Midna1980/t5-small-api

1---2language: 3- en4- fr5- ro6- de7- multilingual8license: apache-2.09tags:10- summarization11- translation12datasets:13- c414---15 16# Model Card for T5 Small17 18![model image](https://camo.githubusercontent.com/623b4dea0b653f2ad3f36c71ebfe749a677ac0a1/68747470733a2f2f6d69726f2e6d656469756d2e636f6d2f6d61782f343030362f312a44304a31674e51663876727255704b657944387750412e706e67)19 20#  Table of Contents21 221. [Model Details](#model-details)232. [Uses](#uses)243. [Bias, Risks, and Limitations](#bias-risks-and-limitations)254. [Training Details](#training-details)265. [Evaluation](#evaluation)276. [Environmental Impact](#environmental-impact)287. [Citation](#citation)298. [Model Card Authors](#model-card-authors)309. [How To Get Started With the Model](#how-to-get-started-with-the-model)31 32# Model Details33 34## Model Description35 36The developers of the Text-To-Text Transfer Transformer (T5) [write](https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html): 37 38> With T5, we propose reframing all NLP tasks into a unified text-to-text-format where the input and output are always text strings, in contrast to BERT-style models that can only output either a class label or a span of the input. Our text-to-text framework allows us to use the same model, loss function, and hyperparameters on any NLP task.39 40T5-Small is the checkpoint with 60 million parameters. 41 42- **Developed by:** Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, Peter J. Liu. See [associated paper](https://jmlr.org/papers/volume21/20-074/20-074.pdf) and [GitHub repo](https://github.com/google-research/text-to-text-transfer-transformer#released-model-checkpoints)43- **Model type:** Language model44- **Language(s) (NLP):** English, French, Romanian, German45- **License:** Apache 2.046- **Related Models:** [All T5 Checkpoints](https://huggingface.co/models?search=t5)47- **Resources for more information:**48  - [Research paper](https://jmlr.org/papers/volume21/20-074/20-074.pdf)49  - [Google's T5 Blog Post](https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html) 50  - [GitHub Repo](https://github.com/google-research/text-to-text-transfer-transformer)51  - [Hugging Face T5 Docs](https://huggingface.co/docs/transformers/model_doc/t5)52  53# Uses54 55## Direct Use and Downstream Use56 57The developers write in a [blog post](https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html) that the model: 58 59> Our text-to-text framework allows us to use the same model, loss function, and hyperparameters on any NLP task, including machine translation, document summarization, question answering, and classification tasks (e.g., sentiment analysis). We can even apply T5 to regression tasks by training it to predict the string representation of a number instead of the number itself.60 61See the [blog post](https://ai.googleblog.com/2020/02/exploring-transfer-learning-with-t5.html) and [research paper](https://jmlr.org/papers/volume21/20-074/20-074.pdf) for further details.62 63## Out-of-Scope Use64 65More information needed.66 67# Bias, Risks, and Limitations68 69More information needed.70 71## Recommendations72 73More information needed.74 75# Training Details76 77## Training Data78 79The model is pre-trained on the [Colossal Clean Crawled Corpus (C4)](https://www.tensorflow.org/datasets/catalog/c4), which was developed and released in the context of the same [research paper](https://jmlr.org/papers/volume21/20-074/20-074.pdf) as T5.80 81The model was pre-trained on a on a **multi-task mixture of unsupervised (1.) and supervised tasks (2.)**.82Thereby, the following datasets were being used for (1.) and (2.):83 841. **Datasets used for Unsupervised denoising objective**:85 86- [C4](https://huggingface.co/datasets/c4)87- [Wiki-DPR](https://huggingface.co/datasets/wiki_dpr)88 89 902. **Datasets used for Supervised text-to-text language modeling objective**91 92- Sentence acceptability judgment93  - CoLA [Warstadt et al., 2018](https://arxiv.org/abs/1805.12471)94- Sentiment analysis 95  - SST-2 [Socher et al., 2013](https://nlp.stanford.edu/~socherr/EMNLP2013_RNTN.pdf)96- Paraphrasing/sentence similarity97  - MRPC [Dolan and Brockett, 2005](https://aclanthology.org/I05-5002)98  - STS-B [Ceret al., 2017](https://arxiv.org/abs/1708.00055)99  - QQP [Iyer et al., 2017](https://quoradata.quora.com/First-Quora-Dataset-Release-Question-Pairs)100- Natural language inference101  - MNLI [Williams et al., 2017](https://arxiv.org/abs/1704.05426)102  - QNLI [Rajpurkar et al.,2016](https://arxiv.org/abs/1606.05250)103  - RTE [Dagan et al., 2005](https://link.springer.com/chapter/10.1007/11736790_9) 104  - CB [De Marneff et al., 2019](https://semanticsarchive.net/Archive/Tg3ZGI2M/Marneffe.pdf)105- Sentence completion106  - COPA [Roemmele et al., 2011](https://www.researchgate.net/publication/221251392_Choice_of_Plausible_Alternatives_An_Evaluation_of_Commonsense_Causal_Reasoning)107- Word sense disambiguation108  - WIC [Pilehvar and Camacho-Collados, 2018](https://arxiv.org/abs/1808.09121)109- Question answering110  - MultiRC [Khashabi et al., 2018](https://aclanthology.org/N18-1023)111  - ReCoRD [Zhang et al., 2018](https://arxiv.org/abs/1810.12885)112  - BoolQ [Clark et al., 2019](https://arxiv.org/abs/1905.10044)113 114## Training Procedure115 116In their [abstract](https://jmlr.org/papers/volume21/20-074/20-074.pdf), the model developers write: 117 118> In this paper, we explore the landscape of transfer learning techniques for NLP by introducing a unified framework that converts every language problem into a text-to-text format. Our systematic study compares pre-training objectives, architectures, unlabeled datasets, transfer approaches, and other factors on dozens of language understanding tasks. 119 120The framework introduced, the T5 framework, involves a training procedure that brings together the approaches studied in the paper. See the [research paper](https://jmlr.org/papers/volume21/20-074/20-074.pdf) for further details.121 122# Evaluation123 124## Testing Data, Factors & Metrics125 126The developers evaluated the model on 24 tasks, see the [research paper](https://jmlr.org/papers/volume21/20-074/20-074.pdf) for full details.127 128## Results 129 130For full results for T5-small, see the [research paper](https://jmlr.org/papers/volume21/20-074/20-074.pdf), Table 14.131 132# Environmental Impact133 134Carbon 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).135 136- **Hardware Type:** Google Cloud TPU Pods137- **Hours used:** More information needed138- **Cloud Provider:** GCP139- **Compute Region:** More information needed140- **Carbon Emitted:** More information needed141 142# Citation143 144**BibTeX:**145 146```bibtex147@article{2020t5,148  author  = {Colin Raffel and Noam Shazeer and Adam Roberts and Katherine Lee and Sharan Narang and Michael Matena and Yanqi Zhou and Wei Li and Peter J. Liu},149  title   = {Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer},150  journal = {Journal of Machine Learning Research},151  year    = {2020},152  volume  = {21},153  number  = {140},154  pages   = {1-67},155  url     = {http://jmlr.org/papers/v21/20-074.html}156}157```158 159**APA:**160- Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., ... & Liu, P. J. (2020). Exploring the limits of transfer learning with a unified text-to-text transformer. J. Mach. Learn. Res., 21(140), 1-67.161 162# Model Card Authors163 164This model card was written by the team at Hugging Face.165 166# How to Get Started with the Model167 168Use the code below to get started with the model.169 170<details>171<summary> Click to expand </summary>172 173```python174from transformers import T5Tokenizer, T5Model175 176tokenizer = T5Tokenizer.from_pretrained("t5-small")177model = T5Model.from_pretrained("t5-small")178 179input_ids = tokenizer(180    "Studies have been shown that owning a dog is good for you", return_tensors="pt"181).input_ids  # Batch size 1182decoder_input_ids = tokenizer("Studies show that", return_tensors="pt").input_ids  # Batch size 1183 184# forward pass185outputs = model(input_ids=input_ids, decoder_input_ids=decoder_input_ids)186last_hidden_states = outputs.last_hidden_state187```188 189See the [Hugging Face T5](https://huggingface.co/docs/transformers/model_doc/t5#transformers.T5Model) docs and a [Colab Notebook](https://colab.research.google.com/github/google-research/text-to-text-transfer-transformer/blob/main/notebooks/t5-trivia.ipynb) created by the model developers for more examples.190</details>191 192