fastT5
mt5-M2M-comparison
| fastT5 | mt5-M2M-comparison | |
|---|---|---|
| 5 | 1 | |
| 540 | 13 | |
| - | - | |
| 0.0 | 3.8 | |
| about 1 year ago | almost 3 years ago | |
| Python | Jupyter Notebook | |
| Apache License 2.0 | - |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
fastT5
-
Speeding up T5
I've tried https://github.com/Ki6an/fastT5 but it works with CPU only.
-
Convert Pegasus model to ONNX
I am working on a project where I fine-tuned a Pegasus model on the Reddit dataset. Now, I need to convert the fine-tuned model to ONNX for the deployment stage. I have followed this guide from Huggingface to convert to the ONNX model for unsupported architects. I got it done but the ONNX model can't generate text. Turned out that Pegasus is an encoder-decoder model and most guides are for either encoder-model (e.g. BERT) or decoder-model (e.g. GPT2). I found the only example of converting an encoder-decoder model to ONNX from here https://github.com/Ki6an/fastT5.
-
[P] What we learned by making T5-large 2X faster than Pytorch (and any autoregressive transformer)
Microsoft Onnx Runtime T5 export tool / FastT5: to support caching, it exports 2 times the decoder part, one with cache, and one without (for the first generated token). So the memory footprint is doubled, which makes the solution difficult to use for these large transformer models.
-
Conceptually, what are the "Past key values" in the T5 Decoder?
Here is the fastT5 model code for reference code:https://github.com/Ki6an/fastT5/blob/master/fastT5/onnx_models.py
-
[P] boost T5 models speed up to 5x & reduce the model size by 3x using fastT5.
for more information on the project refer to the repository here.
mt5-M2M-comparison
-
[D] Comparing M2M to mT5 in low resource translation (10k dataset Yoruba - English)
I found no clear comparison nor a clear guide on how to fine tune both of the models on the translation task, so I decided to write it myself. (code: https://github.com/maroxtn/mt5-M2M-comparison)
What are some alternatives?
Questgen.ai - Question generation using state-of-the-art Natural Language Processing algorithms
keytotext - Keywords to Sentences
json-translate - Translate json files with DeepL or AWS
OpenNMT-Tutorial - Neural Machine Translation (NMT) tutorial. Data preprocessing, model training, evaluation, and deployment.
frame-semantic-transformer - Frame Semantic Parser based on T5 and FrameNet
100DaysOfML - 100 Days Of Machine Learning. New Content in every 1-2 day and projects every week. The massive 100DaysOfML in building
OpenSeeFace - Robust realtime face and facial landmark tracking on CPU with Unity integration
fake-news - Building a fake news detector from initial ideation to model deployment
FasterTransformer - Transformer related optimization, including BERT, GPT
question_generation - Neural question generation using transformers
sparktorch - Train and run Pytorch models on Apache Spark.
transformer-deploy - Efficient, scalable and enterprise-grade CPU/GPU inference server for 🤗 Hugging Face transformer models 🚀