tds

November 20, 2024

[5]: # Import necessary libraries

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split
from sklearn.ensemble import GradientBoostingRegressor, VotingRegressor
from sklearn.metrics import mean_squared_error, mean_absolute_error, r2_score
from xgboost import XGBRegressor
from lightgbm import LGBMRegressor
from google.colab import files
import io

# Step 1: Upload the dataset

print("Please upload your dataset for TDS prediction:")
uploaded = files.upload()
file_name = list(uploaded.keys())[0]
data = pd.read_excel(io.BytesIO(uploaded[file_name]))

# Step 2: Preprocess the dataset

# Specify target and feature columns
target_column = 'TDS' # Target variable
X_columns = ['Mean Chlorophyll Index', 'NDWI', 'Blue Band', 'Green Band',
'Red Band', 'NIR Band', 'SWIR Band'] # Feature variables

# Extract features (X) and target (y)

X = data[X_columns]
y = data[target_column]

# Handle missing values

X.fillna(X.mean(), inplace=True)
y.fillna(y.mean(), inplace=True)

# Step 3: Handle Outliers using IQR

def handle_outliers(df, column):
Q1 = df[column].quantile(0.25)
Q3 = df[column].quantile(0.75)
IQR = Q3 - Q1

1
 lower_bound = Q1 - 1.5 * IQR
upper_bound = Q3 + 1.5 * IQR
return df[(df[column] >= lower_bound) & (df[column] <= upper_bound)]

Please upload your dataset for TDS prediction:

<IPython.core.display.HTML object>
Saving Data set for TDS.xlsx to Data set for TDS (2).xlsx
<ipython-input-5-a4e2705f28f1>:30: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame

See the caveats in the documentation: https://pandas.pydata.org/pandas-

docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
X.fillna(X.mean(), inplace=True)

[6]: # Apply outlier handling to target variable

data_cleaned = handle_outliers(data, target_column)
X = data_cleaned[X_columns]
y = data_cleaned[target_column]

# Step 4: Train-test split

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2,␣
↪random_state=42)

# Step 5: Train Models

# Gradient Boosting
gbr = GradientBoostingRegressor(random_state=42)
gbr.fit(X_train, y_train)

# XGBoost
xgb = XGBRegressor(random_state=42)
xgb.fit(X_train, y_train)

# LightGBM
lgbm = LGBMRegressor(random_state=42)
lgbm.fit(X_train, y_train)

# Ensemble Model
ensemble = VotingRegressor(estimators=[
('gbr', gbr),
('xgb', xgb),
('lgbm', lgbm)
])
ensemble.fit(X_train, y_train)

# Step 6: Evaluate Models

def evaluate_model(model, X_test, y_test, name):

2
 y_pred = model.predict(X_test)
rmse = mean_squared_error(y_test, y_pred, squared=False)
mae = mean_absolute_error(y_test, y_pred)
r2 = r2_score(y_test, y_pred)
mpe = 100 * np.mean(np.abs((y_test - y_pred) / y_test)) # Mean Percentage␣
↪Error

print(f"\n{name} Performance:")
print(f"RMSE: {rmse:.2f}")
print(f"MAE: {mae:.2f}")
print(f"R²: {r2:.2f}")
print(f"Mean Percentage Error (MPE): {mpe:.2f}%")
return y_pred, mpe

[LightGBM] [Warning] Found whitespace in feature_names, replace with underlines

[LightGBM] [Info] Auto-choosing col-wise multi-threading, the overhead of
testing was 0.000392 seconds.
You can set `force_col_wise=true` to remove the overhead.
[LightGBM] [Info] Total Bins 1014
[LightGBM] [Info] Number of data points in the train set: 432, number of used
features: 7
[LightGBM] [Info] Start training from score 132.202315
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf

3
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf

4
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] Found whitespace in feature_names, replace with underlines
[LightGBM] [Info] Auto-choosing col-wise multi-threading, the overhead of
testing was 0.000066 seconds.
You can set `force_col_wise=true` to remove the overhead.
[LightGBM] [Info] Total Bins 1014
[LightGBM] [Info] Number of data points in the train set: 432, number of used
features: 7
[LightGBM] [Info] Start training from score 132.202315
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf

5
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf

6
 [LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf
[LightGBM] [Warning] No further splits with positive gain, best gain: -inf

[7]: # Evaluate individual models and ensemble

y_pred_gbr, mpe_gbr = evaluate_model(gbr, X_test, y_test, "Gradient Boosting␣
↪Regressor")

y_pred_xgb, mpe_xgb = evaluate_model(xgb, X_test, y_test, "XGBoost Regressor")

y_pred_lgbm, mpe_lgbm = evaluate_model(lgbm, X_test, y_test, "LightGBM␣
↪Regressor")

y_pred_ensemble, mpe_ensemble = evaluate_model(ensemble, X_test, y_test,␣

↪"Ensemble Model")

7
 Gradient Boosting Regressor Performance:
RMSE: 24.79
MAE: 18.00
R²: 0.75
Mean Percentage Error (MPE): 18.64%

XGBoost Regressor Performance:

RMSE: 26.29
MAE: 19.40
R²: 0.72
Mean Percentage Error (MPE): 19.82%

LightGBM Regressor Performance:

RMSE: 27.54
MAE: 19.69
R²: 0.69
Mean Percentage Error (MPE): 19.39%

Ensemble Model Performance:

RMSE: 25.07
MAE: 18.27
R²: 0.74
Mean Percentage Error (MPE): 18.65%
/usr/local/lib/python3.10/dist-packages/sklearn/metrics/_regression.py:492:
FutureWarning: 'squared' is deprecated in version 1.4 and will be removed in
1.6. To calculate the root mean squared error, use the
function'root_mean_squared_error'.
warnings.warn(
/usr/local/lib/python3.10/dist-packages/sklearn/metrics/_regression.py:492:
FutureWarning: 'squared' is deprecated in version 1.4 and will be removed in
1.6. To calculate the root mean squared error, use the
function'root_mean_squared_error'.
warnings.warn(
/usr/local/lib/python3.10/dist-packages/sklearn/metrics/_regression.py:492:
FutureWarning: 'squared' is deprecated in version 1.4 and will be removed in
1.6. To calculate the root mean squared error, use the
function'root_mean_squared_error'.
warnings.warn(
/usr/local/lib/python3.10/dist-packages/sklearn/metrics/_regression.py:492:
FutureWarning: 'squared' is deprecated in version 1.4 and will be removed in
1.6. To calculate the root mean squared error, use the
function'root_mean_squared_error'.
warnings.warn(

[8]: # Step 8: Summary of Model Performance

print("\nSummary of Model Performance:")

8
 print(f"Gradient Boosting Regressor: RMSE={mean_squared_error(y_test,␣
↪y_pred_gbr, squared=False):.2f}, MAE={mean_absolute_error(y_test,␣

↪y_pred_gbr):.2f}, R²={r2_score(y_test, y_pred_gbr):.2f}, MPE={mpe_gbr:.2f}%")

print(f"XGBoost Regressor: RMSE={mean_squared_error(y_test, y_pred_xgb,␣

↪squared=False):.2f}, MAE={mean_absolute_error(y_test, y_pred_xgb):.2f},␣

↪R²={r2_score(y_test, y_pred_xgb):.2f}, MPE={mpe_xgb:.2f}%")

print(f"LightGBM Regressor: RMSE={mean_squared_error(y_test, y_pred_lgbm,␣

↪squared=False):.2f}, MAE={mean_absolute_error(y_test, y_pred_lgbm):.2f},␣

↪R²={r2_score(y_test, y_pred_lgbm):.2f}, MPE={mpe_lgbm:.2f}%")

print(f"Ensemble Model: RMSE={mean_squared_error(y_test, y_pred_ensemble,␣

↪squared=False):.2f}, MAE={mean_absolute_error(y_test, y_pred_ensemble):.2f},␣

↪R²={r2_score(y_test, y_pred_ensemble):.2f}, MPE={mpe_ensemble:.2f}%")

Summary of Model Performance:

Gradient Boosting Regressor: RMSE=24.79, MAE=18.00, R²=0.75, MPE=18.64%
XGBoost Regressor: RMSE=26.29, MAE=19.40, R²=0.72, MPE=19.82%
LightGBM Regressor: RMSE=27.54, MAE=19.69, R²=0.69, MPE=19.39%
Ensemble Model: RMSE=25.07, MAE=18.27, R²=0.74, MPE=18.65%
/usr/local/lib/python3.10/dist-packages/sklearn/metrics/_regression.py:492:
FutureWarning: 'squared' is deprecated in version 1.4 and will be removed in
1.6. To calculate the root mean squared error, use the
function'root_mean_squared_error'.
warnings.warn(
/usr/local/lib/python3.10/dist-packages/sklearn/metrics/_regression.py:492:
FutureWarning: 'squared' is deprecated in version 1.4 and will be removed in
1.6. To calculate the root mean squared error, use the
function'root_mean_squared_error'.
warnings.warn(
/usr/local/lib/python3.10/dist-packages/sklearn/metrics/_regression.py:492:
FutureWarning: 'squared' is deprecated in version 1.4 and will be removed in
1.6. To calculate the root mean squared error, use the
function'root_mean_squared_error'.
warnings.warn(
/usr/local/lib/python3.10/dist-packages/sklearn/metrics/_regression.py:492:
FutureWarning: 'squared' is deprecated in version 1.4 and will be removed in
1.6. To calculate the root mean squared error, use the
function'root_mean_squared_error'.
warnings.warn(

[9]: # Step 7: Visualize Predictions vs Actuals (Ensemble Model)

plt.figure(figsize=(10, 6))
plt.scatter(y_test, y_pred_ensemble, alpha=0.7, label="Predictions")
plt.plot([y_test.min(), y_test.max()], [y_test.min(), y_test.max()],␣
↪color='red', linestyle='--', label="Ideal Fit")

plt.xlabel("Actual TDS")

9
 plt.ylabel("Predicted TDS")
plt.title("Predicted vs Actual TDS (Ensemble Model)")
plt.legend()
plt.show()

[10]: # Visualize Predictions vs Actuals for Gradient Boosting Regressor

plt.figure(figsize=(10, 6))
plt.scatter(y_test, y_pred_gbr, alpha=0.7, label="Predictions")
plt.plot([y_test.min(), y_test.max()], [y_test.min(), y_test.max()],␣
↪color='red', linestyle='--', label="Ideal Fit")

plt.xlabel("Actual TDS")
plt.ylabel("Predicted TDS")
plt.title("Predicted vs Actual TDS (Gradient Boosting Regressor)")
plt.legend()
plt.show()

10
[11]: # Visualize Predictions vs Actuals for XGBoost Regressor
plt.figure(figsize=(10, 6))
plt.scatter(y_test, y_pred_xgb, alpha=0.7, label="Predictions")
plt.plot([y_test.min(), y_test.max()], [y_test.min(), y_test.max()],␣
↪color='red', linestyle='--', label="Ideal Fit")

plt.xlabel("Actual TDS")
plt.ylabel("Predicted TDS")
plt.title("Predicted vs Actual TDS (XGBoost Regressor)")
plt.legend()
plt.show()

11
[12]: # Visualize Predictions vs Actuals for LightGBM Regressor
plt.figure(figsize=(10, 6))
plt.scatter(y_test, y_pred_lgbm, alpha=0.7, label="Predictions")
plt.plot([y_test.min(), y_test.max()], [y_test.min(), y_test.max()],␣
↪color='red', linestyle='--', label="Ideal Fit")

plt.xlabel("Actual TDS")
plt.ylabel("Predicted TDS")
plt.title("Predicted vs Actual TDS (LightGBM Regressor)")
plt.legend()
plt.show()

12
[13]: # Visualize Predictions vs Actuals for Ensemble Model
plt.figure(figsize=(10, 6))
plt.scatter(y_test, y_pred_ensemble, alpha=0.7, label="Predictions")
plt.plot([y_test.min(), y_test.max()], [y_test.min(), y_test.max()],␣
↪color='red', linestyle='--', label="Ideal Fit")

plt.xlabel("Actual TDS")
plt.ylabel("Predicted TDS")
plt.title("Predicted vs Actual TDS (Ensemble Model)")
plt.legend()
plt.show()

13
14