[FIXED] How do I predict a specific row of data in scikit-learn?

Issue

I trained a model using the following code

import pandas as pd
from sklearn.model_selection import train_test_split

data = pd.read_csv('sampledata.csv')

cols_to_use = ['OUNdif', 'UFMdif', 'Class']
X = data[cols_to_use]

y = data.W

X_train, X_valid, y_train, y_valid = train_test_split(X, y)

from xgboost import XGBClassifier

my_model = XGBClassifier(n_estimators=1000, learning_rate=0.05)
my_model.fit(X_train, y_train, 
             early_stopping_rounds=5, 
             eval_set=[(X_valid, y_valid)], 
             verbose=False)

from sklearn.metrics import accuracy_score

predictions = my_model.predict(X_valid)

Now if I were to add a new row to the bottom (#355), how would I use my now trained model to predict just that row? (Without accidentally using it as part of the training data)

Solution

You can pass as a list your OUNdif, UFMdif, Class ie: [6, 5, 25] Then predict the value from an np array with just this line and get the first element of the prediction.

input_to_guess = [6,5,25]
print(my_model.predict(np.array(input_to_guess).reshape((1, -1)))[0])

It should return the value you want

Answered By - Tristan LG