[Typing] 修复 dynamic_decode 以及示例中的类型标注错误

[megemini]

PR Category

User Experience

PR Types

Bug fixes

Description

修复#65397 中所有检查出的错误。

以及 dynamic_decode 标注错误。

关联 PR #65008 #65397

@SigureMo

[paddle-bot]

你的PR提交成功，感谢你对开源项目的贡献!
请关注后续CI自动化测试结果，详情请参考Paddle-CI手册。
Your PR has been submitted. Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

[SigureMo]

为什么是这个 DatasetBase？按照示例代码应该是 base 下的？

[megemini]

参考这段示例

import paddle
paddle.enable_static()
dataset = paddle.distributed.InMemoryDataset()
slots = ["slot1", "slot2", "slot3", "slot4"]
slots_vars = []
for slot in slots:
    var = paddle.static.data(
        name=slot, shape=[None, 1], dtype="int64", lod_level=1)
    slots_vars.append(var)
dataset.init(
    batch_size=1,
    thread_num=2,
    input_type=1,
    pipe_command="cat",
    use_var=slots_vars)
filelist = ["a.txt", "b.txt"]
dataset.set_filelist(filelist)
dataset.load_into_memory()
dataset.global_shuffle()
exe = paddle.static.Executor(paddle.CPUPlace())
startup_program = paddle.static.Program()
main_program = paddle.static.Program()
exe.run(startup_program)
exe.train_from_dataset(main_program, dataset)
dataset.release_memory()

base 下面的

Paddle/python/paddle/base/dataset.py

Lines 370 to 396 in 3ad5f26

	class InMemoryDataset(DatasetBase):
	"""
	InMemoryDataset, it will load data into memory
	and shuffle data before training.
	This class should be created by DatasetFactory
	Example:
	dataset = paddle.base.DatasetFactory().create_dataset("InMemoryDataset")
	"""
	@deprecated(since="2.0.0", update_to="paddle.distributed.InMemoryDataset")
	def __init__(self):
	"""Init."""
	super().__init__()
	self.proto_desc.name = "MultiSlotInMemoryDataFeed"
	self.fleet_send_batch_size = None
	self.is_user_set_queue_num = False
	self.queue_num = None
	self.parse_ins_id = False
	self.parse_content = False
	self.parse_logkey = False
	self.merge_by_sid = True
	self.enable_pv_merge = False
	self.merge_by_lineid = False
	self.fleet_send_sleep_seconds = None
	self.trainer_num = -1
	self.pass_id = 0

已经遗弃了，还是说 train_from_dataset 的 dataset 可以支持两个 Dataset ？

[SigureMo]

从执行器的示例代码肯定是支持 paddle.base.dataset.Dataset 的，那就都写吧，import 的时候 as 加个别名避免冲突

离谱的设计，再这样搞直接给他们搞 Any

[SigureMo]

LGTMeow