-
Notifications
You must be signed in to change notification settings - Fork 1.3k
PR of chapter fit_a_line. #28
Changes from all commits
File filter
Filter by extension
Conversations
Jump to
Diff view
Diff view
There are no files selected for viewing
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,5 @@ | ||
data/housing* | ||
data/*.list | ||
*.pyc | ||
data/*.pyc | ||
output |
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -1 +1,172 @@ | ||
TODO: Basing on https://github.com/PaddlePaddle/Paddle/blob/develop/doc/getstarted/basic_usage/index_cn.rst | ||
# 线性回归 | ||
让我们从经典的线性回归(Linear Regression \[[1](#参考文献)\])模型开始这份教程。在这一章里,你将使用真实的数据集建立起一个房价预测模型,并且了解到机器学习中的若干重要概念。 | ||
|
||
## 背景介绍 | ||
给定一个大小为$n$的数据集 ${\{y_{i}, x_{i1}, ..., x_{id}\}}_{i=1}^{n}$,其中$x_{i1}, \ldots, x_{id}$是$d$个属性上的取值,$y_i$是待预测的目标。线性回归模型假设目标$y_i$可以被属性间的线性组合描述,即 | ||
|
||
$$y_i = \omega_1x_{i1} + \omega_2x_{i2} + \ldots + \omega_dx_{id} + b, i=1,\ldots,n$$ | ||
|
||
例如,在我们将要建模的房价预测问题里,$x_{ij}$是描述房子$i$的各种属性(比如房间的个数、周围学校和医院的个数、交通状况等),而 $y_i$是房屋的价格。 | ||
|
||
初看起来,这个假设实在过于简单了,变量间的真实关系很难是线性的。但由于线性回归模型有形式简单和易于建模分析的优点,它在实际问题中得到了大量的应用。很多经典的统计学习、机器学习书籍\[[2,3,4](#参考文献)\]也选择对线性模型独立成章重点讲解。 | ||
|
||
## 效果展示 | ||
我们使用从[UCI Housing Data Set](https://archive.ics.uci.edu/ml/datasets/Housing)获得的波士顿房价数据集进行模型的训练和预测。下面的散点图展示了使用模型对部分房屋价格进行的预测。其中,横轴展示了该类房屋价格的中位数,纵轴为模型的预测结果,当二者值完全相等的时候就会落在虚线上。所以模型预测得越准确,则点离虚线越近。 | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. 其中,横轴展示了该类房屋价格的中位数,纵轴为模型的预测结果 |
||
<p align="center"> | ||
<img src = "image/predictions.png"><br/> | ||
图1. 预测值 V.S. 真实值 | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more.
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. 横纵轴需要标明单位,比如$million,还有前面说明图上表示每平米还是房屋出售值这么多钱 There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. |
||
</p> | ||
|
||
## 模型概览 | ||
在波士顿房价数据集中,和房屋相关的值共有14个:前13个用来描述房屋相关的各种信息,即模型中的 $x_i$;最后一个值为我们要预测的房屋价格的中位数,即模型中的 $y_i$。因此,我们的模型就可以表示成: | ||
|
||
$$\hat{Y} = \omega_1X_{1} + \omega_2X_{2} + \ldots + \omega_{13}X_{13} + b$$ | ||
|
||
$\hat{Y}$ 表示模型的预测结果,用来和真实值$Y$区分。模型要学习的参数即:$\omega_1, \ldots, \omega_{13}, b$。 | ||
|
||
建立模型后,我们需要给模型一个优化目标,使得学到的参数能够让预测值$\hat{Y}$尽可能地接近真实值$Y$。这里我们引入损失函数([Loss Function](https://en.wikipedia.org/wiki/Loss_function),或Cost Function)这个概念。 输入任意一个数据样本的目标值$y_{i}$和模型给出的预测值$\hat{y_{i}}$,损失函数输出一个非负的实值。这个实质通常用来反映模型误差的大小。 | ||
|
||
对于线性回归模型来讲,最常见的损失函数就是均方误差(Mean Squared Error, [MSE](https://en.wikipedia.org/wiki/Mean_squared_error))了,它的形式是: | ||
|
||
$$MSE=\frac{1}{n}\sum_{i=1}^{n}{(\hat{Y_i}-Y_i)}^2$$ | ||
|
||
即对于一个大小为$n$的测试集,$MSE$是$n$个数据预测结果误差平方的均值。 | ||
|
||
## 数据准备 | ||
执行以下命令来准备数据: | ||
```bash | ||
cd data && python prepare_data.py | ||
``` | ||
这段代码将从[UCI Housing Data Set](https://archive.ics.uci.edu/ml/datasets/Housing)下载数据并进行[预处理](#数据预处理),最后数据将被分为训练集和测试集。 | ||
|
||
这份数据集共506行,每行包含了波士顿郊区的一类房屋的相关信息及价格的中位数。其各维属性的意义如下: | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. 价格的中位数 |
||
|
||
| 属性名 | 解释 | 类型 | | ||
| ------| ------ | ------ | | ||
| CRIM | 该镇的人均犯罪率 | 连续值 | | ||
| ZN | 占地面积超过25,000平方呎的住宅用地比例 | 连续值 | | ||
| INDUS | 非零售商业用地比例 | 连续值 | | ||
| CHAS | 是否邻近 Charles River | 离散值,1=邻近;0=不邻近 | | ||
| NOX | 一氧化氮浓度 | 连续值 | | ||
| RM | 每栋房屋的平均客房数 | 连续值 | | ||
| AGE | 1940年之前建成的自用单位比例 | 连续值 | | ||
| DIS | 到波士顿5个就业中心的加权距离 | 连续值 | | ||
| RAD | 到径向公路的可达性指数 | 连续值 | | ||
| TAX | 全值财产税率 | 连续值 | | ||
| PTRATIO | 学生与教师的比例 | 连续值 | | ||
| B | 1000(BK - 0.63)^2,其中BK为黑人占比 | 连续值 | | ||
| LSTAT | 低收入人群占比 | 连续值 | | ||
| MEDV | 房屋价格的中位数 | 连续值 | | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. 房屋价格的中位数 |
||
|
||
### 数据预处理 | ||
#### 连续值与离散值 | ||
观察一下数据,我们的第一个发现是:所有的13维属性中,有12维的连续值和1维的离散值(CHAS)。离散值虽然也常使用类似0、1、2这样的数字表示,但是其含义与连续值是不同的,因为这里的差值没有实际意义。例如,我们用0、1、2来分别表示红色、绿色和蓝色的话,我们并不能因此说“蓝色和红色”比“绿色和红色”的距离更远。所以通常对一个有$d$个可能取值的离散属性,我们会将它们转为$d$个取值为0或1的二值属性。不过就这里而言,因为CHAS本身就是一个二值属性,就省去了这个麻烦。 | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. 为0或1的二值属性 |
||
|
||
#### 属性的归一化 | ||
另外一个稍加观察即可发现的事实是,各维属性的取值范围差别很大(如图2所示)。例如,属性B的取值范围是[0.32, 396.90],而属性NOX的取值范围是[0.3850, 0.8170]。这里就要用到一个常见的操作-归一化(normalization)了。归一化的目标是把各位属性的取值范围放缩到差不多的区间,例如[-0.5,0.5]。这里我们使用一种很常见的操作方法:减掉均值,然后除以原取值范围。 | ||
|
||
做归一化(或 [Feature scaling](https://en.wikipedia.org/wiki/Feature_scaling))至少有以下3个理由: | ||
- 过大或过小的数值范围会导致计算时的浮点上溢或下溢。 | ||
- 不同的数值范围会导致不同属性对模型的重要性不同(至少在训练的初始阶段如此),而这个隐含的假设常常是不合理的。这会对优化的过程造成困难,使训练时间大大的加长。 | ||
- 很多的机器学习技巧/模型(例如L1,L2正则项,向量空间模型-Vector Space Model)都基于这样的假设:所有的属性取值都差不多是以0为均值且取值范围相近的。 | ||
|
||
<p align="center"> | ||
<img src = "image/ranges.png"><br/> | ||
图2. 各维属性的取值范围 | ||
</p> | ||
|
||
#### 整理训练集与测试集 | ||
我们将数据集分割为两份:一份用于调整模型的参数,即进行模型的训练,模型在这份数据集上的误差被称为**训练误差**;另外一份被用来测试,模型在这份数据集上的误差被称为**测试误差**。我们训练模型的目的是为了通过从训练数据中找到规律来预测未知的新数据,所以测试误差是更能反映模型表现的指标。分割数据的比例要考虑到两个因素:更多的训练数据会降低参数估计的方差,从而得到更可信的模型;而更多的测试数据会降低测试误差的方差,从而得到更可信的测试误差。一种常见的分割比例为$8:2$,感兴趣的读者朋友们也可以尝试不同的设置来观察这两种误差的变化。 | ||
|
||
执行如下命令可以分割数据集,并将训练集和测试集的地址分别写入train.list 和 test.list两个文件中,供PaddlePaddle读取。 | ||
```python | ||
python prepare_data.py -r 0.8 #默认使用8:2的比例进行分割 | ||
``` | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. 70行开始的这段预处理代码不用贴, @Zrachel 你觉得呢 There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. 同意,不用贴了,直接说执行python x.py进行预处理吧 |
||
|
||
在更复杂的模型训练过程中,我们往往还会多使用一种数据集:验证集。因为复杂的模型中常常还有一些超参数([Hyperparameter](https://en.wikipedia.org/wiki/Hyperparameter_optimization))需要调节,所以我们会尝试多种超参数的组合来分别训练多个模型,然后对比它们在验证集上的表现选择相对最好的一组超参数,最后才使用这组参数下训练的模型在测试集上评估测试误差。由于本章训练的模型比较简单,我们暂且忽略掉这个过程。 | ||
|
||
### 提供数据给PaddlePaddle | ||
准备好数据之后,我们使用一个Python data provider来为PaddlePaddle的训练过程提供数据。一个 data provider 就是一个Python函数,它会被PaddlePaddle的训练过程调用。在这个例子里,只需要读取已经保存好的数据,然后一行一行地返回给PaddlePaddle的训练进程即可。 | ||
|
||
```python | ||
from paddle.trainer.PyDataProvider2 import * | ||
import numpy as np | ||
#定义数据的类型和维度 | ||
@provider(input_types=[dense_vector(13), dense_vector(1)]) | ||
def process(settings, input_file): | ||
data = np.load(input_file.strip()) | ||
for row in data: | ||
yield row[:-1].tolist(), row[-1:].tolist() | ||
|
||
``` | ||
|
||
## 模型配置说明 | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. 数据定义算法配置网络结构缺少小标题,请依次补充 There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. done |
||
|
||
### 数据定义 | ||
首先,通过 `define_py_data_sources2` 来配置PaddlePaddle从上面的`dataprovider.py`里读入训练数据和测试数据。 PaddlePaddle接受从命令行读入的配置信息,例如这里我们传入一个名为`is_predict`的变量来控制模型在训练和测试时的不同结构。 | ||
```python | ||
from paddle.trainer_config_helpers import * | ||
|
||
is_predict = get_config_arg('is_predict', bool, False) | ||
|
||
define_py_data_sources2( | ||
train_list='data/train.list', | ||
test_list='data/test.list', | ||
module='dataprovider', | ||
obj='process') | ||
|
||
``` | ||
|
||
### 算法配置 | ||
接着,指定模型优化算法的细节。由于线性回归模型比较简单,我们只要设置基本的`batch_size`即可,它指定每次更新参数的时候使用多少条数据计算梯度信息。 | ||
```python | ||
settings(batch_size=2) | ||
``` | ||
|
||
### 网络结构 | ||
最后,使用`fc_layer`和`LinearActivation`来表示线性回归的模型本身。 | ||
```python | ||
#输入数据,13维的房屋信息 | ||
x = data_layer(name='x', size=13) | ||
|
||
y_predict = fc_layer( | ||
input=x, | ||
param_attr=ParamAttr(name='w'), | ||
size=1, | ||
act=LinearActivation(), | ||
bias_attr=ParamAttr(name='b')) | ||
|
||
if not is_predict: #训练时,我们使用MSE,即regression_cost作为损失函数 | ||
y = data_layer(name='y', size=1) | ||
cost = regression_cost(input=y_predict, label=y) | ||
outputs(cost) #训练时输出MSE来监控损失的变化 | ||
else: #测试时,输出预测值 | ||
outputs(y_predict) | ||
``` | ||
|
||
## 训练模型 | ||
在对应代码的根目录下执行PaddlePaddle的命令行训练程序。这里指定模型配置文件为`trainer_config.py`,训练30轮,结果保存在`output`路径下。 | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. 可以直接执行训练脚本 There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. done |
||
```bash | ||
./train.sh | ||
``` | ||
|
||
## 应用模型 | ||
现在来看下如何使用已经训练好的模型进行预测。 | ||
```bash | ||
python predict.py | ||
``` | ||
这里默认使用`output/pass-00029`中保存的模型进行预测,并将数据中的房价与预测结果进行对比,结果保存在 `predictions.png`中。 | ||
如果你想使用别的模型或者其它的数据进行预测,只要传入新的路径即可: | ||
```bash | ||
python predict.py -m output/pass-00020 -t data/housing.test.npy | ||
``` | ||
|
||
## 总结 | ||
在这章里,我们借助波士顿房价这一数据集,介绍了线性回归模型的基本概念,以及如何使用PaddlePaddle实现训练和测试的过程。很多的模型和技巧都是从简单的线性回归模型演化而来,因此弄清楚线性模型的原理和局限非常重要。 | ||
|
||
|
||
## 参考文献 | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. 第一章很重要 There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. 第6行和第7行应该是三个### There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. 抱歉,这个文档还在改,没有完成。之前commit是为了check下公式的格式。我大概今晚会改一个比较完整的版本再ci,之前先不用review。 There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. 缺失总结和参考文献。 There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. done |
||
1. https://en.wikipedia.org/wiki/Linear_regression | ||
2. Friedman J, Hastie T, Tibshirani R. The elements of statistical learning[M]. Springer, Berlin: Springer series in statistics, 2001. | ||
3. Murphy K P. Machine learning: a probabilistic perspective[M]. MIT press, 2012. | ||
4. Bishop C M. Pattern recognition[J]. Machine Learning, 2006, 128. |
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,108 @@ | ||
# Copyright (c) 2016 PaddlePaddle Authors. 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. | ||
from collections import Counter | ||
from urllib2 import urlopen | ||
import argparse | ||
import os | ||
import random | ||
import logging | ||
|
||
import numpy as np | ||
|
||
logging.basicConfig(level=logging.INFO) | ||
data_url = 'https://archive.ics.uci.edu/ml/machine' \ | ||
'-learning-databases/housing/housing.data' | ||
raw_data = 'housing.data' | ||
train_data = 'housing.train.npy' | ||
test_data = 'housing.test.npy' | ||
feature_names = [ | ||
'CRIM', 'ZN', 'INDUS', 'CHAS', 'NOX', 'RM', 'AGE', 'DIS', 'RAD', 'TAX', | ||
'PTRATIO', 'B', 'LSTAT' | ||
] | ||
root_dir = os.path.abspath(os.pardir) | ||
|
||
|
||
def maybe_download(url, file_path): | ||
if not os.path.exists(file_path): | ||
logging.info('data doesn\'t exist on %s, download from [%s]' % | ||
(file_path, url)) | ||
resp = urlopen(url).read() | ||
with open(file_path, 'w') as f: | ||
f.write(resp) | ||
|
||
logging.info('got raw housing data') | ||
|
||
|
||
def save_list(): | ||
with open('train.list', 'w') as f: | ||
f.write('data/' + train_data + '\n') | ||
with open('test.list', 'w') as f: | ||
f.write('data/' + test_data + '\n') | ||
|
||
|
||
def feature_range(maximums, minimums): | ||
import matplotlib | ||
matplotlib.use('Agg') | ||
import matplotlib.pyplot as plt | ||
fig, ax = plt.subplots() | ||
feature_num = len(maximums) | ||
ax.bar(range(feature_num), maximums - minimums, color='r', align='center') | ||
ax.set_title('feature scale') | ||
plt.xticks(range(feature_num), feature_names) | ||
plt.xlim([-1, feature_num]) | ||
fig.set_figheight(6) | ||
fig.set_figwidth(10) | ||
fig.savefig('%s/image/ranges.png' % root_dir, dpi=48) | ||
plt.close(fig) | ||
|
||
|
||
def preprocess(file_path, feature_num=14, shuffle=False, ratio=0.8): | ||
data = np.fromfile(file_path, sep=' ') | ||
data = data.reshape(data.shape[0] / feature_num, feature_num) | ||
maximums, minimums, avgs = data.max(axis=0), data.min(axis=0), data.sum( | ||
axis=0) / data.shape[0] | ||
feature_range(maximums[:-1], minimums[:-1]) | ||
for i in xrange(feature_num - 1): | ||
data[:, i] = (data[:, i] - avgs[i]) / (maximums[i] - minimums[i]) | ||
if shuffle: | ||
np.random.shuffle(data) | ||
offset = int(data.shape[0] * ratio) | ||
np.save(train_data, data[:offset]) | ||
logging.info('saved training data to %s' % train_data) | ||
np.save(test_data, data[offset:]) | ||
logging.info('saved test data to %s' % test_data) | ||
save_list() | ||
|
||
|
||
if __name__ == '__main__': | ||
parser = argparse.ArgumentParser( | ||
description='download boston housing price data set and preprocess the data(normalization and split dataset)' | ||
) | ||
parser.add_argument( | ||
'-r', | ||
'--ratio', | ||
dest='ratio', | ||
default='0.8', | ||
help='ratio of data used for training') | ||
parser.add_argument( | ||
'-s', | ||
'--shuffle', | ||
dest='shuffle', | ||
default='0', | ||
choices={'1', '0'}, | ||
help='shuffle the data before splitting, 1=shuffle, 0=do not shuffle') | ||
args = parser.parse_args() | ||
|
||
maybe_download(data_url, raw_data) | ||
preprocess(raw_data, shuffle=int(args.shuffle), ratio=float(args.ratio)) |
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,24 @@ | ||
# Copyright (c) 2016 PaddlePaddle Authors. 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. | ||
|
||
from paddle.trainer.PyDataProvider2 import * | ||
import numpy as np | ||
|
||
|
||
# define data types of input | ||
@provider(input_types=[dense_vector(13), dense_vector(1)]) | ||
def process(settings, input_file): | ||
data = np.load(input_file.strip()) | ||
for row in data: | ||
yield row[:-1].tolist(), row[-1:].tolist() |
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,79 @@ | ||
# Copyright (c) 2016 PaddlePaddle Authors. 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. | ||
import logging | ||
import argparse | ||
import numpy as np | ||
from py_paddle import swig_paddle, DataProviderConverter | ||
from paddle.trainer.PyDataProvider2 import * | ||
from paddle.trainer.config_parser import parse_config | ||
|
||
logging.basicConfig(level=logging.INFO) | ||
|
||
|
||
def predict(input_file, model_dir): | ||
# prepare PaddlePaddle environment, load models | ||
swig_paddle.initPaddle("--use_gpu=0") | ||
conf = parse_config('trainer_config.py', 'is_predict=1') | ||
network = swig_paddle.GradientMachine.createFromConfigProto( | ||
conf.model_config) | ||
network.loadParameters(model_dir) | ||
slots = [dense_vector(13)] | ||
converter = DataProviderConverter(slots) | ||
|
||
data = np.load(input_file) | ||
ys = [] | ||
for row in data: | ||
result = network.forwardTest(converter([[row[:-1].tolist()]])) | ||
y_true = row[-1:].tolist()[0] | ||
y_predict = result[0]['value'][0][0] | ||
ys.append([y_true, y_predict]) | ||
|
||
ys = np.matrix(ys) | ||
avg_err = np.average(np.square((ys[:, 0] - ys[:, 1]))) | ||
logging.info('MSE of test set is %f' % avg_err) | ||
|
||
# draw a scatter plot | ||
import matplotlib | ||
matplotlib.use('Agg') | ||
import matplotlib.pyplot as plt | ||
fig, ax = plt.subplots() | ||
|
||
ax.scatter(ys[:, 0], ys[:, 1]) | ||
y_range = [ys[:, 0].min(), ys[:, 0].max()] | ||
ax.plot(y_range, y_range, 'k--', lw=4) | ||
ax.set_xlabel('True ($1000)') | ||
ax.set_ylabel('Predicted ($1000)') | ||
ax.set_title('Predictions on boston housing price') | ||
fig.savefig('image/predictions.png', dpi=60) | ||
plt.close(fig) | ||
|
||
|
||
if __name__ == '__main__': | ||
parser = argparse.ArgumentParser( | ||
description='predict house price and save the result as image.') | ||
parser.add_argument( | ||
'-m', | ||
'--model', | ||
dest='model', | ||
default='output/pass-00029', | ||
help='model path') | ||
parser.add_argument( | ||
'-t', | ||
'--test_data', | ||
dest='test_data', | ||
default='data/housing.test.npy', | ||
help='test data path') | ||
args = parser.parse_args() | ||
|
||
predict(input_file=args.test_data, model_dir=args.model) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
其中$x_{i1}, \ldots, x_{id}$是$d$个属性上的取值,$y_i$是待预测的目标。
->
其中$x_{i1}, \ldots, x_{id}$是第$i$个样本$d$个属性上的取值,$y_i$是该样本待预测的目标。