langcache-embed-v3 / README.md

radoslavralev

Add new SentenceTransformer model

f23e239 verified 18 days ago

preview code

raw

history blame

21.1 kB

metadata

language:
  - en
license: apache-2.0
tags:
  - biencoder
  - sentence-transformers
  - text-classification
  - sentence-pair-classification
  - semantic-similarity
  - semantic-search
  - retrieval
  - reranking
  - generated_from_trainer
  - dataset_size:13675
  - loss:ArcFaceInBatchLoss
base_model: Alibaba-NLP/gte-modernbert-base
widget:
  - source_sentence: >-
      Bathurst Street has been the heart of the Jewish community of Toronto for
      decades .
    sentences:
      - >-
        Baron portrayed actress Violet Carson who played Ena Sharples in the
        soap .
      - >-
        Bathurst Street has been the heart of the Jewish community of Toronto
        for many decades .
      - >-
        It stretches approximately 20 miles from Manasquan Inlet in Point
        Pleasant Beach in the north to Island Beach State Park in the south .
  - source_sentence: >-
      All tracks produced by Zack Shada , Jeremy Shada , Logan Charles , John
      Spicer and Seth Renken . All tracks are written by Zack Odom and Kenneth
      Mount .
    sentences:
      - >-
        All tracks produced by Zack Shada , Jeremy Shada , Logan Charles , John
        Spicer and Seth Renken . All tracks are written by Zack Odom and Kenneth
        Mount .
      - >-
        All tracks by Zack Shada , Jeremy Shada , John Spicer , Logan Charles
        and Seth Renken are produced by Zack Odom and Kenneth Mount .
      - Jimmy Connors defeated Eddie Dibbs 7 -- 5 , 7 -- 5
  - source_sentence: >-
      Arque Municipality is situated in the eastern part of the province and
      Tacopaya Municipality is located in the west .
    sentences:
      - >-
        Arque Municipality is situated in the eastern part of the province and
        Tacopaya Municipality is located in the west .
      - >-
        Bangkok International Preparatory and Secondary School , or Bangkok Prep
        , is an independent international school located on the National
        Curriculum of England based in Bangkok , Thailand .
      - >-
        The municipality of Tacopaya is situated in the eastern part of the
        province and municipality of Arque located in the west .
  - source_sentence: Browning is identified as married , but no wife or child is captured .
    sentences:
      - >-
        Alexander Alexander is the grandson of the Sarawak - leader Tun Jugah
        Barieng and the son of former politician Tan Sri Datuk Amar Leonard
        Linggi .
      - Browning is identified as married , but no wife or child is recorded .
      - It was formerly known also as ' Crotto ' .
  - source_sentence: >-
      Actor Charlie Chan , who portrayed Warner Oland when `` The Black Camel ``
      was filmed in Hawaii , he met .
    sentences:
      - >-
        Chang met actor Warner Oland , who portrayed Charlie Chan , when `` The
        Black Camel `` was filmed in Hawaii .
      - >-
        As an actor , he joined the Royal Shakespeare Company of Peter Hall ,
        working with Peggy Ashcroft and Dame Edith Evans .
      - >-
        Actor Charlie Chan , who portrayed Warner Oland when `` The Black Camel
        `` was filmed in Hawaii , he met .
datasets:
  - redis/langcache-sentencepairs-v2
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - cosine_accuracy@1
  - cosine_precision@1
  - cosine_recall@1
  - cosine_ndcg@10
  - cosine_mrr@1
  - cosine_map@100
  - cosine_auc_precision_cache_hit_ratio
  - cosine_auc_similarity_distribution
model-index:
  - name: Redis fine-tuned BiEncoder model for semantic caching on LangCache
    results:
      - task:
          type: custom-information-retrieval
          name: Custom Information Retrieval
        dataset:
          name: test
          type: test
        metrics:
          - type: cosine_accuracy@1
            value: 0.5880219631236443
            name: Cosine Accuracy@1
          - type: cosine_precision@1
            value: 0.5880219631236443
            name: Cosine Precision@1
          - type: cosine_recall@1
            value: 0.5706780985738924
            name: Cosine Recall@1
          - type: cosine_ndcg@10
            value: 0.7717640552650085
            name: Cosine Ndcg@10
          - type: cosine_mrr@1
            value: 0.5880219631236443
            name: Cosine Mrr@1
          - type: cosine_map@100
            value: 0.7213999116625115
            name: Cosine Map@100
          - type: cosine_auc_precision_cache_hit_ratio
            value: 0.35292771304732773
            name: Cosine Auc Precision Cache Hit Ratio
          - type: cosine_auc_similarity_distribution
            value: 0.1674589579463346
            name: Cosine Auc Similarity Distribution

Redis fine-tuned BiEncoder model for semantic caching on LangCache

This is a sentence-transformers model finetuned from Alibaba-NLP/gte-modernbert-base on the LangCache Sentence Pairs (all) dataset. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for sentence pair similarity.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: Alibaba-NLP/gte-modernbert-base
Maximum Sequence Length: 100 tokens
Output Dimensionality: 768 dimensions
Similarity Function: Cosine Similarity
Training Dataset:
- LangCache Sentence Pairs (all)
Language: en
License: apache-2.0

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': 100, 'do_lower_case': False, 'architecture': 'ModernBertModel'})
  (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})
  (mlp_hidden): Dense({'in_features': 768, 'out_features': 768, 'bias': True, 'activation_function': 'torch.nn.modules.activation.ReLU'})
  (mlp_out): Dense({'in_features': 768, 'out_features': 768, 'bias': True, 'activation_function': 'torch.nn.modules.linear.Identity'})
)

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("redis/langcache-embed-v3")
# Run inference
sentences = [
    'Actor Charlie Chan , who portrayed Warner Oland when `` The Black Camel `` was filmed in Hawaii , he met .',
    'Actor Charlie Chan , who portrayed Warner Oland when `` The Black Camel `` was filmed in Hawaii , he met .',
    'Chang met actor Warner Oland , who portrayed Charlie Chan , when `` The Black Camel `` was filmed in Hawaii .',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 1.0000, 0.5313],
#         [1.0000, 1.0000, 0.5313],
#         [0.5313, 0.5313, 1.0000]])

Evaluation

Metrics

Custom Information Retrieval

Dataset: test
Evaluated with ir_evaluator.CustomInformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.588
cosine_precision@1	0.588
cosine_recall@1	0.5707
cosine_ndcg@10	0.7718
cosine_mrr@1	0.588
cosine_map@100	0.7214
cosine_auc_precision_cache_hit_ratio	0.3529
cosine_auc_similarity_distribution	0.1675

Training Details

Training Dataset

LangCache Sentence Pairs (all)

Dataset: LangCache Sentence Pairs (all)
Size: 6,786 training samples
Columns: anchor, positive, and negative

Approximate statistics based on the first 1000 samples:

	anchor	positive	negative
type	string	string	string
details	min: 9 tokens mean: 27.96 tokens max: 50 tokens	min: 9 tokens mean: 27.98 tokens max: 51 tokens	min: 9 tokens mean: 27.56 tokens max: 49 tokens

Samples:

anchor	positive	negative
( 1 ) Lakers vs. ( 2 ) San Antonio Spurs : `` Los Angeles Lakers Win 4-0	`( 1 ) Lakers vs. ( 2 ) San Antonio Spurs : Los Angeles Lakers win series 4-0`	`( 1 ) Los Angeles Lakers vs. ( 2 ) San Antonio Spurs : Lakers win series 4-0`
`( 1 ) Lakers vs. ( 2 ) San Antonio Spurs : Los Angeles Lakers win series 4-0`	( 1 ) Lakers vs. ( 2 ) San Antonio Spurs : `` Los Angeles Lakers Win 4-0	`The study included 752 universities in Pennsylvania , including public schools , public charter schools and traditional public magnet schools .`
`( 1 ) Los Angeles Lakers vs. ( 2 ) San Antonio Spurs : Lakers win series 4-0`	`( 1 ) Los Angeles Lakers vs. ( 2 ) San Antonio Spurs : Lakers win series 4-0`	( 1 ) Lakers vs. ( 2 ) San Antonio Spurs : `` Los Angeles Lakers Win 4-0

Loss: losses.ArcFaceInBatchLoss with these parameters:

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

Evaluation Dataset

LangCache Sentence Pairs (all)

Dataset: LangCache Sentence Pairs (all)
Size: 6,786 evaluation samples
Columns: anchor, positive, and negative

Approximate statistics based on the first 1000 samples:

	anchor	positive	negative
type	string	string	string
details	min: 9 tokens mean: 27.96 tokens max: 50 tokens	min: 9 tokens mean: 27.98 tokens max: 51 tokens	min: 9 tokens mean: 27.56 tokens max: 49 tokens

Samples:

anchor	positive	negative
( 1 ) Lakers vs. ( 2 ) San Antonio Spurs : `` Los Angeles Lakers Win 4-0	`( 1 ) Lakers vs. ( 2 ) San Antonio Spurs : Los Angeles Lakers win series 4-0`	`( 1 ) Los Angeles Lakers vs. ( 2 ) San Antonio Spurs : Lakers win series 4-0`
`( 1 ) Lakers vs. ( 2 ) San Antonio Spurs : Los Angeles Lakers win series 4-0`	( 1 ) Lakers vs. ( 2 ) San Antonio Spurs : `` Los Angeles Lakers Win 4-0	`The study included 752 universities in Pennsylvania , including public schools , public charter schools and traditional public magnet schools .`
`( 1 ) Los Angeles Lakers vs. ( 2 ) San Antonio Spurs : Lakers win series 4-0`	`( 1 ) Los Angeles Lakers vs. ( 2 ) San Antonio Spurs : Lakers win series 4-0`	( 1 ) Lakers vs. ( 2 ) San Antonio Spurs : `` Los Angeles Lakers Win 4-0

Loss: losses.ArcFaceInBatchLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 300
per_device_eval_batch_size: 300
gradient_accumulation_steps: 2
weight_decay: 0.001
adam_beta2: 0.98
adam_epsilon: 1e-06
num_train_epochs: 1
warmup_ratio: 0.05
bf16: True
dataloader_num_workers: 4
dataloader_prefetch_factor: 4
load_best_model_at_end: True
optim: stable_adamw
ddp_find_unused_parameters: False
dataloader_persistent_workers: True
push_to_hub: True
hub_model_id: redis/langcache-embed-v3
eval_on_start: 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: 300
per_device_eval_batch_size: 300
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 2
eval_accumulation_steps: None
torch_empty_cache_steps: None
learning_rate: 5e-05
weight_decay: 0.001
adam_beta1: 0.9
adam_beta2: 0.98
adam_epsilon: 1e-06
max_grad_norm: 1.0
num_train_epochs: 1
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.05
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: 4
dataloader_prefetch_factor: 4
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}
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}
parallelism_config: None
deepspeed: None
label_smoothing_factor: 0.0
optim: stable_adamw
optim_args: None
adafactor: False
group_by_length: False
length_column_name: length
ddp_find_unused_parameters: False
ddp_bucket_cap_mb: None
ddp_broadcast_buffers: False
dataloader_pin_memory: True
dataloader_persistent_workers: True
skip_memory_metrics: True
use_legacy_prediction_loop: False
push_to_hub: True
resume_from_checkpoint: None
hub_model_id: redis/langcache-embed-v3
hub_strategy: every_save
hub_private_repo: None
hub_always_push: False
hub_revision: None
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: True
use_liger_kernel: False
liger_kernel_config: None
eval_use_gather_object: False
average_tokens_across_devices: False
prompts: None
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	Validation Loss	test_cosine_ndcg@10
0	0	1.0850	0.7718

Framework Versions

Python: 3.12.3
Sentence Transformers: 5.1.0
Transformers: 4.56.0
PyTorch: 2.8.0+cu128
Accelerate: 1.10.1
Datasets: 4.0.0
Tokenizers: 0.22.0

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",
}