SentenceTransformer based on sentence-transformers/distilbert-base-nli-mean-tokens

This is a sentence-transformers model finetuned from sentence-transformers/distilbert-base-nli-mean-tokens. It maps sentences & paragraphs to a 768-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/distilbert-base-nli-mean-tokens
Maximum Sequence Length: 128 tokens
Output Dimensionality: 768 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: DistilBertModel 
  (1): Pooling({'word_embedding_dimension': 768, '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})
)

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("DivyaMereddy007/RecipeBert_v5original_epoc50Copy_of_TrainSetenceTransforme-Finetuning_v5_DistilledBert")
# Run inference
sentences = [
    'Watermelon Rind Pickles ["7 lb. watermelon rind", "7 c. sugar", "2 c. apple vinegar", "1/2 tsp. oil of cloves", "1/2 tsp. oil of cinnamon"] ["Trim off green and pink parts of watermelon rind; cut to 1-inch cubes.", "Parboil until tender, but not soft.", "Drain. Combine sugar, vinegar, oil of cloves and oil of cinnamon; bring to boiling and pour over rind.", "Let stand overnight.", "In the morning, drain off syrup.", "Heat and put over rind.", "The third morning, heat rind and syrup; seal in hot, sterilized jars.", "Makes 8 pints.", "(Oil of cinnamon and clove keeps rind clear and transparent.)"]',
    'Summer Chicken ["1 pkg. chicken cutlets", "1/2 c. oil", "1/3 c. red vinegar", "2 Tbsp. oregano", "2 Tbsp. garlic salt"] ["Double recipe for more chicken."]',
    'Summer Spaghetti ["1 lb. very thin spaghetti", "1/2 bottle McCormick Salad Supreme (seasoning)", "1 bottle Zesty Italian dressing"] ["Prepare spaghetti per package.", "Drain.", "Melt a little butter through it.", "Marinate overnight in Salad Supreme and Zesty Italian dressing.", "Just before serving, add cucumbers, tomatoes, green peppers, mushrooms, olives or whatever your taste may want."]',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

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

Training Details

Training Dataset

Unnamed Dataset

Size: 1,746 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: 63 tokens mean: 118.82 tokens max: 128 tokens	min: 63 tokens mean: 118.59 tokens max: 128 tokens	min: 0.0 mean: 0.19 max: 1.0

Samples:

sentence_0	sentence_1	label
`Tuna Macaroni Casserole ["1 box macaroni and cheese", "1 can tuna, drained", "1 small jar pimentos", "1 medium onion, chopped"] ["Prepare macaroni and cheese as directed.", "Add drained tuna, pimento and onion.", "Mix.", "Serve hot or cold."]`	`Easy Fudge ["1 (14 oz.) can sweetened condensed milk", "1 (12 oz.) pkg. semi-sweet chocolate chips", "1 (1 oz.) sq. unsweetened chocolate (if desired)", "1 1/2 c. chopped nuts (if desired)", "1 tsp. vanilla"] ["Butter a square pan, 8 x 8 x 2-inches.", "Heat milk, chocolate chips and unsweetened chocolate over low heat, stirring constantly, until chocolate is melted and mixture is smooth. Remove from heat.", "Stir in nuts and vanilla.", "Spread in pan."]`	`0.05`
`Scalloped Corn ["1 can cream-style corn", "1 can whole kernel corn", "1/2 pkg. (approximately 20) saltine crackers, crushed", "1 egg, beaten", "6 tsp. butter, divided", "pepper to taste"] ["Mix together both cans of corn, crackers, egg, 2 teaspoons of melted butter and pepper and place in a buttered baking dish.", "Dot with remaining 4 teaspoons of butter.", "Bake at 350\u00b0 for 1 hour."]`	`Quick Peppermint Puffs ["8 marshmallows", "2 Tbsp. margarine, melted", "1/4 c. crushed peppermint candy", "1 can crescent rolls"] ["Dip marshmallows in melted margarine; roll in candy. Wrap a crescent triangle around each marshmallow, completely covering the marshmallow and square edges of dough tightly to seal.", "Dip in margarine and place in a greased muffin tin.", "Bake at 375\u00b0 for 10 to 15 minutes; remove from pan."]`	`0.1`
`Beer Bread ["3 c. self rising flour", "1 - 12 oz. can beer", "1 Tbsp. sugar"] ["Stir the ingredients together and put in a greased and floured loaf pan.", "Bake at 425 degrees for 50 minutes.", "Drizzle melted butter on top."]`	`Rhubarb Coffee Cake ["1 1/2 c. sugar", "1/2 c. butter", "1 egg", "1 c. buttermilk", "2 c. flour", "1/2 tsp. salt", "1 tsp. soda", "1 c. buttermilk", "2 c. rhubarb, finely cut", "1 tsp. vanilla"] ["Cream sugar and butter.", "Add egg and beat well.", "To creamed butter, sugar and egg, add alternately buttermilk with mixture of flour, salt and soda.", "Mix well.", "Add rhubarb and vanilla.", "Pour into greased 9 x 13-inch pan and add Topping."]`	`0.4`

Loss: CosineSimilarityLoss with these parameters:

{
    "loss_fct": "torch.nn.modules.loss.MSELoss"
}

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 16
per_device_eval_batch_size: 16
num_train_epochs: 50
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: 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
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: 50
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}
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
batch_sampler: batch_sampler
multi_dataset_batch_sampler: round_robin

Training Logs

Epoch	Step	Training Loss
4.5455	500	0.0092
9.0909	1000	0.0091
13.6364	1500	0.0081
18.1818	2000	0.0074
22.7273	2500	0.0071
27.2727	3000	0.0069
31.8182	3500	0.0066
36.3636	4000	0.0065
40.9091	4500	0.0061
45.4545	5000	0.006
50.0	5500	0.0056

Framework Versions

Python: 3.10.12
Sentence Transformers: 3.0.1
Transformers: 4.41.2
PyTorch: 2.3.0+cu121
Accelerate: 0.31.0
Datasets: 2.19.2
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",
}

DivyaMereddy007
/

RecipeBert_v5original_epoc50Copy_of_TrainSetenceTransforme-Finetuning_v5_DistilledBert