srikarvar/fine_tuned_model_1

SentenceTransformer based on intfloat/multilingual-e5-small

This is a sentence-transformers model finetuned from intfloat/multilingual-e5-small. 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: intfloat/multilingual-e5-small
• Maximum Sequence Length: 512 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': 512, '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("srikarvar/fine_tuned_model_1")
# Run inference
sentences = [
    'How is the weather today?',
    'What is the weather like today?',
    'Who was the first female Prime Minister of the UK?',
]
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

Binary Classification

• Dataset: pair-class-dev
• Evaluated with BinaryClassificationEvaluator

Metric	Value
cosine_accuracy	0.9389
cosine_accuracy_threshold	0.7887
cosine_f1	0.9412
cosine_f1_threshold	0.7887
cosine_precision	0.9573
cosine_recall	0.9256
cosine_ap	0.974
dot_accuracy	0.9389
dot_accuracy_threshold	0.7887
dot_f1	0.9412
dot_f1_threshold	0.7887
dot_precision	0.9573
dot_recall	0.9256
dot_ap	0.974
manhattan_accuracy	0.9389
manhattan_accuracy_threshold	10.1324
manhattan_f1	0.9412
manhattan_f1_threshold	10.1324
manhattan_precision	0.9573
manhattan_recall	0.9256
manhattan_ap	0.9729
euclidean_accuracy	0.9389
euclidean_accuracy_threshold	0.6501
euclidean_f1	0.9412
euclidean_f1_threshold	0.6501
euclidean_precision	0.9573
euclidean_recall	0.9256
euclidean_ap	0.974
max_accuracy	0.9389
max_accuracy_threshold	10.1324
max_f1	0.9412
max_f1_threshold	10.1324
max_precision	0.9573
max_recall	0.9256
max_ap	0.974

Binary Classification

• Dataset: pair-class-test
• Evaluated with BinaryClassificationEvaluator

Metric	Value
cosine_accuracy	0.9389
cosine_accuracy_threshold	0.8208
cosine_f1	0.9421
cosine_f1_threshold	0.8208
cosine_precision	0.9421
cosine_recall	0.9421
cosine_ap	0.9732
dot_accuracy	0.9389
dot_accuracy_threshold	0.8208
dot_f1	0.9421
dot_f1_threshold	0.8208
dot_precision	0.9421
dot_recall	0.9421
dot_ap	0.9732
manhattan_accuracy	0.9345
manhattan_accuracy_threshold	9.3871
manhattan_f1	0.9383
manhattan_f1_threshold	9.5161
manhattan_precision	0.9344
manhattan_recall	0.9421
manhattan_ap	0.9721
euclidean_accuracy	0.9389
euclidean_accuracy_threshold	0.5987
euclidean_f1	0.9421
euclidean_f1_threshold	0.5987
euclidean_precision	0.9421
euclidean_recall	0.9421
euclidean_ap	0.9732
max_accuracy	0.9389
max_accuracy_threshold	9.3871
max_f1	0.9421
max_f1_threshold	9.5161
max_precision	0.9421
max_recall	0.9421
max_ap	0.9732

Training Details

Training Dataset

Unnamed Dataset

• Size: 916 training samples
• Columns: label, sentence2, and sentence1
• Approximate statistics based on the first 1000 samples:

  	label	sentence2	sentence1
  type	int	string	string
  details	0: ~49.56% 1: ~50.44%	min: 4 tokens mean: 10.32 tokens max: 22 tokens	min: 6 tokens mean: 10.92 tokens max: 22 tokens

• Samples:

  label	sentence2	sentence1
  1	What are the potential side effects of this medication?	What are the side effects of this drug?
  0	How to fix a torn pocket?	How to fix a broken zipper?
  0	How to make a chocolate chip cookie dough?	How to bake a chocolate chip cookie?

• Loss: OnlineContrastiveLoss

Evaluation Dataset

Unnamed Dataset

• Size: 229 evaluation samples
• Columns: label, sentence2, and sentence1
• Approximate statistics based on the first 1000 samples:

  	label	sentence2	sentence1
  type	int	string	string
  details	0: ~47.16% 1: ~52.84%	min: 4 tokens mean: 9.95 tokens max: 16 tokens	min: 6 tokens mean: 10.81 tokens max: 20 tokens

• Samples:

  label	sentence2	sentence1
  0	What methods are used to measure a nation's GDP?	How is the GDP of a country measured?
  0	What is the currency of Japan?	What is the currency of China?
  1	Steps to cultivate tomatoes at home	How to grow tomatoes in a garden?

• Loss: OnlineContrastiveLoss

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: epoch
• per_device_train_batch_size: 32
• per_device_eval_batch_size: 32
• gradient_accumulation_steps: 2
• weight_decay: 0.01
• num_train_epochs: 8
• warmup_ratio: 0.1
• load_best_model_at_end: True
• optim: adamw_torch_fused
• batch_sampler: no_duplicates

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: epoch
• 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: 2
• eval_accumulation_steps: None
• learning_rate: 5e-05
• weight_decay: 0.01
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• max_grad_norm: 1.0
• num_train_epochs: 8
• 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: 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: 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}
• 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: 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
• batch_sampler: no_duplicates
• multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	loss	pair-class-dev_max_ap	pair-class-test_max_ap
0	0	-	-	0.7130	-
0.6897	10	3.0972	-	-	-
1.0345	15	-	0.8033	0.9272	-
1.3448	20	1.0451	-	-	-
2.0345	30	0.5786	0.4910	0.9680	-
2.6897	40	0.2996	-	-	-
3.0345	45	-	0.4487	0.9731	-
3.3448	50	0.0901	-	-	-
4.0345	60	0.067	0.4115	0.9732	-
4.6897	70	0.0729	-	-	-
5.0345	75	-	0.4543	0.9727	-
5.3448	80	0.0453	-	-	-
6.0345	90	0.0637	0.4249	0.9736	-
6.6897	100	0.0388	-	-	-
7.0345	105	-	0.4223	0.9740	-
7.3448	110	0.0382	-	-	-
7.4828	112	-	0.4226	0.9740	0.9732

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.10.12
• Sentence Transformers: 3.0.1
• Transformers: 4.41.2
• PyTorch: 2.1.2+cu121
• Accelerate: 0.32.1
• Datasets: 2.19.1
• 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",
}