My Machine Learning Notes
Nice to know (remember) python libraries and their functions, used to build & train succesfull machine learning models. Some of the code examples are taken from Kaggle.
Covered libraries are: Numpy, Pandas, Scikit learn, Tensorflow.
import pandas as pd # load Pandas libraryRead data
# Read the data
X_full = pd.read_csv('../input/train.csv', index_col='Id')
X_test_full = pd.read_csv('../input/test.csv', index_col='Id')Load data (into subsets)
# Remove rows with missing target, separate target from predictors
X_full.dropna(axis=0, subset=['SalePrice'], inplace=True)
y = X_full.SalePrice
X_full.drop(['SalePrice'], axis=1, inplace=True) # Select categorical columns with relatively low cardinality (convenient but arbitrary)
categorical_cols = [cname for cname in X_train_full.columns if X_train_full[cname].nunique() < 10 and X_train_full[cname].dtype == "object"]
# Select numerical columns
numerical_cols = [cname for cname in X_train_full.columns if X_train_full[cname].dtype in ['int64', 'float64']]
# Keep selected columns only
my_cols = categorical_cols + numerical_cols
X_train = X_train_full[my_cols].copy()
X_valid = X_valid_full[my_cols].copy()Write data
# Save test predictions to file
output = pd.DataFrame({'Id': X_test.index,
'SalePrice': preds_test})
output.to_csv('submission.csv', index=False)Pandas DataFrame operations
# Use only numerical predictors
X = X_full.select_dtypes(exclude=['object'])
X_test = X_test_full.select_dtypes(exclude=['object'])
# Inspect columns having missing values
missing_val_count_by_column = (X_train.isnull().sum())
print(missing_val_count_by_column[missing_val_count_by_column > 0]) # Get names of columns with missing values
X_train_missing_columns = [col for col in X_train.columns
if X_train[col].isnull().sum() > 0] # Your code here
# Drop columns in training and validation data
reduced_X_train = X_train.drop(X_train_missing_columns, axis=1)
reduced_X_valid = X_valid.drop(X_train_missing_columns, axis=1)Manipulate data
from sklearn.impute import SimpleImputer
# imputation
my_imputer = SimpleImputer()
imputed_X_train = pd.DataFrame(my_imputer.fit_transform(X_train))
imputed_X_valid = pd.DataFrame(my_imputer.transform(X_valid))
# imputation removed column names; put them back
imputed_X_train.columns = X_train.columns
imputed_X_valid.columns = X_valid.columnsPipelines
from sklearn.compose import ColumnTransformer
from sklearn.pipeline import Pipeline
from sklearn.impute import SimpleImputer
from sklearn.preprocessing import OneHotEncoder
# Preprocessing for numerical data
numerical_transformer = SimpleImputer(strategy='constant')
# Preprocessing for categorical data
categorical_transformer = Pipeline(steps=[
('imputer', SimpleImputer(strategy='most_frequent')),
('onehot', OneHotEncoder(handle_unknown='ignore'))
])
# Bundle preprocessing for numerical and categorical data
preprocessor = ColumnTransformer(
transformers=[
('num', numerical_transformer, numerical_cols),
('cat', categorical_transformer, categorical_cols)
])
from sklearn.metrics import mean_absolute_error
# Bundle preprocessing and modeling code in a pipeline
my_pipeline = Pipeline(steps=[('preprocessor', preprocessor),
('model', model)
])
# Preprocessing of training data, fit model
my_pipeline.fit(X_train, y_train)
# Preprocessing of validation data, get predictions
preds = my_pipeline.predict(X_valid)
# Evaluate the model
score = mean_absolute_error(y_valid, preds)
print('MAE:', score)Evaluate data
Cross-validation
from sklearn.model_selection import cross_val_score
# Multiply by -1 since sklearn calculates *negative* MAE
scores = -1 * cross_val_score(my_pipeline, X, y,
cv=5,
scoring='neg_mean_absolute_error')
print("MAE scores:\n", scores)
print("Average MAE score (across experiments):")
print(scores.mean())Build models
from sklearn.model_selection import train_test_split
# Break off validation set from training data
X_train, X_valid, y_train, y_valid = train_test_split(X, y, train_size=0.8, test_size=0.2)Train models
Random Forest Regression
# Define and fit model
model = RandomForestRegressor(n_estimators=100, random_state=0)
model.fit(final_X_train, y_train)
# Get validation predictions and MAE
preds_valid = model.predict(final_X_valid)
print("MAE (Your approach):")
print(mean_absolute_error(y_valid, preds_valid))Gradiant Boosting Regression
from xgboost import XGBRegressor
from sklearn.metrics import mean_absolute_error
# Define the model
my_model_2 = XGBRegressor(random_state=0,n_estimators=500,learning_rate=0.05,early_stopping_rounds=5) # Your code here
# Fit the model
my_model_2.fit(X_train,y_train,eval_set=[(X_train,y_train),(X_valid,y_valid)]) # Your code here
# Get predictions
predictions_2 = my_model_2.predict(X_valid) # Your code here
# Calculate MAE
mae_2 = mean_absolute_error(predictions_2,y_valid) # Your code here
# Uncomment to print MAE
print("Mean Absolute Error:" , mae_2)Evaluate models
from sklearn.ensemble import RandomForestRegressor
from sklearn.metrics import mean_absolute_error
# Function for comparing different data approaches (on the same model)
def score_dataset(X_train, X_valid, y_train, y_valid):
model = RandomForestRegressor(n_estimators=100, random_state=0)
model.fit(X_train, y_train)
preds = model.predict(X_valid)
return mean_absolute_error(y_valid, preds)Visualization
Two subplots in one figure:
for c in cols[:2]:
import warnings
warnings.filterwarnings("ignore")
fig, axes = plt.subplots(nrows=1, ncols=2, figsize=(12, 5))
sns.kdeplot(x=X_train[c], ax=axes[0], label='Original')
axes[0].set_title(c + ' - Original')
axes[0].legend()
sns.kdeplot(x=X_transformed[c], ax=axes[1], label='Transformed')
axes[1].set_title(c + ' - Transformed')
axes[1].legend()
plt.tight_layout()
plt.show()