Description
More than 100 billion bananas are consumed on our planet each year, making them one of the most popular agricultural products.
And I’m one of those people who buys bananas all year long!
Banana Quality Dataset (Why?)
This dataset was selected because it allows for interesting work in banana classification and prediction of banana quality using regression—so you can buy only tasty bananas!
Data Source — kaggle.com
Data Description
sample_id: A unique identifier assigned to each banana sample in the dataset. This allows for tracking and unique referencing of the samples.
variety: The cultivar or breed of the banana, such as Cavendish, Red Dacca, or Lady Finger. Knowing the specific cultivar provides context for the sample’s physical characteristics and growing conditions.
region: The geographical origin of the banana, such as Ecuador, the Philippines, or Costa Rica. The region can influence factors like climate, soil, and cultivation methods, all of which affect the banana’s quality.
quality_score: A numerical score, likely ranging from 1 to 4, that evaluates the overall quality of the banana sample. This may include factors such as appearance, texture, and the absence of defects.
quality_category: A text label that classifies the quality score into broader categories, such as «Excellent, ” and so on. This simplifies the perception of the quality assessment.
ripeness_index: A numerical index representing the level of banana ripeness, possibly ranging from 1 (green/unripe) to 10 (overripe). This quantifies the fruit’s maturity.
ripeness_category: A text label, such as „Green, ” „Yellow, ” „Ripe, ” or „Overripe, ” corresponding to the ripeness index. This provides a clear qualitative classification of ripeness.
sugar_content_brix: The sugar content in a banana, measured in Brix degrees. This is a common method for assessing the sweetness and quality of the fruit.
firmness_kgf: The firmness of the banana, measured in kilograms-force. This indicates the texture and ripeness of the sample.
length_cm: The physical length of the banana in centimeters. This measurement can vary depending on the variety and growing conditions.
Schedules
It was decided to map the quality distribution of bananas across categories and ripeness using histograms, as histograms effectively display the count of data points for each class, which is essential for understanding potential future training needs.
Graphs showing the relationship between banana quality, region, sugar content, and tree age were presented as line graphs to explore the possibility of predicting banana quality using regression.
(1)
Histogram Distribution of banana quality by category
plt.figure () sns.countplot (x='quality_category', data=df, order=df['quality_category'].value_counts ().index, palette="viridis») plt.title ('Distribution of Banana Quality by Category', fontsize=14) plt.xlabel ('Quality Category', fontsize=12) plt.ylabel ('Count', fontsize=12) plt.show ()
((2))
Histogram Banana Ripeness Distribution
plt.figure () sns.countplot (x='ripeness_category', data=df, order=df['ripeness_category'].value_counts ().index, palette="plasma») plt.title ('Distribution of Banana Ripeness', fontsize=14) plt.xlabel ('Ripeness Category', fontsize=12) plt.ylabel ('Count', fontsize=12) plt.show ()
(((3)))
Line Graph Dependence of Banana Quality on Region
plt.figure () df.groupby ('region')['quality_score'].mean ().sort_values ().plot (kind='line', marker='o', color='teal') plt.title ('Dependence of Banana Quality on Region', fontsize=14) plt.xlabel ('Region', fontsize=12) plt.ylabel ('Average Quality Score', fontsize=12) plt.xticks (rotation=45) plt.show ()
((((4))))
Line Chart Banana Quality vs. Sugar Content
plt.figure () sns.lineplot (x='sugar_content_brix', y='quality_score', data=df, ci=None, color='orange') plt.title ('Banana Quality Dependence on Sugar Content', fontsize=14) plt.xlabel ('Sugar Content (Brix)', fontsize=12) plt.ylabel ('Quality Score', fontsize=12) plt.show ()
(((((5)))))
Line Graph Banana Quality vs. Age
plt.figure () sns.lineplot (x='tree_age_years', y='quality_score', data=df, ci=None, color='purple') plt.title ('Relationship between banana quality and tree age', fontsize=14) plt.xlabel ('Tree Age (years)', fontsize=12) plt.ylabel ('Quality Score', fontsize=12) plt.show ()
Bibliography
PHOTO: https://ru.pinterest.com/search/pins/?q=minion%20with%20bananas&rs=typed
DATASET: https://www.kaggle.com/datasets/mrmars1010/banana-quality-dataset?resource=download
Notebook and Dataset
https://drive.google.com/drive/my-drive?hl=ru
