CPU inference

With some optimizations, it is possible to efficiently run large model inference on a CPU. One of these optimization techniques involves compiling the PyTorch code into an intermediate format for high-performance environments like C++. The other technique fuses multiple operations into one kernel to reduce the overhead of running each operation separately.

You’ll learn how to use BetterTransformer for faster inference, and how to convert your PyTorch code to TorchScript. If you’re using an Intel CPU, you can also use graph optimizations from Intel Extension for PyTorch to boost inference speed even more. Finally, learn how to use 🤗 Optimum to accelerate inference with ONNX Runtime or OpenVINO (if you’re using an Intel CPU).

BetterTransformer

BetterTransformer accelerates inference with its fastpath (native PyTorch specialized implementation of Transformer functions) execution. The two optimizations in the fastpath execution are:

1. fusion, which combines multiple sequential operations into a single “kernel” to reduce the number of computation steps
2. skipping the inherent sparsity of padding tokens to avoid unnecessary computation with nested tensors

BetterTransformer also converts all attention operations to use the more memory-efficient scaled dot product attention.

Before you start, make sure you have 🤗 Optimum installed.

Enable BetterTransformer with the PreTrainedModel.to_bettertransformer() method:

from transformers import AutoModelForCausalLM

model = AutoModelForCausalLM.from_pretrained("bigcode/starcoder")
model.to_bettertransformer()

TorchScript

TorchScript is an intermediate PyTorch model representation that can be run in production environments where performance is important. You can train a model in PyTorch and then export it to TorchScript to free the model from Python performance constraints. PyTorch traces a model to return a ScriptFunction that is optimized with just-in-time compilation (JIT). Compared to the default eager mode, JIT mode in PyTorch typically yields better performance for inference using optimization techniques like operator fusion.

For a gentle introduction to TorchScript, see the Introduction to PyTorch TorchScript tutorial.

With the Trainer class, you can enable JIT mode for CPU inference by setting the --jit_mode_eval flag:

python run_qa.py \
--model_name_or_path csarron/bert-base-uncased-squad-v1 \
--dataset_name squad \
--do_eval \
--max_seq_length 384 \
--doc_stride 128 \
--output_dir /tmp/ \
--no_cuda \
--jit_mode_eval

IPEX graph optimization

Intel® Extension for PyTorch (IPEX) provides further optimizations in JIT mode for Intel CPUs, and we recommend combining it with TorchScript for even faster performance. The IPEX graph optimization fuses operations like Multi-head attention, Concat Linear, Linear + Add, Linear + Gelu, Add + LayerNorm, and more.

To take advantage of these graph optimizations, make sure you have IPEX installed:

pip install intel_extension_for_pytorch

Set the --use_ipex and --jit_mode_eval flags in the Trainer class to enable JIT mode with the graph optimizations:

python run_qa.py \
--model_name_or_path csarron/bert-base-uncased-squad-v1 \
--dataset_name squad \
--do_eval \
--max_seq_length 384 \
--doc_stride 128 \
--output_dir /tmp/ \
--no_cuda \
--use_ipex \
--jit_mode_eval

🤗 Optimum

ONNX Runtime (ORT) is a model accelerator that runs inference on CPUs by default. ORT is supported by 🤗 Optimum which can be used in 🤗 Transformers, without making too many changes to your code. You only need to replace the 🤗 Transformers AutoClass with its equivalent ORTModel for the task you’re solving, and load a checkpoint in the ONNX format.

For example, if you’re running inference on a question answering task, load the optimum/roberta-base-squad2 checkpoint which contains a model.onnx file:

from transformers import AutoTokenizer, pipeline
from optimum.onnxruntime import ORTModelForQuestionAnswering

model = ORTModelForQuestionAnswering.from_pretrained("optimum/roberta-base-squad2")
tokenizer = AutoTokenizer.from_pretrained("deepset/roberta-base-squad2")

onnx_qa = pipeline("question-answering", model=model, tokenizer=tokenizer)

question = "What's my name?"
context = "My name is Philipp and I live in Nuremberg."
pred = onnx_qa(question, context)

If you have an Intel CPU, take a look at 🤗 Optimum Intel which supports a variety of compression techniques (quantization, pruning, knowledge distillation) and tools for converting models to the OpenVINO format for higher performance inference.