README.md

metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:3772
  - loss:MultipleNegativesRankingLoss
  - loss:CosineSimilarityLoss
base_model: jinaai/jina-embedding-b-en-v1
widget:
  - source_sentence: do I possess any funds that are not performing
    sentences:
      - How did my portfolio perform this month?
      - Show me my best performing holdings
      - do I hold any funds that haven't been performing well
  - source_sentence: did I earn any money over the past 18 days
    sentences:
      - What are the returns of my portfolio in the past 6 months?
      - did I gain any profits over the past 18 days
      - How are my investments performing ?
  - source_sentence: What stocks should I expect to rise 100% in the next month?
    sentences:
      - performance of my portfolio over the last 3 years
      - Tell me which of my stocks will go up 100% in next month
      - How much X do I have in my portfolio
  - source_sentence: Mere portfolio kaise chal raha hai?
    sentences:
      - What are my top performing funds?
      - Mere funds kese chal rahe hai?
      - How can I improve my returns ?
  - source_sentence: how many folios do I own
    sentences:
      - Which of my investments have the highest volatility?
      - show my portfolio scheme wise
      - Show all my folios
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - cosine_accuracy@1
  - cosine_accuracy@3
  - cosine_accuracy@5
  - cosine_accuracy@10
  - cosine_precision@1
  - cosine_precision@3
  - cosine_precision@5
  - cosine_precision@10
  - cosine_recall@1
  - cosine_recall@3
  - cosine_recall@5
  - cosine_recall@10
  - cosine_ndcg@10
  - cosine_mrr@10
  - cosine_map@100
model-index:
  - name: SentenceTransformer based on jinaai/jina-embedding-b-en-v1
    results:
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: test eval
          type: test-eval
        metrics:
          - type: cosine_accuracy@1
            value: 0.8756613756613757
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.9894179894179894
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.9973544973544973
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 1
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.8756613756613757
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.3298059964726632
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.19947089947089944
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.09999999999999999
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.8756613756613757
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.9894179894179894
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.9973544973544973
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 1
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.9482145369797552
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.9304232804232803
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.9304232804232804
            name: Cosine Map@100

SentenceTransformer based on jinaai/jina-embedding-b-en-v1

This is a sentence-transformers model finetuned from jinaai/jina-embedding-b-en-v1. 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: jinaai/jina-embedding-b-en-v1
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 768 dimensions
• 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: T5EncoderModel 
  (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})
)

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 = [
    'how many folios do I own',
    'Show all my folios',
    'Which of my investments have the highest volatility?',
]
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]

Evaluation

Metrics

Information Retrieval

• Dataset: test-eval
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.8757
cosine_accuracy@3	0.9894
cosine_accuracy@5	0.9974
cosine_accuracy@10	1.0
cosine_precision@1	0.8757
cosine_precision@3	0.3298
cosine_precision@5	0.1995
cosine_precision@10	0.1
cosine_recall@1	0.8757
cosine_recall@3	0.9894
cosine_recall@5	0.9974
cosine_recall@10	1.0
cosine_ndcg@10	0.9482
cosine_mrr@10	0.9304
cosine_map@100	0.9304

Training Details

Training Datasets

Unnamed Dataset

• Size: 1,886 training samples
• Columns: sentence_0, sentence_1, and label
• Approximate statistics based on the first 1000 samples:

  	sentence_0	sentence_1	label
  type	string	string	float
  details	min: 4 tokens mean: 11.34 tokens max: 26 tokens	min: 4 tokens mean: 10.0 tokens max: 33 tokens	min: 1.0 mean: 1.0 max: 1.0

• Samples:

  sentence_0	sentence_1	label
  how can I reduce the risk in my portfolio?	are there any ways to lower risk in my portfolio ?	1.0
  What stocks do I own that have the highest volatility?	Which of my stocks are most volatile?	1.0
  What are the trends in investments for people like me?	What are others like me investing in?	1.0

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Unnamed Dataset

• Size: 1,886 training samples
• Columns: sentence_0, sentence_1, and label
• Approximate statistics based on the first 1000 samples:

  	sentence_0	sentence_1	label
  type	string	string	float
  details	min: 4 tokens mean: 11.11 tokens max: 26 tokens	min: 4 tokens mean: 9.82 tokens max: 33 tokens	min: 1.0 mean: 1.0 max: 1.0

• Samples:

  sentence_0	sentence_1	label
  What parts of my portfolio are successful?	What's going well in my portfolio	1.0
  Reveal my investment holdings	Show my holdings	1.0
  What do others in my situation invest in?	What are others like me investing in?	1.0

• Loss: CosineSimilarityLoss with these parameters:

  {
      "loss_fct": "torch.nn.modules.loss.MSELoss"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 32
• per_device_eval_batch_size: 32
• num_train_epochs: 10
• multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 32
• per_device_eval_batch_size: 32
• 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
• num_train_epochs: 10
• 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: 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
• 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
• average_tokens_across_devices: False
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: round_robin

Training Logs

Epoch	Step	Training Loss	test-eval_cosine_ndcg@10
1.0	118	-	0.9051
2.0	236	-	0.9165
3.0	354	-	0.9293
4.0	472	-	0.9329
4.2373	500	0.206	0.9338
5.0	590	-	0.9338
6.0	708	-	0.9363
7.0	826	-	0.9403
8.0	944	-	0.9466
8.4746	1000	0.1406	0.9482

Framework Versions

• Python: 3.12.5
• Sentence Transformers: 3.4.1
• Transformers: 4.49.0
• PyTorch: 2.6.0
• Accelerate: 1.5.2
• Datasets: 3.4.1
• 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",
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply},
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}