Commit 4e4f27f5 by xiaotong

create XTrainer

parent feb29d20
......@@ -26,8 +26,6 @@
#ifndef __XLINK_H__
#define __XLINK_H__
#include "XGlobal.h"
namespace nts{ // namespace nts(NiuTrans.Tensor)
/* cross reference */
......
/* NiuTrans.Tensor - an open-source tensor library
* Copyright (C) 2021
* Natural Language Processing Lab, Northeastern University
* and
* NiuTrans Research
* All rights reserved.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/*
*
* $Created by: XIAO Tong (xiaotong@mail.neu.edu.cn) 2021-02-23
*
*/
#include "XTrainer.h"
/* the nts (NiuTrans.Tensor) namespace */
namespace nts {
/* constructor */
XTrainer::XTrainer()
{
}
/* de-constructor */
XTrainer::~XTrainer()
{
}
} /* end of the nts (NiuTrans.Tensor) namespace */
\ No newline at end of file
/* NiuTrans.Tensor - an open-source tensor library
* Copyright (C) 2021
* Natural Language Processing Lab, Northeastern University
* and
* NiuTrans Research
* All rights reserved.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/*
* This class organizes the training process of neural models, e.g., nmt and lm models
* Distributed training is supported.
*
* $Created by: XIAO Tong (xiaotong@mail.neu.edu.cn) 2021-02-23
* I start coding in 2021 after one year since I typed last C code.
* BUT i was a GOOD tex writter in 2020 :)
*/
#ifndef __XTRAINER_H__
#define __XTRAINER_H__
#include "../network/XNet.h"
#include "../tensor/XQueue.h"
namespace nts { // namespace nts(NiuTrans.Tensor)
/*
Training of neural networks with gradient methods. Here we suppose that we
are training NLP models. The routine could be:
1). initialize all we need
2). data preparation
3). loop until convergence
a). read a batch of samples from the input file
b). reset the worker
c). forward computation with the input
d). backward computation with respect to the loss
e). collect the gradient (neccessary when several workers are available)
f). update the model (on the server end)
g). distribute the new model to each worker
Here a worker processes a batch of samples one time, and works with
other workers independently. The server is the origanizer. It distriute
the job to the workers and maintain the model.
*/
class XTrainer
{
private:
public:
/* constructor */
XTrainer();
/* de-constructor */
~XTrainer();
};
}
#endif // __XTRAINER_H__
\ No newline at end of file
Markdown 格式
0%
您添加了 0 到此讨论。请谨慎行事。
请先完成此评论的编辑!
注册 或者 后发表评论