rewrite the text classification demo. #83

lcy-seso · 2017-06-10T06:26:55Z

rewrite the text classification example.

llxxxll · 2017-06-12T04:08:20Z

image_classification/train.py

@@ -51,10 +37,10 @@ def main():
        learning_rate_schedule="discexp", )

    train_reader = paddle.batch(
-        paddle.reader.shuffle(reader.test_reader("train.list"), buf_size=1000),
+        paddle.reader.shuffle(reader.train_reader("train.list"), buf_size=1000),


train.list是文本数据集吗？没有在目录下找到

这处修改用来修改图像分类例子的bug，目前每个例子读取数据的方式确实不统一。后续提PR修改图像分类的例子。

llxxxll · 2017-06-12T04:08:44Z

image_classification/train.py

        batch_size=BATCH_SIZE)
    test_reader = paddle.batch(
-        reader.train_reader("test.list"), batch_size=BATCH_SIZE)
+        reader.test_reader("test.list"), batch_size=BATCH_SIZE)


test.list是文本数据集吗？没有在目录下找到

这处修改用来修改图像分类例子的bug，目前每个例子读取数据的方式确实不统一。后续提PR修改图像分类的例子。

llxxxll · 2017-06-12T04:15:37Z

text_classification/index.html

+├── run.sh              # 运行此脚本，可以以默认参数直接开始训练任务
+├── train.py            # 训练任务脚本
+└── utils.py            # 定义通用的函数，例如：打印日志、解析命令行参数、构建字典、加载字典等
+```


缺少一个快速开始。其中包含上面的目录说明，及一个训练说明（训练过程输出日志里都是什么意思）、和一个预测说明（可以是一段code，参考：https://www.oschina.net/p/jieba/?fromerr=btIKdxHH -功能 1)分词的代码及output、当然也可以是一个gif图）

接下来应该是之前写在末尾的『修改参数说明』，这样可以在新手运行出一个结果的前提下，来对比修改不同参数得到的结果不同。于是顺势引出CNN&DNN该如何选择

其实余下部分都是在解释这些『参数』的含义，顺序也应如此。

llxxxll · 2017-06-12T04:18:58Z

text_classification/index.html

-    cost = paddle.layer.classification_cost(input=output, label=lbl)
-
-    return cost, output, lbl
+```


如果是面向初学者，数据格式的自定义要比其他信息重要得多，因为这可能是唯一"需要考虑"的事，所以优先级需要提到快速开始后面。而详解的这个位置可以对应加一个锚点链接。

llxxxll · 2017-06-13T04:32:17Z

text_classification/run.sh

+#!/bin/sh
+
+python train.py \
+--nn_type="dnn" \


这里的train的方式怎么又改成shell传参了，按照约定都应该写到train.py里？

shell 里面为 train.py 指定的参数，直接运行shell 即可，否则需要在命令行敲长串的参数。

llxxxll · 2017-06-13T04:37:27Z

text_classification/train.py

+
+
+def train(topology,
+          train_data_dir=None,


lm_rnn.py 里有个run_type=GRU #'or LSTM'，在这里是否可以增加一个方法的选择参数比如DNN OR CNN

已有此参数，nn_type 用来指定选择何种模型。

lcy-seso

follow comments.

lcy-seso · 2017-06-13T10:07:51Z

image_classification/train.py

        batch_size=BATCH_SIZE)
    test_reader = paddle.batch(
-        reader.train_reader("test.list"), batch_size=BATCH_SIZE)
+        reader.test_reader("test.list"), batch_size=BATCH_SIZE)


这处修改用来修改图像分类例子的bug，目前每个例子读取数据的方式确实不统一。后续提PR修改图像分类的例子。

lcy-seso · 2017-06-13T10:07:54Z

image_classification/train.py

@@ -51,10 +37,10 @@ def main():
        learning_rate_schedule="discexp", )

    train_reader = paddle.batch(
-        paddle.reader.shuffle(reader.test_reader("train.list"), buf_size=1000),
+        paddle.reader.shuffle(reader.train_reader("train.list"), buf_size=1000),


这处修改用来修改图像分类例子的bug，目前每个例子读取数据的方式确实不统一。后续提PR修改图像分类的例子。

luotao1 · 2017-06-13T10:22:34Z

Readme的目录顺序需要调整么？把模型介绍、模型详解放到最开头的地方？

lcy-seso requested a review from llxxxll June 10, 2017 06:26

lcy-seso force-pushed the rewrite_text_classification branch 21 times, most recently from eeccb28 to f884a61 Compare June 12, 2017 07:59

llxxxll reviewed Jun 12, 2017

View reviewed changes

llxxxll reviewed Jun 13, 2017

View reviewed changes

lcy-seso force-pushed the rewrite_text_classification branch from f884a61 to b1231a5 Compare June 13, 2017 09:07

rewrite the text classification demo.

501ce21

lcy-seso force-pushed the rewrite_text_classification branch from b1231a5 to 501ce21 Compare June 13, 2017 09:13

lcy-seso commented Jun 13, 2017

View reviewed changes

lcy-seso force-pushed the rewrite_text_classification branch from 6068a69 to a0529eb Compare June 13, 2017 10:18

lcy-seso force-pushed the rewrite_text_classification branch 4 times, most recently from 74980fe to ba69ba5 Compare June 13, 2017 11:07

update readme.

136a60d

lcy-seso force-pushed the rewrite_text_classification branch from ba69ba5 to 136a60d Compare June 13, 2017 11:10

llxxxll approved these changes Jun 14, 2017

View reviewed changes

lcy-seso merged commit f27154e into PaddlePaddle:develop Jun 14, 2017

lcy-seso deleted the rewrite_text_classification branch June 16, 2017 10:29

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

rewrite the text classification demo. #83

rewrite the text classification demo. #83

lcy-seso commented Jun 10, 2017

llxxxll Jun 12, 2017

lcy-seso Jun 13, 2017

llxxxll Jun 12, 2017

lcy-seso Jun 13, 2017

llxxxll Jun 12, 2017

llxxxll Jun 12, 2017

llxxxll Jun 12, 2017

llxxxll Jun 12, 2017

llxxxll Jun 13, 2017

lcy-seso Jun 13, 2017

llxxxll Jun 13, 2017

lcy-seso Jun 13, 2017

lcy-seso left a comment

lcy-seso Jun 13, 2017

lcy-seso Jun 13, 2017

luotao1 commented Jun 13, 2017

rewrite the text classification demo. #83

rewrite the text classification demo. #83

Conversation

lcy-seso commented Jun 10, 2017

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

lcy-seso left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

luotao1 commented Jun 13, 2017