I'm using the LightGBM Package.
I have successfully created a new tree using "create_tree_digraph" but I face some trouble understanding the result.
There is "leaf_value" in a leaf node. I don't know what it means. Please, somebody help me understand this. Thanks. :)
I used this example code from here: https://www.analyticsvidhya.com/blog/2017/06/which-algorithm-takes-the-crown-light-gbm-vs-xgboost/
#importing standard libraries
import numpy as np
import pandas as pd
from pandas import Series, DataFrame
import graphviz
import lightgbm as lgb
#loading our training dataset 'adult.csv' with name 'data' using pandas
data=pd.read_csv('./adult.csv',header=None)
#Assigning names to the columns
data.columns=['age','workclass','fnlwgt','education','education-num','marital_Status','occupation','relationship','race','sex','capital_gain','capital_loss','hours_per_week','native_country','Income']
# Label Encoding our target variable
from sklearn.preprocessing import LabelEncoder,OneHotEncoder
l=LabelEncoder()
l.fit(data.Income)
data.Income=Series(l.transform(data.Income)) #label encoding our target variable
#One Hot Encoding of the Categorical features
one_hot_workclass=pd.get_dummies(data.workclass)
one_hot_education=pd.get_dummies(data.education)
#removing categorical features
data.drop(['workclass','education','marital_Status','occupation','relationship','race','sex','native_country'],axis=1,inplace=True)
#Merging one hot encoded features with our dataset 'data'
data=pd.concat([data,one_hot_workclass,one_hot_education],axis=1)
#Here our target variable is 'Income' with values as 1 or 0.
#Separating our data into features dataset x and our target dataset y
x=data.drop('Income',axis=1)
y=data.Income
#Imputing missing values in our target variable
y.fillna(y.mode()[0],inplace=True)
#Now splitting our dataset into test and train
from sklearn.model_selection import train_test_split
x_train,x_test,y_train,y_test=train_test_split(x,y,test_size=.3)
train_data=lgb.Dataset(x_train,label=y_train)
#setting parameters for lightgbm
param = {'num_leaves':150, 'objective':'binary','max_depth':3,'learning_rate':.05,'max_bin':200}
param['metric'] = ['auc', 'binary_logloss']
#training our model using light gbm
num_round=50
lgbm=lgb.train(param,train_data,num_round)
graph = lgb.create_tree_digraph(lgbm)
graph.render(view=True)
Then I applied 'create_tree_digraph' function.
Pics
These are the raw predicted probabilities before the sigmoid function is applied. However, one thing to be aware of is your image is only showing 1 tree out of the entire model so it will not be the same as the actual outcome (unless your model is just this 1 tree).
This Image is showing what it would look like if you applied the sigmoid to the leaf values prior to creating the plots.
Related
I'm new to python but I'm trying to run a regression with a bunch of different variables. So far I've got it down to Scikit. I've been searching for hours but can't seem to find a way to import the data and then run a linear regression on it while returning the coefficients of each variable. Any help is much appreciated. I have 15 columns that I want to run against the X.
X = Margin
Ys = A1, B1, C1, D1 etc.
Example set below:
Margin,A1
-8,110.7
-10,112
-1,106.7
9,109
-9,107.5
1,108.1
-19,109.2
Here's what I've got so far I know it's not much
import pandas as pd
data = pd.read_csv("NBA.csv")
As a convention in machine learning we consider X as the features and Y as the target.
If you want to run a linear regression and extract the coefficients, you can do the following :
# import the needed libraries
import pandas as pd
from sklearn.linear_model import LinearRegression
# Import the data
data = pd.read_csv("NBA.csv")
# Specify the features and the target
target = 'Margin'
features = data.columns.tolist() # This is the column names of your data as a list
features.remove(target) # We remove the target from the list of features
# Train the model
model = LinearRegression() # Instantiate the model
model.fit(data[features].values, data[target].values) # fit the model to the data
print(features) # Returns the name of each feature
print(model.coef_) # Returns the coefficients for each feature (in the same order of your features)
I do have the below dataset.
I've created Logistic Regression out of it and checked Accuracy and is working fine. So now requirement is I've a new data with Age 30 and EstimatedSalary 50000 and I would like to predict whether Purchased will be 0 or 1. How to pass the new values 30 and 50000 in my python code.
Below is the python code which I've used.
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import accuracy_score
%matplotlib inline
dataset = pd.read_csv(r"suv_data.csv")
X=dataset.iloc[:,[0,1]].values
y=dataset.iloc[:,2].values
X_train,X_test,y_train,y_test=train_test_split(X, y, test_size=0.2, random_state=1)
sc=StandardScaler()
X_train=sc.fit_transform(X_train)
X_test=sc.transform(X_test)
classifier=LogisticRegression(random_state=0)
classifier.fit(X_train,y_train)
y_pred=classifier.predict(X_test)
accuracy_score(y_test,y_pred)*100
Regards,
Bharath Vikas
In general, to evaluate (i.e. call .predict in sklearn) a trained model, you need to input samples that have the same shape as the samples the model was trained on.
In your case I suppose (see my comment on your question) you wanted to have samples with Age and EstimatedSalary in the training set using Purchased as label.
Then, to test on a single sample just try this:
single_test_sample = pd.DataFrame({'Age':[30], 'EstimatedSalary':[50000]}).iloc[:,[0,1]].values
single_test_sample = sc.transform(single_test_sample)
single_test_prediction = classifier.predict(single_test_sample)
Note that you can also add more values in the test dataframe Age and EstimatedSalary columns, now I only added the sample you were interested in. If you add more, the model will output a prediction for each row in the test dataframe.
Also note that your code and mine, will also work without this .values at the end of the train/test set as sklearn already provides functionality with pandas dataframes.
Your question is not clear but I understand that you need to use the fitted model to predict a new sample.
After having fitted your model just use this:
new_sample = np.array([[30,50000]]) # 2D numpy array
new_sample_sc = sc.transform(new_sample)
y_pred_new = classifier.predict(new_sample_sc)
print(y_pred_new)
I have data in the following form
Class Feature set list
classlabel1 - [size,time] example:[6780.3,350.00]
classlabel2 - [size,time]
classlabel3 - [size,time]
classlabel4 - [size,time]
How do I save this data in excel sheet and how can I train the model using this feature set? Currently I am working on SVM classifier.
I have tried saving the feature set list in a dataframe and saving this dataframe to a csv file. But the size and time are getting split into two different columns.
The dataframe is getting saved in csv file in the following way:
col 0 col1 col2
62309 396.5099154 label1
I would like to train and test on the feature vector [size,time] combined. Is it possible and is this a right way? If it is possible, how can I do it?
Firstly responding to your question:
I would like to train and test on the feature vector [size,time]
combined. Is it possible and is this a right way? If it is possible,
how can I do it?
Combining the two is not the right thing to do because both are in two different scales (if they are actually what there name suggests) and also combining them will result in loss of information which they will provide, so they are two totally independent features for any ML supervised algorithm. So I would suggest to treat these two features separately rather than combining into one.
Now let's move onto to next section:
How do I save this data in excel sheet and how can I train the model
using this feature set? Currently I am working on SVM classifier.
Storing data : In my opinion, you can store data in whichever format you want but I would prefer storing data in csv format as it is convenient and loading of data file is faster.
sample_data.csv
size,time,class_label
100,150,label1
200,250,label2
240,180,label1
Below is the code for reading the data from csv and training SVM :
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn import preprocessing
from sklearn.svm import SVC
from sklearn.metrics import accuracy_score
# loading data
data = pd.read_csv("sample_data.csv", error_bad_lines=True,
warn_bad_lines=True)
# Dividing into dependent and independent features
Y = data.class_label_col.values
X = data.drop("class_label_col", axis=1).values
# encode the class column values
label_encoded_Y = preprocessing.LabelEncoder().fit_transform(list(Y))
# split training and testing data
x_train,x_test,y_train,y_test=train_test_split(X,label_encoded_Y,
train_size=0.8,
test_size=0.2)
# Now use the whichever trainig algo you want
clf = SVC(gamma='auto')
clf.fit(x_train, y_train)
# Using the predictor
y_pred = clf.predict(x_test)
Since size and time are different features, you should separate them into 2 different columns so your model could set separate weight to each of them, i.e.
# data.csv
size time label
6780.3 3,350.00 classLabel1
...
If you want to transform the data you have into the format above you could use pandas.read_excel and use ast to transform the string list into python list object.
import pandas as pd
import ast
df = pd.read_excel("data.xlsx")
size_time = [(ast.literal_eval(x)[0], ast.literal_eval(x)[1]) for x in df["Feature set list"]]
size = [x[0] for x in size_time]
time = [x[1] for x in size_time]
label = df["Class"]
new_df = pd.DataFrame({"size":size, "time":time, "label":label})
# This will result in the DataFrame below.
# size time label
# 6780.3 350.0 classlabel1
# Save DataFrame to csv
new_df.to_csv("data_fix.csv")
# Use it
x = new_df.drop("label", axis=1)
y = new_df.label
# Further data preparation, such as split the dataset
# into train and test set, etc.
...
Hope this helps
This is an example of my data:
filename,2,3,4,5,6,7,class
a.txt,0,0,0,0,0,0,0
b.txt,0,0,0,0,0,1,0
c.txt,0,0,0,0,1,0,0
d.txt,1,0,1,0,0,1,1
When I train my data, I just use the columns from 2 -> 7 as input, class as output. But when I test the model after it trained and save, I need to know that which files are belong to which class. I mean like how to know d.txt is class 1.
I use pandas to import the data from .csv file, I use train set and test set in 2 different csv files. In the train phase, I uses columns 2-7 as input, and column class as target, these columns are numerical. The filename class is just text. In the test phase, I need to know the filename with the predicted class. But I don't know how to do that.
Thanks
P/s: I used MLP,SVM, NB as classifier.
Assuming your data is in .csv format:
filename,2,3,4,5,6,7,class
a.txt,0,0,0,0,0,0,0
b.txt,0,0,0,0,0,1,0
c.txt,0,0,0,0,1,0,0
d.txt,1,0,1,0,0,1,1
You can output the corresponding filename to a predicted class using:
features=[1,0,1,0,0,1] #input
output=clf.predict([features])[0] #predicted class
print(df[df["class"]==output]["filename"]) #corresponding filename
Note that in your example you're facing the problem where the amount of features is greater than the amount of examples, therefore the classifier may deteriorate.
Hopefully you just gave a sample of your data. In this case you're likely to be good. Just watch out for what classifier to use.
Full code:
import numpy as np
import pandas as pd
from sklearn import svm
df=pd.read_csv('file.csv')
X = df.iloc[:,1:7].values
y = df.iloc[:,7:8].values
clf = svm.SVC() #using SVM as classifier
clf.fit(X, y)
features=[1,0,1,0,0,1]
output=clf.predict([features])[0]
print(df[df["class"]==output]["filename"])
With R, one can ignore a variable (column) while building a model with the following syntax:
model = lm(dependant.variable ~ . - ignored.variable, data=my.training,set)
It's very handy when your data set contains indexes or ID.
How would you do that with SKlearn in python, assuming your data are Pandas data frames ?
So this is from my own code I used to do some prediction on StackOverflow last year:
from __future__ import division
from pandas import *
from sklearn import cross_validation
from sklearn import metrics
from sklearn.ensemble import GradientBoostingClassifier
basic_feature_names = [ 'BodyLength'
, 'NumTags'
, 'OwnerUndeletedAnswerCountAtPostTime'
, 'ReputationAtPostCreation'
, 'TitleLength'
, 'UserAge' ]
fea = # extract the features - removed for brevity
# construct our classifier
clf = GradientBoostingClassifier(n_estimators=num_estimators, random_state=0)
# now fit
clf.fit(fea[basic_feature_names], orig_data['OpenStatusMod'].values)
# now
priv_fea = # this was my test dataset
# now calculate the predicted classes
pred = clf.predict(priv_fea[basic_feature_names])
So if we wanted a subset of the features for classification I could have done this:
# want to train using fewer features so remove 'BodyLength'
basic_feature_names.remove('BodyLength')
clf.fit(fea[basic_feature_names], orig_data['OpenStatusMod'].values)
So the idea here is that a list can be used to select a subset of the columns in the pandas dataframe, as such we can construct a new list or remove a value and use this for selection
I'm not sure how you could do this easily using numpy arrays as indexing is done differently.