
__author__ = "Donald Ghazi"
__email__ = "donald@donaldghazi.com"
__website__ = "donaldghazi.com"


import warnings
import matplotlib.pyplot as plt
import pandas as pd

from sklearn.linear_model import LinearRegression # Build my model 
from sklearn.metrics import mean_absolute_error # Evaluate my model 
from sklearn.utils.validation import check_is_fitted
warnings.simplefilter(action="ignore", category=FutureWarning) # Eliminate warnings that pop-up (visual distractions)


# Write a function named wrangle that takes a file path as an argument and returns a DataFrame
def wrangle(filepath):
    # Read CSV file into DataFrame
    df = pd.read_csv(filepath)
    return df


# Use wrangle function to create a DataFrame df from the CSV file "data/buenos-aires-real-estate-1.csv"
df = wrangle("data/buenos-aires-real-estate-1.csv")
print ("df shape:", df.shape)
df.head()

df shape: (8606, 16)


# Look at "place_with_parent_names" column to see if "Capital Federal" is within that string
mask_ba = df["place_with_parent_names"].str.contains("Capital Federal")
mask_ba.head()

0     True
1    False
2    False
3    False
4     True
Name: place_with_parent_names, dtype: bool


mask_ba = df["place_with_parent_names"].str.contains("Capital Federal")

# Show a DataFrame where everything in that DataFrame is or are properties in Buenos Aires proper 
df[mask_ba].head()


# Add to wrangle function so that the DataFrame returns only includes apartments in Buenos Aires "Capital Federal"

def wrangle(filepath):
    # Read CSV file into DataFrame
    df = pd.read_csv(filepath)
    
    # Subset to properties in "Capital Federal"
    mask_ba = df["place_with_parent_names"].str.contains("Capital Federal")
    df = df[mask_ba]
    
    return df


df = wrangle("data/buenos-aires-real-estate-1.csv")
print ("df shape:", df.shape)
df.head()

df shape: (2524, 16)


# See what sorts of properties in my dataset 
df["property_type"].unique()

array(['apartment', 'house', 'PH', 'store'], dtype=object)


# Check work to make sure all of the properties are apartments
mask_apt = df["property_type"] == "apartment"
mask_apt.head()

0     True
4     True
9     True
11    True
18    True
Name: property_type, dtype: bool


def wrangle(filepath):
    # Read CSV file into DataFrame
    df = pd.read_csv(filepath)
    
    # Subset to properties in "Capital Federal"
    mask_ba = df["place_with_parent_names"].str.contains("Capital Federal")
    
    # Subset to "apartments"
    mask_apt = df["property_type"] == "apartment"
    
    df = df[mask_ba & mask_apt]
    
    return df


df = wrangle("data/buenos-aires-real-estate-1.csv")
print ("df shape:", df.shape)
df.head()

df shape: (2068, 16)


# Should be only apartment
df["property_type"].unique()

array(['apartment'], dtype=object)


df["currency"].unique()

array(['USD', nan, 'ARS'], dtype=object)


df.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 2068 entries, 0 to 8604
Data columns (total 16 columns):
 #   Column                      Non-Null Count  Dtype  
---  ------                      --------------  -----  
 0   operation                   2068 non-null   object 
 1   property_type               2068 non-null   object 
 2   place_with_parent_names     2068 non-null   object 
 3   lat-lon                     1984 non-null   object 
 4   price                       1944 non-null   float64
 5   currency                    1944 non-null   object 
 6   price_aprox_local_currency  1944 non-null   float64
 7   price_aprox_usd             1944 non-null   float64
 8   surface_total_in_m2         1430 non-null   float64
 9   surface_covered_in_m2       1859 non-null   float64
 10  price_usd_per_m2            1314 non-null   float64
 11  price_per_m2                1752 non-null   float64
 12  floor                       533 non-null    float64
 13  rooms                       1580 non-null   float64
 14  expenses                    521 non-null    object 
 15  properati_url               2068 non-null   object 
dtypes: float64(9), object(7)
memory usage: 274.7+ KB


mask_price = df["price_aprox_usd"] < 400_000
mask_price.head()

0      True
4      True
9      True
11     True
18    False
Name: price_aprox_usd, dtype: bool


def wrangle(filepath):
    # Read CSV file into DataFrame
    df = pd.read_csv(filepath)
    
    # Subset to properties in "Capital Federal"
    mask_ba = df["place_with_parent_names"].str.contains("Capital Federal")
    
    # Subset to "apartments"
    mask_apt = df["property_type"] == "apartment"
    
    # Subset to properties where "price_aprox_usd" < 400,000
    mask_price = df["price_aprox_usd"] < 400_000
    
    df = df[mask_ba & mask_apt & mask_price]
    
    return df


df = wrangle("data/buenos-aires-real-estate-1.csv")
print ("df shape:", df.shape)
df.head()

df shape: (1781, 16)


# Create a histogram of "surface_covered_in_m2"
plt.hist(df["surface_covered_in_m2"])
plt.xlabel(["Area [sq meters]"])
plt.title("Distribution of Apartment Sizes");


# Calculate the summary statistics for df using the describe method
df.describe()


# Look at the summary statistics for the "surface_covered_in_m2" feature
df.describe()["surface_covered_in_m2"]

count     1635.000000
mean        97.877064
std       1533.057610
min          0.000000
25%         38.000000
50%         50.000000
75%         73.000000
max      62034.000000
Name: surface_covered_in_m2, dtype: float64


# All observations should fall between the 0.1 and 0.9 quantiles for "surface_covered_in_m2"
df["surface_covered_in_m2"].quantile([0.1, 0.9])

0.1     31.0
0.9    100.0
Name: surface_covered_in_m2, dtype: float64


low, high = df["surface_covered_in_m2"].quantile([0.1, 0.9])
low

31.0


low, high = df["surface_covered_in_m2"].quantile([0.1, 0.9])
high

100.0


mask_area = df["surface_covered_in_m2"].between(low,high)
mask_area.head()

0     False
4      True
9      True
11    False
20    False
Name: surface_covered_in_m2, dtype: bool


# Add to wrangle function so that it removes observations that are outliers in the "surface_covered_in_m2"

def wrangle(filepath):
    # Read CSV file into DataFrame
    df = pd.read_csv(filepath)
    
    # Subset to properties in "Capital Federal"
    mask_ba = df["place_with_parent_names"].str.contains("Capital Federal")
    
    # Subset to "apartments"
    mask_apt = df["property_type"] == "apartment"
    
    # Subset to properties where "price_aprox_usd" < 400,000
    mask_price = df["price_aprox_usd"] < 400_000
    
    # Subset
    df = df[mask_ba & mask_apt & mask_price]
    
    # Remove outliers by "surface_covered_in_m2"
    low, high = df["surface_covered_in_m2"].quantile([0.1, 0.9])
    mask_area = df["surface_covered_in_m2"].between(low,high)
    df = df[mask_area]
    
    return df


df = wrangle("data/buenos-aires-real-estate-1.csv")
print("df shape:", df.shape)
df.head()

df shape: (1343, 16)


# Create a histogram of "surface_covered_in_m2"
plt.hist(df["surface_covered_in_m2"])
plt.xlabel(["Area [sq meters]"])
plt.title("Distribution of Apartment Sizes");


# Create a scatter plot that shows price ("price_aprox_usd") vs area ("surface_covered_in_m2") in my dataset
plt.scatter(x=df["surface_covered_in_m2"], y=df["price_aprox_usd"])
plt.xlabel("Area [sq meters]")
plt.ylabel("Price [USD]")
plt.title("Buenos Aires: Price vs. Area");


# Create the feature matrix named X_train, which I'll use to train my model
features = ["surface_covered_in_m2"] # It should contain one feature only: ["surface_covered_in_m2"]
X_train = df[features] # Feature matrix should always be two-dimensional
X_train.head()


X_train.shape

(1343, 1)


# Create the target vector named y_train, which I'll use to train my model
target = "price_aprox_usd" # Target should be "price_aprox_usd"
y_train = df[target]
y_train.head()

4     129000.0
9      87000.0
29    118000.0
40     57000.0
41     90000.0
Name: price_aprox_usd, dtype: float64


y_train.shape

(1343,)


# Calculate the mean of target vector y_train and assign it to the variable y_mean
y_mean = y_train.mean()
y_mean

135527.83871928515


# Create a list named y_pred_baseline containing the value of y_mean repeated so it's the same length at y
y_pred_baseline = y_mean
y_pred_baseline

135527.83871928515


# I actually need a list so take y_mean and put it inside a list
y_pred_baseline = [y_mean]
y_pred_baseline

[135527.83871928515]


# Multiply it by the length of y_train
y_pred_baseline = [y_mean] * len(y_train)
len(y_pred_baseline)

1343


# Making sure above is the samze size as my y_train
y_pred_baseline = [y_mean] * len(y_train)
len(y_pred_baseline) == len(y_train)

True


# Show relationship between the observations X_train and my dumb model's predictions y_pred_baseline
plt.plot(X_train, y_pred_baseline, color="orange", label="Baseline Model")
plt.scatter(X_train, y_train)
plt.xlabel("Area [sq meters]")
plt.ylabel("Price [USD]")
plt.title("Buenos Aires: Price vs. Area")
plt.legend();


# Calculate the baseline mean absolute error for predictions in y_pred_baseline as compared to the true targets in y

mae_baseline = mean_absolute_error(y_train, y_pred_baseline)

print("Mean apt price:", round(y_mean, 2))
print("Baseline MAE:", round(mae_baseline, 2))

Mean apt price: 135527.84
Baseline MAE: 45199.46


# Instantiate a LinearRegression model named model
model = LinearRegression()


# Fit my model to the data, X_train and y_train
model.fit(X_train, y_train)

LinearRegression()

LinearRegression()


# Checking my work
check_is_fitted(model)


# Use model's predict method to create a list of predictions for the observations in feature matrix X_train
y_pred_training = model.predict(X_train)  # Name this array y_pred_training.
y_pred_training[:5]

array([169151.87330223, 106064.44707446, 133101.91545779, 106064.44707446,
       124089.42599668])


# Calculate training mean absolute error for predictions in y_pred_training 
mae_training = mean_absolute_error(y_train, y_pred_training) # As compared to the true targets in y_train
print("Training MAE:", round(mae_training, 2))

Training MAE: 31248.26


# Import test data buenos-aires-test-features.csv into a DataFrame
X_test = pd.read_csv("data/buenos-aires-test-features.csv")[features]

# Generate a Series of predictions using the model
y_pred_test = pd.Series(model.predict(X_test)) # Make sure the X_train is same column order as X_test
y_pred_test.head()

0    117330.058901
1    135355.037823
2     88039.468152
3     88039.468152
4    106064.447074
dtype: float64


# Extract the intercept from the model, and assign it to the variable intercept
intercept = round(model.intercept_, 2) # Round it to two decimal points
print("Model Intercept:", intercept)
assert any([isinstance(intercept, int), isinstance(intercept, float)])

Model Intercept: 11433.31


# Extract the coefficient associated "surface_covered_in_m2" in my model
coefficient = round(model.coef_[0], 2) # Assign it to the variable `coefficient`
print('Model coefficient for "surface_covered_in_m2":', coefficient)
assert any([isinstance(coefficient, int), isinstance(coefficient, float)])

Model coefficient for "surface_covered_in_m2": 2253.12


# Print the equation that the model has determined for predicting apartment price based on size
print(f"apt_price = {intercept} + {coefficient} * surface_covered") # F string allows me to print out character values like "apt_price" and also insert variable names

apt_price = 11433.31 + 2253.12 * surface_covered


# Plot the relationship between the observations in X_train and my model's predictions y_pred_training
plt.plot(X_train, model.predict(X_train), color="r", label="Linear Model")
plt.scatter(X_train, y_train)
plt.xlabel("surface covered [sq meters]")
plt.ylabel("price [usd]")
plt.legend();

	operation	property_type	place_with_parent_names	lat-lon	price	currency	price_aprox_local_currency	price_aprox_usd	surface_total_in_m2	surface_covered_in_m2	price_usd_per_m2	price_per_m2	floor	rooms	expenses	properati_url
0	sell	apartment	\|Argentina\|Capital Federal\|Villa Crespo\|	-34.6047834183,-58.4586812499	180000.0	USD	2729232.0	180000.0	120.0	110.0	1500.000000	1636.363636	NaN	4.0	NaN	http://villa-crespo.properati.com.ar/12egq_ven...
1	sell	house	\|Argentina\|Bs.As. G.B.A. Zona Oeste\|La Matanza...	NaN	250000.0	USD	3790600.0	250000.0	117.0	120.0	2136.752137	2083.333333	NaN	4.0	NaN	http://ramos-mejia.properati.com.ar/s7pd_venta...
2	sell	house	\|Argentina\|Bs.As. G.B.A. Zona Oeste\|Morón\|Cast...	-34.6497002,-58.658073	410000.0	USD	6216584.0	410000.0	410.0	220.0	1000.000000	1863.636364	NaN	NaN	NaN	http://castelar-moron.properati.com.ar/11vgn_v...
3	sell	house	\|Argentina\|Bs.As. G.B.A. Zona Oeste\|Tres de Fe...	-34.5957086,-58.5669503	180000.0	USD	2729232.0	180000.0	200.0	135.0	900.000000	1333.333333	NaN	5.0	NaN	http://tres-de-febrero.properati.com.ar/7f7u_v...
4	sell	apartment	\|Argentina\|Capital Federal\|Chacarita\|	-34.5846508988,-58.4546932614	129000.0	USD	1955949.6	129000.0	76.0	70.0	1697.368421	1842.857143	NaN	NaN	NaN	http://chacarita.properati.com.ar/10qlv_venta_...

	operation	property_type	place_with_parent_names	lat-lon	price	currency	price_aprox_local_currency	price_aprox_usd	surface_total_in_m2	surface_covered_in_m2	price_usd_per_m2	price_per_m2	floor	rooms	expenses	properati_url
0	sell	apartment	\|Argentina\|Capital Federal\|Villa Crespo\|	-34.6047834183,-58.4586812499	180000.0	USD	2729232.0	180000.0	120.0	110.0	1500.000000	1636.363636	NaN	4.0	NaN	http://villa-crespo.properati.com.ar/12egq_ven...
4	sell	apartment	\|Argentina\|Capital Federal\|Chacarita\|	-34.5846508988,-58.4546932614	129000.0	USD	1955949.6	129000.0	76.0	70.0	1697.368421	1842.857143	NaN	NaN	NaN	http://chacarita.properati.com.ar/10qlv_venta_...
9	sell	apartment	\|Argentina\|Capital Federal\|Villa Luro\|	-34.6389789,-58.500115	87000.0	USD	1319128.8	87000.0	48.0	42.0	1812.500000	2071.428571	NaN	NaN	NaN	http://villa-luro.properati.com.ar/12m82_venta...
11	sell	apartment	\|Argentina\|Capital Federal\|Once\|	-34.6050060697,-58.4001162302	60000.0	USD	909744.0	60000.0	28.0	28.0	2142.857143	2142.857143	NaN	1.0	NaN	http://once.properati.com.ar/zz0q_venta_depart...
18	sell	apartment	\|Argentina\|Capital Federal\|Belgrano\|	-34.5646269,-58.4412741	600000.0	USD	9097440.0	600000.0	220.0	NaN	2727.272727	NaN	NaN	5.0	NaN	http://belgrano.properati.com.ar/jvge_venta_de...

	operation	property_type	place_with_parent_names	lat-lon	price	currency	price_aprox_local_currency	price_aprox_usd	surface_total_in_m2	surface_covered_in_m2	price_usd_per_m2	price_per_m2	floor	rooms	expenses	properati_url
0	sell	apartment	\|Argentina\|Capital Federal\|Villa Crespo\|	-34.6047834183,-58.4586812499	180000.0	USD	2729232.0	180000.0	120.0	110.0	1500.000000	1636.363636	NaN	4.0	NaN	http://villa-crespo.properati.com.ar/12egq_ven...
4	sell	apartment	\|Argentina\|Capital Federal\|Chacarita\|	-34.5846508988,-58.4546932614	129000.0	USD	1955949.6	129000.0	76.0	70.0	1697.368421	1842.857143	NaN	NaN	NaN	http://chacarita.properati.com.ar/10qlv_venta_...
9	sell	apartment	\|Argentina\|Capital Federal\|Villa Luro\|	-34.6389789,-58.500115	87000.0	USD	1319128.8	87000.0	48.0	42.0	1812.500000	2071.428571	NaN	NaN	NaN	http://villa-luro.properati.com.ar/12m82_venta...
11	sell	apartment	\|Argentina\|Capital Federal\|Once\|	-34.6050060697,-58.4001162302	60000.0	USD	909744.0	60000.0	28.0	28.0	2142.857143	2142.857143	NaN	1.0	NaN	http://once.properati.com.ar/zz0q_venta_depart...
18	sell	apartment	\|Argentina\|Capital Federal\|Belgrano\|	-34.5646269,-58.4412741	600000.0	USD	9097440.0	600000.0	220.0	NaN	2727.272727	NaN	NaN	5.0	NaN	http://belgrano.properati.com.ar/jvge_venta_de...

	operation	property_type	place_with_parent_names	lat-lon	price	currency	price_aprox_local_currency	price_aprox_usd	surface_total_in_m2	surface_covered_in_m2	price_usd_per_m2	price_per_m2	floor	rooms	expenses	properati_url
0	sell	apartment	\|Argentina\|Capital Federal\|Villa Crespo\|	-34.6047834183,-58.4586812499	180000.0	USD	2729232.0	180000.0	120.0	110.0	1500.000000	1636.363636	NaN	4.0	NaN	http://villa-crespo.properati.com.ar/12egq_ven...
4	sell	apartment	\|Argentina\|Capital Federal\|Chacarita\|	-34.5846508988,-58.4546932614	129000.0	USD	1955949.6	129000.0	76.0	70.0	1697.368421	1842.857143	NaN	NaN	NaN	http://chacarita.properati.com.ar/10qlv_venta_...
9	sell	apartment	\|Argentina\|Capital Federal\|Villa Luro\|	-34.6389789,-58.500115	87000.0	USD	1319128.8	87000.0	48.0	42.0	1812.500000	2071.428571	NaN	NaN	NaN	http://villa-luro.properati.com.ar/12m82_venta...
11	sell	apartment	\|Argentina\|Capital Federal\|Once\|	-34.6050060697,-58.4001162302	60000.0	USD	909744.0	60000.0	28.0	28.0	2142.857143	2142.857143	NaN	1.0	NaN	http://once.properati.com.ar/zz0q_venta_depart...
18	sell	apartment	\|Argentina\|Capital Federal\|Belgrano\|	-34.5646269,-58.4412741	600000.0	USD	9097440.0	600000.0	220.0	NaN	2727.272727	NaN	NaN	5.0	NaN	http://belgrano.properati.com.ar/jvge_venta_de...

	operation	property_type	place_with_parent_names	lat-lon	price	currency	price_aprox_local_currency	price_aprox_usd	surface_total_in_m2	surface_covered_in_m2	price_usd_per_m2	price_per_m2	floor	rooms	expenses	properati_url
0	sell	apartment	\|Argentina\|Capital Federal\|Villa Crespo\|	-34.6047834183,-58.4586812499	180000.0	USD	2729232.0	180000.0	120.0	110.0	1500.000000	1636.363636	NaN	4.0	NaN	http://villa-crespo.properati.com.ar/12egq_ven...
4	sell	apartment	\|Argentina\|Capital Federal\|Chacarita\|	-34.5846508988,-58.4546932614	129000.0	USD	1955949.6	129000.0	76.0	70.0	1697.368421	1842.857143	NaN	NaN	NaN	http://chacarita.properati.com.ar/10qlv_venta_...
9	sell	apartment	\|Argentina\|Capital Federal\|Villa Luro\|	-34.6389789,-58.500115	87000.0	USD	1319128.8	87000.0	48.0	42.0	1812.500000	2071.428571	NaN	NaN	NaN	http://villa-luro.properati.com.ar/12m82_venta...
11	sell	apartment	\|Argentina\|Capital Federal\|Once\|	-34.6050060697,-58.4001162302	60000.0	USD	909744.0	60000.0	28.0	28.0	2142.857143	2142.857143	NaN	1.0	NaN	http://once.properati.com.ar/zz0q_venta_depart...
20	sell	apartment	\|Argentina\|Capital Federal\|San Nicolás\|	-34.603898,-58.378617	69000.0	USD	1046205.6	69000.0	NaN	22.0	NaN	3136.363636	23.0	2.0	NaN	http://san-nicolas.properati.com.ar/rnju_venta...

Overview¶

Prepare Data¶

Import¶

Explore¶

Split¶

Build Model¶

Baseline¶

Iterate¶

Evaluate¶

Communicate Results¶

	price	price_aprox_local_currency	price_aprox_usd	surface_total_in_m2	surface_covered_in_m2	price_usd_per_m2	price_per_m2	floor	rooms
count	1.781000e+03	1.781000e+03	1781.000000	1247.000000	1635.000000	1198.000000	1618.000000	482.000000	1385.000000
mean	2.285410e+05	2.178037e+06	143647.215239	114.417001	97.877064	2231.534726	4340.345113	10.085062	2.434657
std	4.100189e+05	1.141710e+06	75298.758506	1755.263895	1533.057610	650.139587	8126.940810	44.783320	1.065193
min	0.000000e+00	0.000000e+00	0.000000	0.000000	0.000000	0.000000	2.224587	1.000000	1.000000
25%	9.000000e+04	1.349454e+06	89000.000000	40.000000	38.000000	1815.485830	1973.750000	2.000000	2.000000
50%	1.300000e+05	1.849813e+06	122000.000000	55.000000	50.000000	2151.642336	2411.239496	4.000000	2.000000
75%	1.911340e+05	2.653420e+06	175000.000000	79.000000	73.000000	2577.003583	2958.157462	7.000000	3.000000
max	5.487000e+06	6.034635e+06	398000.000000	62034.000000	62034.000000	5975.609756	57156.250000	616.000000	6.000000