使用导出/重新加载模型进行 fastai 错误预测:“输入类型和权重类型应该相同”

Max*_*wer 6 python deep-learning pytorch fast-ai

每当我导出 fastai 模型并重新加载它时,当我尝试使用重新加载的模型在新测试集上生成预测时,我会收到此错误(或非常相似的错误):

RuntimeError: Input type (torch.cuda.FloatTensor) and weight type (torch.cuda.HalfTensor) should be the same
Run Code Online (Sandbox Code Playgroud)

下面的最小可重复代码示例,您只需要将FILES_DIR变量更新为 MNIST 数据存放在系统上的位置:

from fastai import *
from fastai.vision import *

# download data for reproduceable example
untar_data(URLs.MNIST_SAMPLE)
FILES_DIR = '/home/mepstein/.fastai/data/mnist_sample'  # this is where command above deposits the MNIST data for me


# Create FastAI databunch for model training
tfms = get_transforms()
tr_val_databunch = ImageDataBunch.from_folder(path=FILES_DIR,  # location of downloaded data shown in log of prev command
                                train = 'train',
                                valid_pct = 0.2,
                                ds_tfms = tfms).normalize()

# Create Model
conv_learner = cnn_learner(tr_val_databunch, 
                           models.resnet34, 
                           metrics=[error_rate]).to_fp16()

# Train Model
conv_learner.fit_one_cycle(4)

# Export Model
conv_learner.export()  # saves model as 'export.pkl' in path associated with the learner

# Reload Model and use it for inference on new hold-out set
reloaded_model = load_learner(path = FILES_DIR,
                              test = ImageList.from_folder(path = f'{FILES_DIR}/valid'))

preds = reloaded_model.get_preds(ds_type=DatasetType.Test)
Run Code Online (Sandbox Code Playgroud)

输出:

“运行时错误:输入类型(torch.cuda.FloatTensor)和权重类型(torch.cuda.HalfTensor)应该相同”

逐条执行代码语句,一切正常,直到最后一行pred = ...出现上面的火炬错误。

相关软件版本:

Python 3.7.3 fastai 1.0.57
火炬 1.2.0 火炬
视觉 0.4.0

Max*_*wer 3

所以这个问题的答案相对简单:

1)正如我的评论中所述,混合精度模式(设置conv_learner to_fp16())下的训练导致导出/重新加载模型出现错误

2) 要在混合精度模式下进行训练(比常规训练更快)并能够无错误地导出/重新加载模型,只需在导出之前将模型设置回默认精度即可。

...在代码中,只需更改上面的示例:

# Export Model
conv_learner.export()
Run Code Online (Sandbox Code Playgroud)

到:

# Export Model (after converting back to default precision for safe export/reload
conv_learner = conv_learner.to_fp32()
conv_learner.export()
Run Code Online (Sandbox Code Playgroud)

...现在上面的完整(可重现)代码示例运行没有错误,包括模型重新加载后的预测。