paddle-bot[bot]
commented
2 years ago 您好，我们已经收到了您的问题，会安排技术人员尽快解答您的问题，请耐心等待。请您再次检查是否提供了清晰的问题描述、复现代码、环境&版本、报错信息等。同时，您也可以通过查看官网API文档、常见问题、历史Issue、AI社区来寻求解答。祝您生活愉快～
Hi! We've received your issue and please be patient to get responded. We will arrange technicians to answer your questions as soon as possible. Please make sure that you have posted enough message to demo your request. You may also check out the API，FAQ，Github Issue and AI community to get the answer.Have a nice day!
zhiboniu
kevinluoc
rainyfly
bug描述 Describe the Bug
我们想在visualdl中输出模型结构，但一直报错，直接使用官方提供的示例也一样出错:"Failed to save model graph, error: ''"。
以下代码中有写下相关注释，即时尝试一个线性回归的简单模型，也一样出错
import numpy as np import paddle import paddle.nn as nn import paddle.optimizer as opt from paddle.static import InputSpec from visualdl import LogWriter
BATCH_SIZE = 16 BATCH_NUM = 4 EPOCH_NUM = 4 IMAGE_SIZE = 784 CLASS_NUM = 10
writer = LogWriter(logdir="./log/graph_test02/")
define a random dataset
class RandomDataset(paddle.io.Dataset): def init(self, num_samples): self.num_samples = num_samples
class LinearNet(nn.Layer): def init(self): super(LinearNet, self).init() self._linear = nn.Linear(IMAGE_SIZE, CLASS_NUM)
def train(layer, loader, loss_fn, opt): for epoch_id in range(EPOCH_NUM): for batch_id, (image, label) in enumerate(loader()): out = layer(image) loss = loss_fn(out, label) loss.backward() opt.step() opt.clear_grad() print("Epoch {} batch {}: loss = {}".format(epoch_id, batch_id, np.mean(loss.numpy())))
create network
layer = LinearNet() loss_fn = nn.CrossEntropyLoss() adam = opt.Adam(learning_rate=0.001, parameters=layer.parameters())
create data loader
dataset = RandomDataset(BATCH_NUM * BATCH_SIZE) loader = paddle.io.DataLoader(dataset, batch_size=BATCH_SIZE, shuffle=True, drop_last=True, num_workers=2)
train
train(layer, loader, loss_fn, adam)
save
path = "example.dy_model/linear" paddle.jit.save( layer=layer, path=path, input_spec=[InputSpec(shape=[None, 784], dtype='float32')])
writer.add_graph(model=layer, input_spec=[paddle.static.InputSpec(shape=[None, 784], dtype='float32')], verbose=False) writer.close()
报错同样发生在这里
_writer.add_graph(model=layer, input_spec=[paddle.static.InputSpec(shape=[None, 784], dtype='float32')], verbose=False)_
报错时，运行中的代码如下： import paddle from paddle import Tensor import paddle.fluid as fluid from typing import * import numpy as np def forward(self, x): return paddle.jit.dy2static.convert_call(self._linear)(x)
此內容为paddle运行过程中产生，此外报错并非发生在训练过程中，而是单纯发生在add_graph的方法上。
其他补充信息 Additional Supplementary Information
No response