Nessrine9/finetuned2-MiniLM-L12-v2

SentenceTransformer based on sentence-transformers/all-MiniLM-L12-v2

This is a sentence-transformers model finetuned from sentence-transformers/all-MiniLM-L12-v2. It maps sentences & paragraphs to a 384-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: sentence-transformers/all-MiniLM-L12-v2
• Maximum Sequence Length: 128 tokens
• Output Dimensionality: 384 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': 128, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 384, '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("Nessrine9/finetuned2-MiniLM-L12-v2")
# Run inference
sentences = [
    'A large crowd watches as a couple tap dances together on a wooden floor.',
    'A man swings a golf club.',
    'A man crashes his car into the grocery store.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Evaluation

Metrics

Semantic Similarity

• Dataset: snli-dev
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.5007
spearman_cosine	0.4931
pearson_manhattan	0.4738
spearman_manhattan	0.4923
pearson_euclidean	0.475
spearman_euclidean	0.4931
pearson_dot	0.5007
spearman_dot	0.4931
pearson_max	0.5007
spearman_max	0.4931

Training Details

Training Dataset

Unnamed Dataset

• Size: 100,000 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: 7 tokens mean: 16.85 tokens max: 67 tokens	min: 5 tokens mean: 10.61 tokens max: 29 tokens	min: 0.0 mean: 0.5 max: 1.0

• Samples:

  sentence_0	sentence_1	label
  A biker is practicing a trick while his friend watch him as his audience.	man riding the bike to show his talent to his girlfriend.	0.5
  A man in a brown jacket standing in front of an open porch door.	A man is standing in front of the porch door.	0.0
  Two men and three children are at the beach.	Five people enjoying their vacation.	0.5

• 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: 16
• per_device_eval_batch_size: 16
• num_train_epochs: 4
• fp16: True
• 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: 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
• 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: 4
• 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: 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
• eval_use_gather_object: False
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: round_robin

Training Logs

Epoch	Step	Training Loss	snli-dev_spearman_max
0.08	500	0.1807	0.3001
0.16	1000	0.1497	0.3646
0.24	1500	0.1443	0.3652
0.32	2000	0.1394	0.3860
0.4	2500	0.1369	0.3810
0.48	3000	0.1346	0.3895
0.56	3500	0.1358	0.4147
0.64	4000	0.1387	0.4190
0.72	4500	0.131	0.4254
0.8	5000	0.1314	0.4219
0.88	5500	0.1288	0.4342
0.96	6000	0.1299	0.4135
1.0	6250	-	0.4393
1.04	6500	0.1306	0.4565
1.12	7000	0.1253	0.4433
1.2	7500	0.1275	0.4486
1.28	8000	0.1265	0.4616
1.3600	8500	0.1237	0.4462
1.44	9000	0.1223	0.4573
1.52	9500	0.123	0.4609
1.6	10000	0.1251	0.4678
1.6800	10500	0.1262	0.4500
1.76	11000	0.1194	0.4696
1.8400	11500	0.1206	0.4733
1.92	12000	0.118	0.4701
2.0	12500	0.1238	0.4688
2.08	13000	0.1191	0.4646
2.16	13500	0.1179	0.4757
2.24	14000	0.1177	0.4652
2.32	14500	0.1176	0.4873
2.4	15000	0.115	0.4674
2.48	15500	0.1141	0.4784
2.56	16000	0.1143	0.4824
2.64	16500	0.1184	0.4898
2.7200	17000	0.1124	0.4818
2.8	17500	0.1141	0.4905
2.88	18000	0.1115	0.4850
2.96	18500	0.1123	0.4867
3.0	18750	-	0.4867
3.04	19000	0.1149	0.4849
3.12	19500	0.1114	0.4888
3.2	20000	0.1124	0.4903
3.2800	20500	0.1124	0.4900
3.36	21000	0.1088	0.4871
3.44	21500	0.1065	0.4835
3.52	22000	0.1075	0.4912
3.6	22500	0.1115	0.4944
3.68	23000	0.1122	0.4932
3.76	23500	0.1074	0.4917
3.84	24000	0.1081	0.4923
3.92	24500	0.1057	0.4921
4.0	25000	0.1118	0.4931

Framework Versions

• Python: 3.10.12
• Sentence Transformers: 3.2.1
• Transformers: 4.44.2
• PyTorch: 2.5.0+cu121
• Accelerate: 0.34.2
• Datasets: 3.0.2
• 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",
}