Switch Transformers is a Mixture of Experts (MoE) model trained on Masked Language Modeling (MLM) task. The model architecture is similar to the classic T5, but with the Feed Forward layers replaced by the Sparse MLP layers containing "experts" MLP. According to the original paper the model enables faster training (scaling properties) while being better than T5 on fine-tuned tasks. As mentioned in the first few lines of the abstract :
we advance the current scale of language models by pre-training up to trillion parameter models on the “Colossal Clean Crawled Corpus”, and achieve a 4x speedup over the T5-XXL model.
Disclaimer: Content from this model card has been written by the Hugging Face team, and parts of it were copy pasted from the original paper.
Note that these checkpoints has been trained on Masked-Language Modeling (MLM) task. Therefore the checkpoints are not "ready-to-use" for downstream tasks. You may want to check FLAN-T5 for running fine-tuned weights or fine-tune your own MoE following this notebook
Find below some example scripts on how to use the model in transformers:
from transformers import AutoTokenizer, SwitchTransformersForConditionalGeneration
tokenizer = AutoTokenizer.from_pretrained("google/switch-base-16")
model = SwitchTransformersForConditionalGeneration.from_pretrained("google/switch-base-16")
input_text = "A <extra_id_0> walks into a bar a orders a <extra_id_1> with <extra_id_2> pinch of <extra_id_3>."
input_ids = tokenizer(input_text, return_tensors="pt").input_ids
outputs = model.generate(input_ids)
print(tokenizer.decode(outputs[0]))
>>> <pad> <extra_id_0> man<extra_id_1> beer<extra_id_2> a<extra_id_3> salt<extra_id_4>.</s>
# pip install accelerate
from transformers import AutoTokenizer, SwitchTransformersForConditionalGeneration
tokenizer = AutoTokenizer.from_pretrained("google/switch-base-16")
model = SwitchTransformersForConditionalGeneration.from_pretrained("google/switch-base-16", device_map="auto")
input_text = "A <extra_id_0> walks into a bar a orders a <extra_id_1> with <extra_id_2> pinch of <extra_id_3>."
input_ids = tokenizer(input_text, return_tensors="pt").input_ids.to(0)
outputs = model.generate(input_ids)
print(tokenizer.decode(outputs[0]))
>>> <pad> <extra_id_0> man<extra_id_1> beer<extra_id_2> a<extra_id_3> salt<extra_id_4>.</s>
# pip install accelerate
from transformers import AutoTokenizer, SwitchTransformersForConditionalGeneration
tokenizer = AutoTokenizer.from_pretrained("google/switch-base-16")
model = SwitchTransformersForConditionalGeneration.from_pretrained("google/switch-base-16", device_map="auto", torch_dtype=torch.float16)
input_text = "A <extra_id_0> walks into a bar a orders a <extra_id_1> with <extra_id_2> pinch of <extra_id_3>."
input_ids = tokenizer(input_text, return_tensors="pt").input_ids.to(0)
outputs = model.generate(input_ids)
print(tokenizer.decode(outputs[0]))
>>> <pad> <extra_id_0> man<extra_id_1> beer<extra_id_2> a<extra_id_3> salt<extra_id_4>.</s>
# pip install bitsandbytes accelerate
from transformers import AutoTokenizer, SwitchTransformersForConditionalGeneration
tokenizer = AutoTokenizer.from_pretrained("google/switch-base-16")
model = SwitchTransformersForConditionalGeneration.from_pretrained("google/switch-base-16", device_map="auto")
input_text = "A <extra_id_0> walks into a bar a orders a <extra_id_1> with <extra_id_2> pinch of <extra_id_3>."
input_ids = tokenizer(input_text, return_tensors="pt").input_ids.to(0)
outputs = model.generate(input_ids)
print(tokenizer.decode(outputs[0]))
>>> <pad> <extra_id_0> man<extra_id_1> beer<extra_id_2> a<extra_id_3> salt<extra_id_4>.</s>
See the research paper for further details.
More information needed.
More information needed.
More information needed.
More information needed.
More information needed.
The model was trained on a Masked Language Modeling task, on Colossal Clean Crawled Corpus (C4) dataset, following the same procedure as T5.
According to the model card from the original paper the model has been trained on TPU v3 or TPU v4 pods, using t5x codebase together with jax.
The authors evaluated the model on various tasks and compared the results against T5. See the table below for some quantitative evaluation:
For full details, please check the research paper.
For full results for Switch Transformers, see the research paper, Table 5.
Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).
BibTeX:
@misc{https://doi.org/10.48550/arxiv.2101.03961,
doi = {10.48550/ARXIV.2101.03961},
url = {https://arxiv.org/abs/2101.03961},
author = {Fedus, William and Zoph, Barret and Shazeer, Noam},
keywords = {Machine Learning (cs.LG), Artificial Intelligence (cs.AI), FOS: Computer and information sciences, FOS: Computer and information sciences},
title = {Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity},
publisher = {arXiv},
year = {2021},
copyright = {arXiv.org perpetual, non-exclusive license}
}