krishanusinha20/multi-agentic-sql-generator-model
6106
1---2tags:3- sentence-transformers4- sentence-similarity5- feature-extraction6- generated_from_trainer7- dataset_size:108- loss:CosineSimilarityLoss9base_model: sentence-transformers/all-MiniLM-L6-v210widget:11- source_sentence: Find the most popular payment method used in 2024.12 sentences:13 - SELECT * FROM orders WHERE customer_id = 42;14 - SELECT customer_id, COUNT(order_id) AS order_count FROM orders WHERE order_date15 BETWEEN '2024-01-01' AND '2024-12-31' GROUP BY customer_id HAVING order_count16 >= 3;17 - SELECT payment_method, COUNT(*) AS usage_count FROM payments WHERE payment_date18 BETWEEN '2024-01-01' AND '2024-12-31' GROUP BY payment_method ORDER BY usage_count19 DESC LIMIT 1;20- source_sentence: Which products sold the most in 2024?21 sentences:22 - SELECT COUNT(*) AS total_orders FROM orders WHERE order_date >= DATE('now', '-623 months');24 - SELECT p.category, SUM(oi.subtotal) AS total_revenue FROM order_items oi JOIN25 products p ON oi.product_id = p.product_id GROUP BY p.category ORDER BY total_revenue26 DESC LIMIT 3;27 - SELECT product_id, SUM(quantity) AS total_sold FROM order_items JOIN orders ON28 order_items.order_id = orders.order_id WHERE order_date BETWEEN '2024-01-01' AND29 '2024-12-31' GROUP BY product_id ORDER BY total_sold DESC LIMIT 10;30pipeline_tag: sentence-similarity31library_name: sentence-transformers32---33 34# SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v235 36This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-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.37 38## Model Details39 40### Model Description41- **Model Type:** Sentence Transformer42- **Base model:** [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) <!-- at revision fa97f6e7cb1a59073dff9e6b13e2715cf7475ac9 -->43- **Maximum Sequence Length:** 256 tokens44- **Output Dimensionality:** 384 dimensions45- **Similarity Function:** Cosine Similarity46<!-- - **Training Dataset:** Unknown -->47<!-- - **Language:** Unknown -->48<!-- - **License:** Unknown -->49 50### Model Sources51 52- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)53- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)54- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)55 56### Full Model Architecture57 58```59SentenceTransformer(60 (0): Transformer({'max_seq_length': 256, 'do_lower_case': False}) with Transformer model: BertModel 61 (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})62 (2): Normalize()63)64```65 66## Usage67 68### Direct Usage (Sentence Transformers)69 70First install the Sentence Transformers library:71 72```bash73pip install -U sentence-transformers74```75 76Then you can load this model and run inference.77```python78from sentence_transformers import SentenceTransformer79 80# Download from the 🤗 Hub81model = SentenceTransformer("krishanusinha20/multi-agentic-sql-generator-model")82# Run inference83sentences = [84 'Which products sold the most in 2024?',85 "SELECT product_id, SUM(quantity) AS total_sold FROM order_items JOIN orders ON order_items.order_id = orders.order_id WHERE order_date BETWEEN '2024-01-01' AND '2024-12-31' GROUP BY product_id ORDER BY total_sold DESC LIMIT 10;",86 "SELECT COUNT(*) AS total_orders FROM orders WHERE order_date >= DATE('now', '-6 months');",87]88embeddings = model.encode(sentences)89print(embeddings.shape)90# [3, 384]91 92# Get the similarity scores for the embeddings93similarities = model.similarity(embeddings, embeddings)94print(similarities.shape)95# [3, 3]96```97 98<!--99### Direct Usage (Transformers)100 101<details><summary>Click to see the direct usage in Transformers</summary>102 103</details>104-->105 106<!--107### Downstream Usage (Sentence Transformers)108 109You can finetune this model on your own dataset.110 111<details><summary>Click to expand</summary>112 113</details>114-->115 116<!--117### Out-of-Scope Use118 119*List how the model may foreseeably be misused and address what users ought not to do with the model.*120-->121 122<!--123## Bias, Risks and Limitations124 125*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*126-->127 128<!--129### Recommendations130 131*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*132-->133 134## Training Details135 136### Training Dataset137 138#### Unnamed Dataset139 140* Size: 10 training samples141* Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code>142* Approximate statistics based on the first 10 samples:143 | | sentence_0 | sentence_1 | label |144 |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:--------------------------------------------------------------|145 | type | string | string | float |146 | details | <ul><li>min: 11 tokens</li><li>mean: 13.0 tokens</li><li>max: 15 tokens</li></ul> | <ul><li>min: 13 tokens</li><li>mean: 45.5 tokens</li><li>max: 72 tokens</li></ul> | <ul><li>min: 1.0</li><li>mean: 1.0</li><li>max: 1.0</li></ul> |147* Samples:148 | sentence_0 | sentence_1 | label |149 |:-------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------|150 | <code>Find the total revenue generated in 2024.</code> | <code>SELECT SUM(total_amount) AS total_revenue FROM orders WHERE order_date BETWEEN '2024-01-01' AND '2024-12-31';</code> | <code>1.0</code> |151 | <code>Find the top 3 product categories with the highest sales revenue.</code> | <code>SELECT p.category, SUM(oi.subtotal) AS total_revenue FROM order_items oi JOIN products p ON oi.product_id = p.product_id GROUP BY p.category ORDER BY total_revenue DESC LIMIT 3;</code> | <code>1.0</code> |152 | <code>How many orders were placed in the last 6 months?</code> | <code>SELECT COUNT(*) AS total_orders FROM orders WHERE order_date >= DATE('now', '-6 months');</code> | <code>1.0</code> |153* Loss: [<code>CosineSimilarityLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosinesimilarityloss) with these parameters:154 ```json155 {156 "loss_fct": "torch.nn.modules.loss.MSELoss"157 }158 ```159 160### Training Hyperparameters161#### Non-Default Hyperparameters162 163- `per_device_train_batch_size`: 4164- `per_device_eval_batch_size`: 4165- `num_train_epochs`: 5166- `multi_dataset_batch_sampler`: round_robin167 168#### All Hyperparameters169<details><summary>Click to expand</summary>170 171- `overwrite_output_dir`: False172- `do_predict`: False173- `eval_strategy`: no174- `prediction_loss_only`: True175- `per_device_train_batch_size`: 4176- `per_device_eval_batch_size`: 4177- `per_gpu_train_batch_size`: None178- `per_gpu_eval_batch_size`: None179- `gradient_accumulation_steps`: 1180- `eval_accumulation_steps`: None181- `torch_empty_cache_steps`: None182- `learning_rate`: 5e-05183- `weight_decay`: 0.0184- `adam_beta1`: 0.9185- `adam_beta2`: 0.999186- `adam_epsilon`: 1e-08187- `max_grad_norm`: 1188- `num_train_epochs`: 5189- `max_steps`: -1190- `lr_scheduler_type`: linear191- `lr_scheduler_kwargs`: {}192- `warmup_ratio`: 0.0193- `warmup_steps`: 0194- `log_level`: passive195- `log_level_replica`: warning196- `log_on_each_node`: True197- `logging_nan_inf_filter`: True198- `save_safetensors`: True199- `save_on_each_node`: False200- `save_only_model`: False201- `restore_callback_states_from_checkpoint`: False202- `no_cuda`: False203- `use_cpu`: False204- `use_mps_device`: False205- `seed`: 42206- `data_seed`: None207- `jit_mode_eval`: False208- `use_ipex`: False209- `bf16`: False210- `fp16`: False211- `fp16_opt_level`: O1212- `half_precision_backend`: auto213- `bf16_full_eval`: False214- `fp16_full_eval`: False215- `tf32`: None216- `local_rank`: 0217- `ddp_backend`: None218- `tpu_num_cores`: None219- `tpu_metrics_debug`: False220- `debug`: []221- `dataloader_drop_last`: False222- `dataloader_num_workers`: 0223- `dataloader_prefetch_factor`: None224- `past_index`: -1225- `disable_tqdm`: False226- `remove_unused_columns`: True227- `label_names`: None228- `load_best_model_at_end`: False229- `ignore_data_skip`: False230- `fsdp`: []231- `fsdp_min_num_params`: 0232- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}233- `fsdp_transformer_layer_cls_to_wrap`: None234- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}235- `deepspeed`: None236- `label_smoothing_factor`: 0.0237- `optim`: adamw_torch238- `optim_args`: None239- `adafactor`: False240- `group_by_length`: False241- `length_column_name`: length242- `ddp_find_unused_parameters`: None243- `ddp_bucket_cap_mb`: None244- `ddp_broadcast_buffers`: False245- `dataloader_pin_memory`: True246- `dataloader_persistent_workers`: False247- `skip_memory_metrics`: True248- `use_legacy_prediction_loop`: False249- `push_to_hub`: False250- `resume_from_checkpoint`: None251- `hub_model_id`: None252- `hub_strategy`: every_save253- `hub_private_repo`: None254- `hub_always_push`: False255- `gradient_checkpointing`: False256- `gradient_checkpointing_kwargs`: None257- `include_inputs_for_metrics`: False258- `include_for_metrics`: []259- `eval_do_concat_batches`: True260- `fp16_backend`: auto261- `push_to_hub_model_id`: None262- `push_to_hub_organization`: None263- `mp_parameters`: 264- `auto_find_batch_size`: False265- `full_determinism`: False266- `torchdynamo`: None267- `ray_scope`: last268- `ddp_timeout`: 1800269- `torch_compile`: False270- `torch_compile_backend`: None271- `torch_compile_mode`: None272- `dispatch_batches`: None273- `split_batches`: None274- `include_tokens_per_second`: False275- `include_num_input_tokens_seen`: False276- `neftune_noise_alpha`: None277- `optim_target_modules`: None278- `batch_eval_metrics`: False279- `eval_on_start`: False280- `use_liger_kernel`: False281- `eval_use_gather_object`: False282- `average_tokens_across_devices`: False283- `prompts`: None284- `batch_sampler`: batch_sampler285- `multi_dataset_batch_sampler`: round_robin286 287</details>288 289### Framework Versions290- Python: 3.11.11291- Sentence Transformers: 3.4.1292- Transformers: 4.48.3293- PyTorch: 2.5.1+cu124294- Accelerate: 1.3.0295- Datasets: 3.3.2296- Tokenizers: 0.21.0297 298## Citation299 300### BibTeX301 302#### Sentence Transformers303```bibtex304@inproceedings{reimers-2019-sentence-bert,305 title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",306 author = "Reimers, Nils and Gurevych, Iryna",307 booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",308 month = "11",309 year = "2019",310 publisher = "Association for Computational Linguistics",311 url = "https://arxiv.org/abs/1908.10084",312}313```314 315<!--316## Glossary317 318*Clearly define terms in order to be accessible across audiences.*319-->320 321<!--322## Model Card Authors323 324*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*325-->326 327<!--328## Model Card Contact329 330*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*331-->