EccomerceEmbeddingModel / README.md

IshTale

Rename model/README.md to README.md

14305f7 verified 5 months ago

preview code

raw

history blame contribute delete

13.7 kB

metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - dense
  - generated_from_trainer
  - dataset_size:269012
  - loss:CoSENTLoss
base_model: intfloat/e5-large-v2
widget:
  - source_sentence: smart cutting machine for crafts
    sentences:
      - HyperX Cloud Alpha Wireless Gaming Headset
      - Rubbermaid Brilliance 20-Piece Food Storage Set
      - >-
        Men's Wick Short Sleeve Crew - Light Merino Wool Camo Hunting Shirt, UV
        Protection Moisture Management Base Layer
  - source_sentence: high capacity portable hard drive
    sentences:
      - Mr. Heater Big Buddy Portable Propane Heater
      - Samsung Galaxy Watch 5 Pro
      - >-
        Sun Bum Original SPF 45 Sunscreen Mist - Broad Spectrum Moisturizing
        Facial Sunscreen Spray with Vitamin E - Hawaii 104 Act Compliant (Made
        without Octinoxate & Oxybenzone) - Travel Friendly - 3.4 oz
  - source_sentence: fluid acrylics for pouring art
    sentences:
      - >-
        Linen Suit for Men 2 Pieces Slim Fit Casual Suits Groomsmen Tuxedos
        Wedding Party Blazer Pants Set Beige
      - Mejuri Small Hoop Earrings in Gold
      - Singer Start 1304 Sewing Machine
  - source_sentence: premium wireless gaming headset
    sentences:
      - Vornado MVH Whole Room Heater
      - >-
        Westinghouse 11000 Peak Watt Tri-Fuel Portable Inverter Generator,
        Remote Start, Transfer Switch Ready, Gas/Propane/Natural Gas Powered,
        Low THD, Safe for Electronics, Parallel Capable, CO Sensor
      - >-
        Rattaner Patio Wicker Furniture Set 6 Pieces Outdoor HDPE Wicker
        Conversation Couch Sectional Chair Sofa Set with Grey Cushions
  - source_sentence: travel system with stroller and car seat
    sentences:
      - Chemex Classic Series Pour-Over Glass Coffeemaker
      - David Yurman Cable Classic Bracelet
      - Legion Stonehenge Paper Pad
pipeline_tag: sentence-similarity
library_name: sentence-transformers

SentenceTransformer based on intfloat/e5-large-v2

This is a sentence-transformers model finetuned from intfloat/e5-large-v2. It maps sentences & paragraphs to a 1024-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/e5-large-v2
Maximum Sequence Length: 512 tokens
Output Dimensionality: 1024 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, 'architecture': 'BertModel'})
  (1): Pooling({'word_embedding_dimension': 1024, '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("sentence_transformers_model_id")
# Run inference
sentences = [
    'travel system with stroller and car seat',
    'Chemex Classic Series Pour-Over Glass Coffeemaker',
    'Legion Stonehenge Paper Pad',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.5295, 0.5210],
#         [0.5295, 1.0000, 0.5429],
#         [0.5210, 0.5429, 1.0000]])

Training Details

Training Dataset

Unnamed Dataset

Size: 269,012 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: 5 tokens
mean: 9.64 tokens
max: 21 tokens

min: 6 tokens
mean: 20.82 tokens
max: 99 tokens

min: -1.0
mean: 0.04
max: 0.99

	sentence_0	sentence_1	label
type	string	string	float
details	min: 5 tokens mean: 9.64 tokens max: 21 tokens	min: 6 tokens mean: 20.82 tokens max: 99 tokens	min: -1.0 mean: 0.04 max: 0.99

Samples:

sentence_0	sentence_1	label
`razor set with handle and blades`	`Hahnemühle Watercolor Journal`	`-0.8008412511835391`
`mini perfume atomizer for refillable travel scent`	`LISAPACK Perfume Travel Refillable Bottle - Atomizer Cologne Spray for Men Portable - Mini Sprayer Empty for Refill - Small Size 8ML Striped (Grey, Black, Silver)`	`0.85625`
`pour-over glass coffeemaker`	`Shark Navigator Lift-Away Professional NV356E Vacuum`	`0.131319533933279`

Loss: CoSENTLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 32
per_device_eval_batch_size: 32
num_train_epochs: 1
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: 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: 1
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}
parallelism_config: None
deepspeed: None
label_smoothing_factor: 0.0
optim: adamw_torch_fused
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
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: False
use_liger_kernel: False
liger_kernel_config: None
eval_use_gather_object: False
average_tokens_across_devices: False
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: round_robin
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss
0.0595	500	5.606
0.1189	1000	5.5059
0.1784	1500	5.4614
0.2379	2000	5.4299
0.2974	2500	5.415
0.3568	3000	5.4104
0.4163	3500	5.3718
0.4758	4000	5.3755
0.5353	4500	5.3545
0.5947	5000	5.3498
0.6542	5500	5.3392
0.7137	6000	5.3521
0.7732	6500	5.3248
0.8326	7000	5.3044
0.8921	7500	5.2916
0.9516	8000	5.2891

Framework Versions

Python: 3.12.11
Sentence Transformers: 5.1.0
Transformers: 4.56.0
PyTorch: 2.8.0+cu126
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",
}

CoSENTLoss

@online{kexuefm-8847,
    title={CoSENT: A more efficient sentence vector scheme than Sentence-BERT},
    author={Su Jianlin},
    year={2022},
    month={Jan},
    url={https://kexue.fm/archives/8847},
}