Training-aware Quantization of image classification model - quick start¶

This tutorial shows how to do training-aware quantization using API in PaddleSlim. We use MobileNetV1 to train image classification model as example. The tutorial contains follow sections:

Necessary imports
Model architecture
Train normal model
Quantization
Train model after quantization
Save model after quantization

1. Necessary imports¶

PaddleSlim depends on Paddle1.7. Please make true that you have installed Paddle correctly. Then do the necessary imports:

import paddle
import paddle.fluid as fluid
import paddleslim as slim
import numpy as np

2. Model architecture¶

The section constructs a classification model, which use MobileNetV1 and MNIST dataset. The model‘s input size is [1, 28, 28] and output size is 10. In order to show tutorial conveniently, we pre-defined a method to get image classification model in paddleslim.models.

note: The APIs in paddleslim.models are not formal inferface in PaddleSlim. They are defined to simplify the tutorial such as the definition of model structure and the construction of Program.

exe, train_program, val_program, inputs, outputs = \
    slim.models.image_classification("MobileNet", [1, 28, 28], 10, use_gpu=True)

3. Train normal model¶

The section shows how to define model inputs, train and test model. The reason for training the normal image classification model first is that the quantization model‘s training process is performed on the well-trained model. We add quantization and dequantization operators in well-trained model and finetune using smaller learning rate.

3.1 input data definition¶

To speed up training process, we select MNIST dataset to train image classification model. The API paddle.dataset.mnist in Paddle framework contains downloading and reading the images in dataset.

import paddle.dataset.mnist as reader
train_reader = paddle.batch(
        reader.train(), batch_size=128, drop_last=True)
test_reader = paddle.batch(
        reader.train(), batch_size=128, drop_last=True)
train_feeder = fluid.DataFeeder(inputs, fluid.CPUPlace())

3.2 training model and testing¶

Define functions to train and test model. We only need call the functions when formal model training and quantization model training. The function does one epoch training because that MNIST dataset is small and top1 accuracy will reach 95% after one epoch.

def train(prog):
    iter = 0
    for data in train_reader():
        acc1, acc5, loss = exe.run(prog, feed=train_feeder.feed(data), fetch_list=outputs)
        if iter % 100 == 0:
            print('train iter={}, top1={}, top5={}, loss={}'.format(iter, acc1.mean(), acc5.mean(), loss.mean()))
        iter += 1

def test(prog):
    iter = 0
    res = [[], []]
    for data in train_reader():
        acc1, acc5, loss = exe.run(prog, feed=train_feeder.feed(data), fetch_list=outputs)
        if iter % 100 == 0:
            print('test iter={}, top1={}, top5={}, loss={}'.format(iter, acc1.mean(), acc5.mean(), loss.mean()))
        res[0].append(acc1.mean())
        res[1].append(acc5.mean())
        iter += 1
    print('final test result top1={}, top5={}'.format(np.array(res[0]).mean(), np.array(res[1]).mean()))

Call train function to train normal classification model. train_program is defined in 2. Model architecture.

train(train_program)

Call test function to test normal classification model. val_program is defined in 2. Model architecture.

test(val_program)

4. Quantization¶

We call quant_aware API to add quantization and dequantization operators in train_program and val_program according to default configuration.

quant_program = slim.quant.quant_aware(train_program, exe.place, for_test=False)
val_quant_program = slim.quant.quant_aware(val_program, exe.place, for_test=True)

5. Train model after quantization¶

Finetune the model after quantization. Test model after one epoch training.

train(quant_program)

Test model after quantization. The top1 and top5 accuracy are close to result in 3.2 training model and testing. We preform the training-aware quantization without loss on this image classification model.

test(val_quant_program)

6. Save model after quantization¶

The model in 4. Quantization after calling slim.quant.quant_aware API is only suitable to train. To get the inference model, we should use slim.quant.convert API to change model architecture and use fluid.io.save_inference_model to save model. float_prog‘s parameters are float32 dtype but in int8‘s range which can be used in fluid or paddle-lite. paddle-lite will change the parameters‘ dtype from float32 to int8 first when loading the inference model. int8_prog‘s parameters are int8 dtype and we can get model size after quantization by saving it. int8_prog cannot be used in fluid or paddle-lite.

float_prog, int8_prog = slim.quant.convert(val_quant_program, exe.place, save_int8=True)
target_vars = [float_prog.global_block().var(name) for name in outputs]
fluid.io.save_inference_model(dirname='./inference_model/float',
        feeded_var_names=[var.name for var in inputs],
        target_vars=target_vars,
        executor=exe,
        main_program=float_prog)
fluid.io.save_inference_model(dirname='./inference_model/int8',
        feeded_var_names=[var.name for var in inputs],
        target_vars=target_vars,
        executor=exe,
        main_program=int8_prog)