Use scaled data on test file - python

I want to fit a model of logistic regression on my first file(F1) of data and i want to test it
on another file named F2(The same exersise of another year).
Code on F1:
sc = preprocessing.StandardScaler()
X_train = sc.fit_transform(X_train)
X_test = sc.transform(X_test)
logistic = LogisticRegression(random_state =0,max_iter = 300 ,penalty = 'l2')
model = logistic.fit(X,y)
ScaledObj = X_train
How can i do to use the scaled data in my test file please
I did this but i don"t knwo how to use ScaledObj on my test
Code on my File Test(F2)
F2 = pd.read_csv("F2.csv", sep =',')
y_test = F2['y']
X_test = F2.copy()
del X_test['y']
y_pred = model.predict(X_test)
proba= model.predict_proba(X_test)[:, 1]
Auc_Test = metrics.roc_auc_score(y_test, proba)

For best practice in a machine learning project, the typical workflow goes like this:
fit the scaler to the training data separated from the testing data
transform the training data (you did this already with your fit_transform step)
transform the test data using the already-fitted scaler*. This prevents any data leakage between your training and testing data
Use the same fitted scaler* to transform any other validation or production data.
*-Note that the scaler only lives in memory, so if you want to use it in another script, you can use something like pickle or joblib to save the object for later use
You've done steps 1-3 correctly in your code above, and you can execute step 4 the same way. One thing I would recommend, however, is not to overwrite your variables, as this can be confusing when reading the code later.
F2 = pd.read_csv("F2.csv", sep =',')
y_test1 = F2['y']
X_test1 = F2.copy()
del X_test1['y']
#add this line, same as you did before
X_test1 = sc.transform(X_test1)
y_pred = model.predict(X_test1)

Related

Using seperated test and train files with train_test_split()

I have two .csv files that one of them is test.csv and the other one is train.csv. However, as you can predict the test file does not have the target column ('y' in this case) while train file has.
What I wanted to do is first using train file to train the system entirely, then using the test file just to see predictions.
I'm using from sklearn.model_selection import train_test_split() to create train and test examples but it accepts 1 file path only. I want to train the system using train file first, then when it finished I want to get test datas from test.csv file and make the predictions.
So first I tried classic way but decreasing test size so It'll be like "this file used for train only",
import pandas as pd
from sklearn.svm import SVC
dataset = pd.read_csv(r'path\train.csv', sep=",")
X_train, X_test, y_train, y_test = train_test_split(
X, y, test_size = 0.001, random_state = 45)
clf = SVC(kernel = 'rbf')
clf.fit(X_train, y_train)
but then, when it comes to real test part(which I want to use the data in test.csv that doesn't have target values), how can I import test.csv somehow I can use the test data in trained model above
#get data from test.csv as somehow X_test
clfPredict = clf.predict(X_test)
If this is not possible using train_test_split(), what's the proper way to accomplish this task?
You need to load the train CSV and split it to:
y_train = df1['Y column']
X_train = df1.drop('Y Column', axis = 1)
And regarding test:
X_test = df2
and y_test will be the result from clf.predict(X_test)

standardize data with K-Fold cross validation

I'm using StratifiedKFold so my code looks like this
def train_model(X,y,X_test,folds,model):
scores=[]
for fold_n, (train_index, valid_index) in enumerate(folds.split(X, y)):
X_train,X_valid = X[train_index],X[valid_index]
y_train,y_valid = y[train_index],y[valid_index]
model.fit(X_train,y_train)
y_pred_valid = model.predict(X_valid).reshape(-1,)
scores.append(roc_auc_score(y_valid, y_pred_valid))
print('CV mean score: {0:.4f}, std: {1:.4f}.'.format(np.mean(scores), np.std(scores)))
folds = StratifiedKFold(10,shuffle=True,random_state=0)
lr = LogisticRegression(class_weight='balanced',penalty='l1',C=0.1,solver='liblinear')
train_model(X_train,y_train,X_test,repeted_folds,lr)
now before train the model I want to standardize the data so which is the correct way?
1)
scaler = StandardScaler()
X_train = scaler.fit_transform(X_train)
X_test = scaler.transform(X_test)
doing this before calling train_model function
2)
doing standardization inside function like this
def train_model(X,y,X_test,folds,model):
scores=[]
for fold_n, (train_index, valid_index) in enumerate(folds.split(X, y)):
X_train,X_valid = X[train_index],X[valid_index]
y_train,y_valid = y[train_index],y[valid_index]
scaler = StandardScaler()
X_train = scaler.fit_transform(X_train)
X_vaid = scaler.transform(X_valid)
X_test = scaler.transform(X_test)
model.fit(X_train,y_train)
y_pred_valid = model.predict(X_valid).reshape(-1,)
scores.append(roc_auc_score(y_valid, y_pred_valid))
print('CV mean score: {0:.4f}, std: {1:.4f}.'.format(np.mean(scores), np.std(scores)))
As per my knowlwdge in 2nd option I'm not leaking the data.so which way is correct if I'm not using pipeline and also how to use pipeline if i want to use cross validation?
Indeed the second option is better because the scaler does not see the values of X_valid to scale X_train.
Now if you were to use a pipeline, you can do:
from sklearn.pipeline import make_pipeline
def train_model(X,y,X_test,folds,model):
pipeline = make_pipeline(StandardScaler(), model)
...
And then use pipeline instead of model. At every fit or predict call, it will automatically standardize the data at hand.
Note that you can also use the cross_val_score function from scikit-learn, with the parameter scoring='roc_auc'.
When to standardize your data may be a question better suited for Cross Validated.
IMO if your data are large then it probably doesn't matter too much (if you're using k-fold this may not be the case) but since you can, it's better to do it within your cross validation (k-fold), or option 2.
Also, see this for more information on overfitting in cross validation.

How to pass different set of data to train and test without splitting a dataframe. (python)?

I have gone through multiple questions that help divide your dataframe into train and test, with scikit, without etc.
But my question is I have 2 different csvs ( 2 different dataframes from different years). I want to use one as train and other as test?
How to do so for LinearRegression / any model?
Load the datasets individually.
Make sure they are in the same format of rows and columns (features).
Use the train set to fit the model.
Use the test set to predict the output after training.
# Load the data
train = pd.read_csv('train.csv')
test = pd.read_csv('test.csv')
# Split features and value
# when trying to predict column "target"
X_train, y_train = train.drop("target"), train["target"]
X_test, y_test = test.drop("target"), test["target"]
# Fit (train) model
reg = LinearRegression()
reg.fit(X_train, y_train)
# Predict
pred = reg.predict(X_test)
# Score
accuracy = reg.socre(X_test, y_test)
please skillsmuggler what about the X_train and X_Test how I can define it because when I try to do that it said NameError: name 'X_train' is not defined
I couldn't edit the first answer which is almost there. There is some code missing though...
# Load the data
train = pd.read_csv('train.csv')
test = pd.read_csv('test.csv')
y_train = train[:, :1] #if y is only one column
X_train = train[:, 1:]
# Fit (train) model
reg = LinearRegression()
reg.fit(X_train, y_train)
# Predict
pred = reg.predict(X_test)
# Score
accuracy = reg.socre(X_test, y_test)

using saved sklearn model to make prediction

I have a saved logistic regression model which I trained with training data and saved using joblib. I am trying to load this model in a different script, pass it new data and make a prediction based on the new data.
I am getting the following error "sklearn.exceptions.NotFittedError: CountVectorizer - Vocabulary wasn't fitted." Do I need to fit the data again ? I would have thought that the point of being able to save the model would be to not have to do this.
The code I am using is below excluding the data cleaning section. Any help to get the prediction to work would be appreciated.
new_df = pd.DataFrame(latest_tweets,columns=['text'])
new_df.to_csv('new_tweet.csv',encoding='utf-8')
csv = 'new_tweet.csv'
latest_df = pd.read_csv(csv)
latest_df.dropna(inplace=True)
latest_df.reset_index(drop=True,inplace=True)
new_x = latest_df.text
loaded_model = joblib.load("finalized_mode.sav")
tfidf_transformer = TfidfTransformer()
cvec = CountVectorizer()
x_val_vec = cvec.transform(new_x)
X_val_tfidf = tfidf_transformer.transform(x_val_vec)
result = loaded_model.predict(X_val_tfidf)
print (result)
Your training part have 3 parts which are fitting the data:
CountVectorizer: Learns the vocabulary of the training data and returns counts
TfidfTransformer: Learns the counts of the vocabulary from previous part, and returns tfidf
LogisticRegression: Learns the coefficients for features for optimum classification performance.
Since each part is learning something about the data and using it to output the transformed data, you need to have all 3 parts while testing on new data. But you are only saving the lr with joblib, so the other two are lost and with it is lost the training data vocabulary and count.
Now in your testing part, you are initializing new CountVectorizer and TfidfTransformer, and calling fit() (fit_transform()), which will learn the vocabulary only from this new data. So the words will be less than the training words. But then you loaded the previously saved LR model, which expects the data according to features like training data. Hence this error:
ValueError: X has 130 features per sample; expecting 223086
What you need to do is this:
During training:
filename = 'finalized_model.sav'
joblib.dump(lr, filename)
filename = 'finalized_countvectorizer.sav'
joblib.dump(cvec, filename)
filename = 'finalized_tfidftransformer.sav'
joblib.dump(tfidf_transformer, filename)
During testing
loaded_model = joblib.load("finalized_model.sav")
loaded_cvec = joblib.load("finalized_countvectorizer.sav")
loaded_tfidf_transformer = joblib.load("finalized_tfidftransformer.sav")
# Observe that I only use transform(), not fit_transform()
x_val_vec = loaded_cvec.transform(new_x)
X_val_tfidf = loaded_tfidf_transformer.transform(x_val_vec)
result = loaded_model.predict(X_val_tfidf)
Now you wont get that error.
Recommendation:
You should use TfidfVectorizer in place of both CountVectorizer and TfidfTransformer, so that you dont have to use two objects all the time.
And along with that you should use Pipeline to combine the two steps:- TfidfVectorizer and LogisticRegression, so that you only have to use a single object (which is easier to save and load and generic handling).
So edit the training part like this:
tfidf_vectorizer = TfidfVectorizer()
lr = LogisticRegression()
tfidf_lr_pipe = Pipeline([('tfidf', tfidf_vectorizer), ('lr', lr)])
# Internally your X_train will be automatically converted to tfidf
# and that will be passed to lr
tfidf_lr_pipe.fit(X_train, y_train)
# Similarly here only transform() will be called internally for tfidfvectorizer
# And that data will be passed to lr.predict()
y_preds = tfidf_lr_pipe.predict(x_test)
# Now you can save this pipeline alone (which will save all its internal parts)
filename = 'finalized_model.sav'
joblib.dump(tfidf_lr_pipe, filename)
During testing, do this:
loaded_pipe = joblib.load("finalized_model.sav")
result = loaded_model.predict(new_x)
You have not fit the CountVectorizer.
You should do like this..
cvec = CountVectorizer()
x_val_vec = cvec.fit_transform(new_x)
Similarly, TfidTransformer must be used like this..
X_val_tfidf = tfidf_transformer.fit_transform(x_val_vec)

Keras Regression using Scikit Learn StandardScaler with Pipeline and without Pipeline

I am comparing the performance of two programs about KerasRegressor using Scikit-Learn StandardScaler: one program with Scikit-Learn Pipeline and one program without the Pipeline.
Program 1:
estimators = []
estimators.append(('standardise', StandardScaler()))
estimators.append(('multiLayerPerceptron', KerasRegressor(build_fn=build_nn, nb_epoch=num_epochs, batch_size=10, verbose=0)))
pipeline = Pipeline(estimators)
log = pipeline.fit(X_train, Y_train)
Y_deep = pipeline.predict(X_test)
Program 2:
scale = StandardScaler()
X_train = scale.fit_transform(X_train)
X_test = scale.fit_transform(X_test)
model_np = KerasRegressor(build_fn=build_nn, nb_epoch=num_epochs, batch_size=10, verbose=0)
log = model_np.fit(X_train, Y_train)
Y_deep = model_np.predict(X_test)
My problem is that Program 1 can achieve R2 score as 0.98 (3 trials on average) while Program 2 only achieve R2 score as 0.84 (3 trials on average.) Can anyone explain the difference between these two programs?
In the second case, you are calling StandardScaler.fit_transform() on both X_train and X_test. Its wrong usage.
You should call fit_transform() on X_train and then call only transform() on the X_test. Because thats what the Pipeline does.
The Pipeline as the documentation states, will:
fit():
Fit all the transforms one after the other and transform the data,
then fit the transformed data using the final estimator
predict():
Apply transforms to the data, and predict with the final estimator
So you see, it will only apply transform() to the test data, not fit_transform().
So elaborate my point, your code should be:
scale = StandardScaler()
X_train = scale.fit_transform(X_train)
#This is the change
X_test = scale.transform(X_test)
model_np = KerasRegressor(build_fn=build_nn, nb_epoch=num_epochs, batch_size=10, verbose=0)
log = model_np.fit(X_train, Y_train)
Y_deep = model_np.predict(X_test)
Calling fit() or fit_transform() on test data wrongly scales it to a different scale than what was used on train data. And is a source of change in prediction.
Edit: To answer the question in comment:
See, fit_transform() is just a shortcut function for doing fit() and then transform(). For StandardScaler, fit() doesnt return anything, just learns the mean and standard deviation of data. And then transform() applies the learning on the data to return new scaled data.
So what you are saying leads to below two scenarios:
Scenario 1: Wrong
1) X_scaled = scaler.fit_transform(X)
2) Divide the X_scaled into X_scaled_train, X_scaled_test and run your model.
No need to scale again.
Scenario 2: Wrong (Basically equal to Scenario 1, reversing the scaling and spitting operations)
1) Divide the X into X_train, X_test
2) scale.fit_transform(X) [# You are not using the returned value, only fitting the data, so equivalent to scale.fit(X)]
3.a) X_train_scaled = scale.transform(X_train) #[Equals X_scaled_train in scenario 1]
3.b) X_test_scaled = scale.transform(X_test) #[Equals X_scaled_test in scenario 1]
You can try any of the scenario and maybe it will increase the performance of your model.
But there is one very important thing which is missing in them. When you do scaling on the whole data and then divide them into train and test, it is assumed that you know the test (unseen) data, which will not be true in real world cases. And will give you results which will not be according to real world results. Because in the real world, whole of the data will be our training data. It may also lead to over-fitting because the model has some information about the test data already.
So when evaluating the performance of machine learning models, it is recommended that you keep aside the test data before performing any operations on it. Because it is our unseen data, we know nothing about it. So ideal path of operations would be the one I answered, ie.:
1) Divide X into X_train and X_test (same for y)
2) X_train_scaled = scale.fit_transform(X_train) [#Learn the mean and SD of train data]
3) X_test_scaled = scale.transform(X_test) [#Use the mean and SD learned in step2 to convert test data]
4) Use the X_train_scaled for training the model and X_test_scaled in evaluation.
Hope it makes sense to you.

Categories

Resources