fm / README.md

Add new SentenceTransformer model.

26d543e verified almost 2 years ago

24.9 kB

base_model: BAAI/bge-base-en-v1.5
datasets: []
language: []
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
pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:26
  - loss:MultipleNegativesRankingLoss
widget:
  - source_sentence: >-
      The Supplier shall deliver the Batteries to the Manufacturer within 5 days
      of receipt of each
    sentences:
      - according to the MOU?
      - What is the Delivery Schedule for the Batteries?
      - single order?
  - source_sentence: >-
      The Employee agrees to abide by the Employer’s rules, regulations,
      guidelines, policies, and
    sentences:
      - When does this Agreement terminate?
      - What rules and policies must the Employee abide by?
      - Which law governs this Agreement, and where would disputes be resolved?
  - source_sentence: >-
      Answer: Deepak Babbar agrees to pay Rs 5,10,000 as a full and final
      settlement to Ayushi
    sentences:
      - What are the Payment Terms for the Batteries?
      - What financial settlement does Deepak Babbar agree to in the MOU?
      - order?
  - source_sentence: >-
      The Supplier agrees to supply 60,000 Batteries over the course of one
      year, as specified in
    sentences:
      - When does the Employee commence employment with the Employer?
      - When does the Company employ the Employee?
      - >-
        How many Batteries are Supplier obligated to supply under this
        Agreement?
  - source_sentence: >-
      The term of this Agreement shall continue until terminated by either party
      in accordance with
    sentences:
      - What is the pricing per Battery under this Agreement?
      - What events constitute Force Majeure under this Agreement?
      - What is the term of the Agreement?
model-index:
  - name: SentenceTransformer based on BAAI/bge-base-en-v1.5
    results:
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 768
          type: dim_768
        metrics:
          - type: cosine_accuracy@1
            value: 0.3333333333333333
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.3333333333333333
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.3333333333333333
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.6666666666666666
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.3333333333333333
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.1111111111111111
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.06666666666666667
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.06666666666666667
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.3333333333333333
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.3333333333333333
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.3333333333333333
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.6666666666666666
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.4336766652213271
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.3703703703703704
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.3862433862433863
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 512
          type: dim_512
        metrics:
          - type: cosine_accuracy@1
            value: 0.3333333333333333
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.3333333333333333
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.3333333333333333
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.6666666666666666
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.3333333333333333
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.1111111111111111
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.06666666666666667
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.06666666666666667
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.3333333333333333
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.3333333333333333
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.3333333333333333
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.6666666666666666
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.4336766652213271
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.3703703703703704
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.38703703703703707
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 256
          type: dim_256
        metrics:
          - type: cosine_accuracy@1
            value: 0.3333333333333333
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.3333333333333333
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.3333333333333333
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.6666666666666666
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.3333333333333333
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.1111111111111111
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.06666666666666667
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.06666666666666667
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.3333333333333333
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.3333333333333333
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.3333333333333333
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.6666666666666666
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.4336766652213271
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.3703703703703704
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.38791423001949316
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 128
          type: dim_128
        metrics:
          - type: cosine_accuracy@1
            value: 0.3333333333333333
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.3333333333333333
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.3333333333333333
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 1
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.3333333333333333
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.1111111111111111
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.06666666666666667
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.10000000000000002
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.3333333333333333
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.3333333333333333
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.3333333333333333
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 1
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.5524123942573345
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.425925925925926
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.425925925925926
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 64
          type: dim_64
        metrics:
          - type: cosine_accuracy@1
            value: 0.3333333333333333
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.6666666666666666
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.6666666666666666
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.6666666666666666
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.3333333333333333
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.2222222222222222
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.13333333333333333
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.06666666666666667
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.3333333333333333
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.6666666666666666
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.6666666666666666
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.6666666666666666
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.5
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.4444444444444444
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.47008547008547
            name: Cosine Map@100

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

This is a sentence-transformers model finetuned from BAAI/bge-base-en-v1.5. 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: BAAI/bge-base-en-v1.5
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 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': 768, '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("vineet10/fm")
# Run inference
sentences = [
    'The term of this Agreement shall continue until terminated by either party in accordance with',
    'What is the term of the Agreement?',
    'What events constitute Force Majeure under this Agreement?',
]
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: dim_768
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.3333
cosine_accuracy@3	0.3333
cosine_accuracy@5	0.3333
cosine_accuracy@10	0.6667
cosine_precision@1	0.3333
cosine_precision@3	0.1111
cosine_precision@5	0.0667
cosine_precision@10	0.0667
cosine_recall@1	0.3333
cosine_recall@3	0.3333
cosine_recall@5	0.3333
cosine_recall@10	0.6667
cosine_ndcg@10	0.4337
cosine_mrr@10	0.3704
cosine_map@100	0.3862

Information Retrieval

Dataset: dim_512
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.3333
cosine_accuracy@3	0.3333
cosine_accuracy@5	0.3333
cosine_accuracy@10	0.6667
cosine_precision@1	0.3333
cosine_precision@3	0.1111
cosine_precision@5	0.0667
cosine_precision@10	0.0667
cosine_recall@1	0.3333
cosine_recall@3	0.3333
cosine_recall@5	0.3333
cosine_recall@10	0.6667
cosine_ndcg@10	0.4337
cosine_mrr@10	0.3704
cosine_map@100	0.387

Information Retrieval

Dataset: dim_256
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.3333
cosine_accuracy@3	0.3333
cosine_accuracy@5	0.3333
cosine_accuracy@10	0.6667
cosine_precision@1	0.3333
cosine_precision@3	0.1111
cosine_precision@5	0.0667
cosine_precision@10	0.0667
cosine_recall@1	0.3333
cosine_recall@3	0.3333
cosine_recall@5	0.3333
cosine_recall@10	0.6667
cosine_ndcg@10	0.4337
cosine_mrr@10	0.3704
cosine_map@100	0.3879

Information Retrieval

Dataset: dim_128
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.3333
cosine_accuracy@3	0.3333
cosine_accuracy@5	0.3333
cosine_accuracy@10	1.0
cosine_precision@1	0.3333
cosine_precision@3	0.1111
cosine_precision@5	0.0667
cosine_precision@10	0.1
cosine_recall@1	0.3333
cosine_recall@3	0.3333
cosine_recall@5	0.3333
cosine_recall@10	1.0
cosine_ndcg@10	0.5524
cosine_mrr@10	0.4259
cosine_map@100	0.4259

Information Retrieval

Dataset: dim_64
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.3333
cosine_accuracy@3	0.6667
cosine_accuracy@5	0.6667
cosine_accuracy@10	0.6667
cosine_precision@1	0.3333
cosine_precision@3	0.2222
cosine_precision@5	0.1333
cosine_precision@10	0.0667
cosine_recall@1	0.3333
cosine_recall@3	0.6667
cosine_recall@5	0.6667
cosine_recall@10	0.6667
cosine_ndcg@10	0.5
cosine_mrr@10	0.4444
cosine_map@100	0.4701

Training Details

Training Dataset

Unnamed Dataset

Size: 26 training samples
Columns: context and question
Approximate statistics based on the first 1000 samples:
context question
type string string
details
min: 2 tokens
mean: 19.19 tokens
max: 28 tokens

min: 4 tokens
mean: 11.27 tokens
max: 18 tokens

	context	question
type	string	string
details	min: 2 tokens mean: 19.19 tokens max: 28 tokens	min: 4 tokens mean: 11.27 tokens max: 18 tokens

Samples:

context	question
`Answer: Deepak Babbar makes the final payment of Rs 2,60,000 at the time of quashing FIR`	`MOU?`
`This Agreement is governed by the laws of Indiana, and any disputes arising out of or in`	`Which law governs this Agreement, and where would disputes be resolved?`
`Answer: After the first motion, both parties must file petitions for quashing FIRs and`	`according to the MOU?`

Loss: MultipleNegativesRankingLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 16
per_device_eval_batch_size: 16
num_train_epochs: 5
warmup_ratio: 0.1
fp16: 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
learning_rate: 5e-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: 5
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: 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	dim_128_cosine_map@100	dim_256_cosine_map@100	dim_512_cosine_map@100	dim_64_cosine_map@100	dim_768_cosine_map@100
0	0	0.4259	0.3879	0.3870	0.4701	0.3862

Framework Versions

Python: 3.10.12
Sentence Transformers: 3.0.1
Transformers: 4.42.4
PyTorch: 2.3.1+cu121
Accelerate: 0.32.1
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",
}

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}
}