year, and desired activities (such as hiking or sightseeing). Please use the following data format to represent the available destinations:\n```csharp\nList<Destination> destinations = new List<Destination>\n{\n    new Destination\n    {\n        Name = "Paris",\n        Country = "France",\n        Activities = new List<string> {"sightseeing", "shopping", "museums"},\n        Cost = 5000,\n        Season = "spring"\n    },\n    new Destination\n    {\n        Name = "Tokyo",\n        Country = "Japan",\n        Activities = new List<string> {"sightseeing", "food", "temples"},\n        Cost = 8000,\n        Season = "fall"\n    },\n    new Destination\n    {\n        Name = "Sydney",\n        Country = "Australia",\n        Activities = new List<string> {"beaches", "hiking", "wildlife"},\n        Cost = 7000,\n        Season = "summer"\n    },\n    new Destination\n    {\n        Name = "Marrakesh",\n        Country = "Morocco",\n        Activities = new List<string> {"sightseeing", "shopping", "food"},\n        Cost = 4000,\n        Season = "winter"\n    }\n};\npublic class Destination\n{\n    public string Name { get; set; }\n    public string Country { get; set; }\n    public List<string> Activities { get; set; }\n    public int Cost { get; set; }\n    public string Season { get; set; }\n}\n```\nPlease provide step-by-step instructions for using the program and any necessary inputs. ',
]
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: 1,000 training samples
Columns: sentence and label
Approximate statistics based on the first 1000 samples:
sentence label
type string int
details
min: 8 tokens
mean: 95.61 tokens
max: 512 tokens

0: ~50.00%
1: ~50.00%

	sentence	label
type	string	int
details	min: 8 tokens mean: 95.61 tokens max: 512 tokens	0: ~50.00% 1: ~50.00%

Samples:

sentence	label
`请输出所有跟政企市场相关的关键词列表`	`0`
`开发一个定制的JavaScript解决方案，用于有效地平衡和排序一个二叉树。你可以假设输入是一个平衡因子擯至2的大O()为Log(N)的AVL树。专注于实现自我调整二叉搜索树的变换，当面对不平衡操作时，如插入或删除节点。确保你的解决方案为潜在的边缘案例做好准备，并具有健壮的错误处理策略。你的代码应该清晰地记录和优化效率。`	`0`
`在一个尚未被公开的领域中，描述五个最具创新性的产品概念。`	`0`

Loss: BatchAllTripletLoss

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 16
per_device_eval_batch_size: 16
learning_rate: 2e-05
num_train_epochs: 2
warmup_ratio: 0.1
bf16: True
batch_sampler: no_duplicates

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: 16
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 1
eval_accumulation_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: 2
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: 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
1.5873	100	0.0963

Framework Versions

Python: 3.10.12
Sentence Transformers: 3.0.1
Transformers: 4.42.4
PyTorch: 2.3.1+cu121
Accelerate: 0.33.0.dev0
Datasets: 2.20.0
Tokenizers: 0.19.1

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",
}

BatchAllTripletLoss

@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}
}