# ML3
Introduction
ML3是TechYoung课程辅助工具包.
## Distribution |
Run the following commands to register, build and upload the package to PYPI. |
python3 setup.py sdist upload |
The home page on PYPI is: https://pypi.org/project/wcc/ |
Install
sudo pip3 install ml3
Usage
After installation, run the following command:
import ml3
Methods:
plot.histplot(data, column_name, **kwargs):
*data:* dataframe
*column_name:* column name of dataframe, 例如 “hr_mean”
*kwargs:* “xmin”, “xmax”
plot.gmmplot(data, column_names, k_range, **kwargs):
*data:* dataframe
*column_names:* list of columns name of dataframe, 例如 [“hr_mean”, “hr_std”]
*k_range:* the range of components (k), 例如 [2, 11] or (2, 11)
*kwargs:* “xmin”, “xmax”, “ymax”, “bins”
plot.kmeansplot(data, column_names, k_range, **kwargs):
*data:* dataframe
*column_names:* list of columns name of dataframe, 例如 [“hr_mean”, “hr_std”]
*k_range:* the range of clusters (k), 例如 [2, 11] or (2, 11)
*kwargs:* “xmin”, “xmax” “ymax”, “bins”
plot.metricplot(n_clusters_range, scores, scores2=[], **kwargs):
*n_clusters_range：* tuple or list of range，例如 (2, 10)
*scores:* list of score
*scores:* list of score2 (option)
*kwargs:* “x_label”, “y_label”
plot.errorbarplot(data, x, y=[], y2=[], **kwargs):
*data:* dataframe
*x:* x-axis column name，例如 “ctime”
*y:* y column name，例如 [“hr_mean”, “hr_std”]
*y2:* y2 column name，例如 [“br_mean”, “br_std”] (option)
*kwargs:* “X_LABEL”, “Y_LABEL”, “TITLE”, “LIMIT”
plot.pcaplot(data, column_names, **kwargs):
*data:* dataframe
*column_names:* list of columns name of dataframe, 例如 [“hr_mean”, “hr_std”]
*kwargs:* “n_components”
plot.tsenplot(data, column_names, **kwargs):
*data:* dataframe
*column_names:* list of columns name of dataframe, 例如 [“hr_mean”, “hr_std”]
*kwargs:* “n_components”
plot.kalmanplot(data, column_names, dim_x=2, dim_z=1, x=[], p=[], f=[], q=[], h=[], r=1, **kwargs):
*data:* dataframe
*column_names:* list of columns name of rawdata dataframe, 例如 [“hr”]
*dim_x:* the size of the state vector，状态空间维度
默认为2
*dim_z:* the size of the measurement vector，观测矩阵维度
默认为1
*x:* filter state estimate，初始状态预测矩阵
默认为[1, 0.1]，分别为心率和心率变化率
*p:* covariance matrix，协方差矩阵
默认为[[1, 0.1], [0.1, 1]]，心率变化率和人的心率是一定的关系，根据运动状态或者濒死会有很明显的差别，选择0.1代表有一定关系，但是不关系大
*q:* process uncertainty/noise，噪声矩阵，此矩阵不能为0，否则数据会异常
默认为[[0.0001, 0], [0, 0.0001]]，因为数据都是在cpu中进行，不会产生噪音
*r:* measurement uncertainty/noise，测量误差
默认为1，测量误差，医疗器械心率误差规定为+-1
*h:* measurement function
Sometimes certain states are measured, when others are not. For example, the first, third and fifth states of a five-dimensional state vector are measurable, while second and fourth states are not measurable H = [[1, 0, 0, 0, 0], [0, 0, 1, 0, 0], [0, 0, 0, 0, 1]]
*f:* state transistion matrix，状态转移矩阵
默认为[[1, 0.5], [0, 1]]，此矩阵不能对称，否则会计算异常
*kwargs:* None
seaborn.boxplot(x, y, **kwargs):
此函数需要ml4进行对原始数据进行窗口化分类
*x:* the UNIX timestamp list from ml4
*y:* the data list from ml4
*kwargs:* “X_LABEL”, “Y_LABEL”, “TITLE”
seaborn.violinplot(x, y, **kwargs):
此函数需要ml4进行对原始数据进行窗口化分类
*x:* the UNIX timestamp list from ml4
*y:* the data list from ml4
*kwargs:* “X_LABEL”, “Y_LABEL”, “TITLE”
Example
import ml3
import ml4
import pandas as pd
data = pd.read_csv("feature.csv")
# histogram
ml3.plot.histplot(data, "hr_mean")
# error bar
ml3.plot.errorbarplot(data, "ctime", ["hr_mean", "hr_std"], ["br_mean", "br_std"])
# single feature
ml3.plot.kmeansplot(data, "hr_mean", (2, 10))
ml3.plot.gmmplot(data, "hr_mean", (2, 10))
# multiple features
ml3.plot.gmmplot(data, ["hr_mean", "hr_std", "br_mean", "br_std", "mo_mean", "mo_std"], (2, 10))
ml3.plot.kmeansplot(data, ["hr_mean", "hr_std", "br_mean", "br_std", "mo_mean", "mo_std"], (2, 10))
# two scores metricplot
scores = [110704, 75304, 60731, 52297, 45675, 41231, 37744, 35247, 33263]
scores2 = [0.05, 0.09, 0.15, 0.2, 0.3, 0.5, 0.6, 0.9, 1]
ml3.plot.metricplot((2, 11), scores, scores2)
# boxplot and violoinplot
x, y = ml4.ml4.getWindowData(data, "ctime", "hr")
timeList = []
for i in x:
tmp = datetime.fromtimestamp(i)
timeList.append(tmp.strftime("%H:%M"))
ml3.seaborn.boxplot(timeList, y)
ml3.seaborn.violinplot(timeList, y)
data = pd.read_csv("rawdata.csv")
ml3.plot.kalmanplot(data, ["hr"])
data["log1p"] = np.log1p(data["br_std"])
ml3.plot.kmeansplot(data, ["log1p"], (2, 10), ymax=15, bins=0.01)Note
版本里的1.2.4是旧的版本。1.2.5和以后的版本是用于函数计算的版本。 1.2.5以及以后版本将去掉wcc自动框架. 目录下的子目录：libwebp-0.4.1-linux-x86-64 需要从网上下载，然后把里面的bin下的gif2webp放到/usr/bin里。这样就可以在wcc里调用了.
2 maintainers
Metadata
2 maintainers
Release files for ml3 1.2.7
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| ml3-1.2.7.tar.gz | 12.4 kB | Without provenance | Details for ml3-1.2.7.tar.gz |
Release files / ml3-1.2.7.tar.gz
| Download URL | ml3-1.2.7.tar.gz |
|---|---|
| Size | 12.4 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
a503aa7be912ab76e826053109c45833bec4f756fb290be2abae5a699baac67f
|
|
BLAKE2b-256 checksum How to use checksums |
d7f56b2461ab97809acd71d4da13b62f3d7397e92c773bd57dca0cd30eb7016c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
Python-urllib/3.6
|
Security for ml3 1.2.7
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.
