| --- |
| tags: |
| - question-answering |
| - bert |
| --- |
| |
| # Model Card for dynamic_tinybert |
| |
| # Model Details |
| |
| ## Model Description |
| |
| Dynamic-TinyBERT: Boost TinyBERT’s Inference Efficiency by Dynamic Sequence Length |
| |
| |
| - **Developed by:** Intel |
| - **Shared by [Optional]:** Intel |
| - **Model type:** Question Answering |
| - **Language(s) (NLP):** More information needed |
| - **License:** More information needed |
| - **Parent Model:** BERT |
| - **Resources for more information:** |
| - [Associated Paper](https://neurips2021-nlp.github.io/papers/16/CameraReady/Dynamic_TinyBERT_NLSP2021_camera_ready.pdf) |
| |
| |
| |
| |
| # Uses |
| |
| |
| ## Direct Use |
| This model can be used for the task of question answering. |
| |
| ## Downstream Use [Optional] |
| |
| More information needed. |
| |
| ## Out-of-Scope Use |
| |
| The model should not be used to intentionally create hostile or alienating environments for people. |
| |
| # Bias, Risks, and Limitations |
| |
| |
| Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al. (2021)](https://aclanthology.org/2021.acl-long.330.pdf) and [Bender et al. (2021)](https://dl.acm.org/doi/pdf/10.1145/3442188.3445922)). Predictions generated by the model may include disturbing and harmful stereotypes across protected classes; identity characteristics; and sensitive, social, and occupational groups. |
| |
| |
| |
| ## Recommendations |
| |
| |
| Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. |
| |
| # Training Details |
| |
| ## Training Data |
| |
| The model authors note in the [associated paper](https://neurips2021-nlp.github.io/papers/16/CameraReady/Dynamic_TinyBERT_NLSP2021_camera_ready.pdf): |
| > All our experiments are evaluated on the challenging question-answering benchmark SQuAD1.1 [11]. |
| |
| |
| ## Training Procedure |
| |
| |
| ### Preprocessing |
| |
| The model authors note in the [associated paper](https://neurips2021-nlp.github.io/papers/16/CameraReady/Dynamic_TinyBERT_NLSP2021_camera_ready.pdf): |
| > We start with a pre-trained general-TinyBERT student, which was trained to learn the general knowledge of BERT using the general-distillation method presented by TinyBERT. We perform transformer distillation from a fine- tuned BERT teacher to the student, following the same training steps used in the original TinyBERT: (1) **intermediate-layer distillation (ID)** — learning the knowledge residing in the hidden states and attentions matrices, and (2) **prediction-layer distillation (PD)** — fitting the predictions of the teacher. |
| |
| |
| |
| |
| ### Speeds, Sizes, Times |
| |
| The model authors note in the [associated paper](https://neurips2021-nlp.github.io/papers/16/CameraReady/Dynamic_TinyBERT_NLSP2021_camera_ready.pdf): |
| >For our Dynamic-TinyBERT model we use the architecture of TinyBERT6L: a small BERT model with 6 layers, a hidden size of 768, a feed forward size of 3072 and 12 heads. |
| |
| # Evaluation |
| |
| |
| ## Testing Data, Factors & Metrics |
| |
| ### Testing Data |
| |
| More information needed |
| |
| ### Factors |
| More information needed |
| |
| ### Metrics |
| |
| More information needed |
| |
| |
| ## Results |
| |
| The model authors note in the [associated paper](https://neurips2021-nlp.github.io/papers/16/CameraReady/Dynamic_TinyBERT_NLSP2021_camera_ready.pdf): |
| |
| | Model | Max F1 (full model) | Best Speedup within BERT-1% | |
| |------------------|---------------------|-----------------------------| |
| | Dynamic-TinyBERT | 88.71 | 3.3x | |
| |
| |
| |
| |
| # Model Examination |
| |
| More information needed |
| |
| # Environmental Impact |
| |
| Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). |
| |
| - **Hardware Type:** Titan GPU |
| - **Hours used:** More information needed |
| - **Cloud Provider:** More information needed |
| - **Compute Region:** More information needed |
| - **Carbon Emitted:** More information needed |
| |
| # Technical Specifications [optional] |
| |
| ## Model Architecture and Objective |
| |
| More information needed |
| |
| ## Compute Infrastructure |
| |
| More information needed |
| |
| ### Hardware |
| |
| |
| More information needed |
| |
| ### Software |
| |
| More information needed. |
| |
| # Citation |
| |
| |
| **BibTeX:** |
| |
| ```bibtex |
| @misc{https://doi.org/10.48550/arxiv.2111.09645, |
| doi = {10.48550/ARXIV.2111.09645}, |
| |
| url = {https://arxiv.org/abs/2111.09645}, |
| |
| author = {Guskin, Shira and Wasserblat, Moshe and Ding, Ke and Kim, Gyuwan}, |
| |
| keywords = {Computation and Language (cs.CL), Machine Learning (cs.LG), FOS: Computer and information sciences, FOS: Computer and information sciences}, |
| |
| title = {Dynamic-TinyBERT: Boost TinyBERT's Inference Efficiency by Dynamic Sequence Length}, |
| |
| publisher = {arXiv}, |
| |
| year = {2021}, |
| ``` |
| |
| |
| |
| **APA:** |
| |
| More information needed |
| |
| # Glossary [optional] |
| |
| More information needed |
| |
| # More Information [optional] |
| More information needed |
| |
| # Model Card Authors [optional] |
| |
| Intel in collaboration with Ezi Ozoani and the Hugging Face team |
| |
| # Model Card Contact |
| |
| More information needed |
| |
| # How to Get Started with the Model |
| |
| Use the code below to get started with the model. |
| |
| <details> |
| <summary> Click to expand </summary> |
| |
| ```python |
| from transformers import AutoTokenizer, AutoModelForQuestionAnswering |
| |
| tokenizer = AutoTokenizer.from_pretrained("Intel/dynamic_tinybert") |
| |
| model = AutoModelForQuestionAnswering.from_pretrained("Intel/dynamic_tinybert") |
| ``` |
| </details> |
| |
| |
