sentenceTransformer_nepali_embedding

This is a sentence-transformers model finetuned from jangedoo/all-MiniLM-L6-v2-nepali on the json dataset. 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: jangedoo/all-MiniLM-L6-v2-nepali
Maximum Sequence Length: 256 tokens
Output Dimensionality: 384 dimensions
Similarity Function: Cosine Similarity
Training Dataset:
- json
Language: nep
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': 256, '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("ritesh-07/fine_tuned_model_02")
# Run inference
sentences = [
    'कुटनीतिक राहदानीको लागि निवेदनमा कस्तो ठेगाना विवरण चाहिन्छ?',
    'कुटनीतिक राहदानीको लागि निवेदनमा जिल्ला, गाउँ/नगरपालिका, वडा नम्बर, गाउँ/सडक, र घर नम्बरको ठेगाना विवरण चाहिन्छ।',
    'राहदानीको लागि कागजात धुल्याउने प्रक्रिया महानिर्देशकको स्वीकृतिमा हुन्छ।',
]
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

Information Retrieval

Dataset: dim_384
Evaluated with InformationRetrievalEvaluator with these parameters:
```
{
    "truncate_dim": 384
}
```

Metric	Value
cosine_accuracy@1	0.4103
cosine_accuracy@3	0.6581
cosine_accuracy@5	0.735
cosine_accuracy@10	0.8462
cosine_precision@1	0.4103
cosine_precision@3	0.2194
cosine_precision@5	0.147
cosine_precision@10	0.0846
cosine_recall@1	0.4103
cosine_recall@3	0.6581
cosine_recall@5	0.735
cosine_recall@10	0.8462
cosine_ndcg@10	0.6218
cosine_mrr@10	0.5504
cosine_map@100	0.5572

Information Retrieval

Dataset: dim_256
Evaluated with InformationRetrievalEvaluator with these parameters:
```
{
    "truncate_dim": 256
}
```

Metric	Value
cosine_accuracy@1	0.4274
cosine_accuracy@3	0.641
cosine_accuracy@5	0.7179
cosine_accuracy@10	0.8291
cosine_precision@1	0.4274
cosine_precision@3	0.2137
cosine_precision@5	0.1436
cosine_precision@10	0.0829
cosine_recall@1	0.4274
cosine_recall@3	0.641
cosine_recall@5	0.7179
cosine_recall@10	0.8291
cosine_ndcg@10	0.616
cosine_mrr@10	0.5488
cosine_map@100	0.5564

Information Retrieval

Dataset: dim_128
Evaluated with InformationRetrievalEvaluator with these parameters:
```
{
    "truncate_dim": 128
}
```

Metric	Value
cosine_accuracy@1	0.3932
cosine_accuracy@3	0.5812
cosine_accuracy@5	0.6752
cosine_accuracy@10	0.8034
cosine_precision@1	0.3932
cosine_precision@3	0.1937
cosine_precision@5	0.135
cosine_precision@10	0.0803
cosine_recall@1	0.3932
cosine_recall@3	0.5812
cosine_recall@5	0.6752
cosine_recall@10	0.8034
cosine_ndcg@10	0.5799
cosine_mrr@10	0.51
cosine_map@100	0.5176

Information Retrieval

Dataset: dim_64
Evaluated with InformationRetrievalEvaluator with these parameters:
```
{
    "truncate_dim": 64
}
```

Metric	Value
cosine_accuracy@1	0.3846
cosine_accuracy@3	0.5812
cosine_accuracy@5	0.641
cosine_accuracy@10	0.7607
cosine_precision@1	0.3846
cosine_precision@3	0.1937
cosine_precision@5	0.1282
cosine_precision@10	0.0761
cosine_recall@1	0.3846
cosine_recall@3	0.5812
cosine_recall@5	0.641
cosine_recall@10	0.7607
cosine_ndcg@10	0.5652
cosine_mrr@10	0.5037
cosine_map@100	0.514

Training Details

Training Dataset

json

Dataset: json
Size: 1,046 training samples
Columns: anchor and positive
Approximate statistics based on the first 1000 samples:
anchor positive
type string string
details
min: 18 tokens
mean: 40.9 tokens
max: 103 tokens

min: 23 tokens
mean: 65.74 tokens
max: 235 tokens

	anchor	positive
type	string	string
details	min: 18 tokens mean: 40.9 tokens max: 103 tokens	min: 23 tokens mean: 65.74 tokens max: 235 tokens

Samples:

anchor	positive
`राहदानी नियमावली, २०७७ मा अभिलेखको गोपनीयताको उल्लङ्घनको जाँचको नतिजाको अपील कसले जाँच गर्छ?`	`राहदानी नियमावली, २०७७ मा अभिलेखको गोपनीयताको उल्लङ्घनको जाँचको नतिजाको अपील मन्त्रालयले तोकेको समितिले जाँच गर्छ।`
`राहदानी नियमावली, २०७७ मा सत्यापनको लागि कस्तो सही चाहिन्छ?`	`राहदानी नियमावली, २०७७ मा सत्यापनको लागि निवेदकको सही, र नाबालकको हकमा बाबु, आमा, वा संरक्षकको सही चाहिन्छ।`
`राहदानी नियमावली, २०७७ मा कस्तो निकायले राहदानी जारी गर्छ?`	`राहदानी नियमावली, २०७७ मा विभाग, नियोग, वा जिल्ला प्रशासन कार्यालयले राहदानी जारी गर्छ।`

Loss: MatryoshkaLoss with these parameters:

{
    "loss": "MultipleNegativesRankingLoss",
    "matryoshka_dims": [
        384,
        256,
        128,
        64
    ],
    "matryoshka_weights": [
        1,
        1,
        1,
        1
    ],
    "n_dims_per_step": -1
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: epoch
per_device_train_batch_size: 32
per_device_eval_batch_size: 16
gradient_accumulation_steps: 16
learning_rate: 2e-05
num_train_epochs: 4
lr_scheduler_type: cosine
warmup_ratio: 0.1
bf16: True
tf32: False
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: 16
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 16
eval_accumulation_steps: None
torch_empty_cache_steps: None
learning_rate: 2e-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: 4
max_steps: -1
lr_scheduler_type: cosine
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: True
fp16: False
fp16_opt_level: O1
half_precision_backend: auto
bf16_full_eval: False
fp16_full_eval: False
tf32: False
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: 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: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	dim_384_cosine_ndcg@10	dim_256_cosine_ndcg@10	dim_128_cosine_ndcg@10	dim_64_cosine_ndcg@10
1.0	3	-	0.5232	0.5074	0.4679	0.4451
2.0	6	-	0.5891	0.5703	0.5555	0.5275
3.0	9	-	0.6108	0.6052	0.5815	0.5594
3.4848	10	2.5112	-	-	-	-
4.0	12	-	0.6218	0.6160	0.5799	0.5652

The bold row denotes the saved checkpoint.

Framework Versions

Python: 3.11.13
Sentence Transformers: 4.1.0
Transformers: 4.53.2
PyTorch: 2.6.0+cu124
Accelerate: 1.9.0
Datasets: 4.0.0
Tokenizers: 0.21.2

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",
}

MatryoshkaLoss

@misc{kusupati2024matryoshka,
    title={Matryoshka Representation Learning},
    author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
    year={2024},
    eprint={2205.13147},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

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}
}

Downloads last month: 10

Safetensors

Model size

22.7M params

Tensor type

F32

Model tree for ritesh-07/fine_tuned_model_02

Unable to build the model tree, the base model loops to the model itself. Learn more.

Evaluation results

Cosine Accuracy@1 on dim 384
self-reported

0.410
Cosine Accuracy@3 on dim 384
self-reported

0.658
Cosine Accuracy@5 on dim 384
self-reported

0.735
Cosine Accuracy@10 on dim 384
self-reported

0.846
Cosine Precision@1 on dim 384
self-reported

0.410
Cosine Precision@3 on dim 384
self-reported

0.219
Cosine Precision@5 on dim 384
self-reported

0.147
Cosine Precision@10 on dim 384
self-reported

0.085
Cosine Recall@1 on dim 384
self-reported

0.410
Cosine Recall@3 on dim 384
self-reported

0.658