e5-fine-tuned / README.md
DariaaaS's picture
Add new SentenceTransformer model.
538f62d verified
metadata
base_model: intfloat/multilingual-e5-base
datasets: []
language: []
library_name: sentence-transformers
pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:100
  - loss:TripletLoss
widget:
  - source_sentence: What is the average household income in the city known as "Danzig"?
    sentences:
      - the most bad aliases refer to MAX(COUNT(bad_alias));
      - Greeneville is the city;
      - >-
        average household income refers to avg_income_per_household; city known
        as "Danzig" refers to bad_alias = 'Danzig';
  - source_sentence: What is the average household income in the city known as "Danzig"?
    sentences:
      - '"Berlin, NH" is the CBSA_name'
      - '"Puerto Rico" refers to state = ''PR'''
      - >-
        average household income refers to avg_income_per_household; city known
        as "Danzig" refers to bad_alias = 'Danzig';
  - source_sentence: What is the country and state of the city named Dalton?
    sentences:
      - median age over 40 refers to median_age > 40
      - >-
        DIVIDE(SUBTRACT(SUM(population_2020)), SUM(population_2010)),
        SUM(population_2010) as percentage where county = 'ARROYO';
      - Dalton is the city;
  - source_sentence: What is the country and state of the city named Dalton?
    sentences:
      - >-
        community post office type refers to type = 'Community Post Office';
        elevation above 6000 refers to elevation > 6000;
      - Dalton is the city;
      - '"Berlin, NH" is the CBSA_name'
  - source_sentence: >-
      List 10 cities with a median age over 40. Include their zip codes and area
      codes.
    sentences:
      - '"URB San Joaquin" is the bad_alias'
      - >-
        in California refers to name = 'California' and state = 'CA'; 'Community
        Post Office' is the Type
      - median age over 40 refers to median_age > 40

SentenceTransformer based on intfloat/multilingual-e5-base

This is a sentence-transformers model finetuned from intfloat/multilingual-e5-base. 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: intfloat/multilingual-e5-base
  • Maximum Sequence Length: 512 tokens
  • Output Dimensionality: 768 tokens
  • Similarity Function: Cosine Similarity

Model Sources

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: XLMRobertaModel 
  (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("DariaaaS/e5-fine-tuned")
# Run inference
sentences = [
    'List 10 cities with a median age over 40. Include their zip codes and area codes.',
    'median age over 40 refers to median_age > 40',
    "in California refers to name = 'California' and state = 'CA'; 'Community Post Office' is the Type",
]
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

Unnamed Dataset

  • Size: 100 training samples
  • Columns: sentence_0, sentence_1, and sentence_2
  • Approximate statistics based on the first 1000 samples:
    sentence_0 sentence_1 sentence_2
    type string string string
    details
    • min: 11 tokens
    • mean: 19.8 tokens
    • max: 31 tokens
    • min: 8 tokens
    • mean: 21.3 tokens
    • max: 39 tokens
    • min: 8 tokens
    • mean: 27.57 tokens
    • max: 54 tokens
  • Samples:
    sentence_0 sentence_1 sentence_2
    Among all the residential areas in Delaware, how many of them implement daylight saving? "Delaware" is a county; implement daylight savings refers to daylight_saving = 'Yes' DIVIDE(COUNT(zip_code where type = 'Post Office'), COUNT(zip_code)) as percentage where name = 'California';
    What is the country and state of the city named Dalton? Dalton is the city; average male median age refers to Divide (Sum(male_median_age), Count(county)); 'WINDHAM' is the county
    Among the residential areas with the bad alias "Internal Revenue Service", how many of them are in the Eastern time zone? "Internal Revenue Service" is the bad_alias; in Eastern time zone refers to time_zone = 'Eastern' "Berlin, NH" is the CBSA_name
  • Loss: TripletLoss with these parameters:
    {
        "distance_metric": "TripletDistanceMetric.EUCLIDEAN",
        "triplet_margin": 5
    }
    

Training Hyperparameters

Non-Default Hyperparameters

  • per_device_train_batch_size: 10
  • per_device_eval_batch_size: 10
  • num_train_epochs: 5
  • multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand
  • overwrite_output_dir: False
  • do_predict: False
  • eval_strategy: no
  • prediction_loss_only: True
  • per_device_train_batch_size: 10
  • per_device_eval_batch_size: 10
  • 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
  • num_train_epochs: 5
  • max_steps: -1
  • 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
  • batch_sampler: batch_sampler
  • multi_dataset_batch_sampler: round_robin

Framework Versions

  • Python: 3.10.12
  • Sentence Transformers: 3.0.1
  • Transformers: 4.42.4
  • PyTorch: 2.4.0+cu121
  • Accelerate: 0.32.1
  • Datasets: 2.21.0
  • 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",
}

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}
}