app21-final-updates

.streamlit/config.toml

@@ -3,4 +3,7 @@ primaryColor="#4F7D7B"
 backgroundColor="#2B2934"
 secondaryBackgroundColor="#1C1B21"
 textColor="#ffffff"
-font="sans serif"
+font="sans serif"
+
+[deprecation]
+showPyplotGlobalUse = false

README.md

@@ -49,8 +49,9 @@
 | [Day_#17](https://github.com/silvermete0r/31github/blob/master/projects/day17.py)  | 📝 Named Entity Recognition (NER) using BERT model from HuggingFace | This app helps to identify the entities in a given text. The `bert-base-NER` model used from HuggingFace. | HuggingFace Transformers |
 | [Day_#18](https://github.com/silvermete0r/31github/blob/master/projects/day18.py)  | 📝 Text English to French Language Translation App | This app uses the `Helsinki-NLP/opus-mt-en-fr` model to translate English text to French. | HuggingFace Transformers |
 | [Day_#19](https://github.com/silvermete0r/31github/blob/master/projects/day19.py)  | 📷 Instagram Analytics Web App | This app uses the `instaloader` library to download Instagram profile data. | Instaloader |
-| [Day_#20](https://github.com/silvermete0r/31github/blob/master/projects/day20.py)  | ₿ Cryptocurrency Analytics Web App | This app is cryptocurrency analytics dashboard that allows you to analyze price, volume, and volatility of different cryptocurrencies. | Cryptocompare API |
-| [Day_#21](https://github.com/silvermete0r/31github/blob/master/projects/day21.py)  |  |  |  |
+| [Day_#20](https://github.com/silvermete0r/31github/blob/master/projects/day20.py)  | 🥮 Cryptocurrency Analytics Web App | This app is cryptocurrency analytics dashboard that allows you to analyze price, volume, and volatility of different cryptocurrencies. | Cryptocompare API |
+| [Day_#21](https://github.com/silvermete0r/31github/blob/master/projects/day21.py)  | 💰 Data Scientists Salary Analysis | This app shows how to use Streamlit to create charts and plots for data visualization on the example of Data Scientists Salary Analysis. | Streamlit Docs |
+| [Day_#22](https://github.com/silvermete0r/31github/blob/master/projects/day22.py)  |  |  |  |
 
 
 <!-- LICENSE -->

app.py

@@ -49,8 +49,9 @@ def main():
             'Day_#17': '📝 Named Entity Recognition (NER) using BERT model from HuggingFace',
             'Day_#18': '📝 Text English to French Language Translation App',
             'Day_#19': '📷 Instagram Analytics Web App',
-            'Day_#20': '₿ Cryptocurrency Analytics Web App',
-            'Day_#21': '',
+            'Day_#20': '🥮 Cryptocurrency Analytics Web App',
+            'Day_#21': '💰 Data Scientists Salary Analysis',
+            'Day_#22': '',
         },
         'Description': {
             'Day_#1': 'This app analyze data about NBA Basketball Player Stats in Regular Seasons! Data taken from official resource using web scrapping!',
@@ -73,7 +74,8 @@ def main():
             'Day_#18': 'This app uses the `Helsinki-NLP/opus-mt-en-fr` model to translate English text to French.',
             'Day_#19': 'This app uses the `instaloader` library to download Instagram profile data.',
             'Day_#20': 'This app is cryptocurrency analytics dashboard that allows you to analyze price, volume, and volatility of different cryptocurrencies.',
-            'Day_#21': '',
+            'Day_#21': 'This app shows how to use Streamlit to create charts and plots for data visualization on the example of Data Scientists Salary Analysis.',
+            'Day_#22': '',
         },
         'Reference': {
             'Day_#1': 'Data Professor',
@@ -96,7 +98,8 @@ def main():
             'Day_#18': 'HuggingFace Transformers',
             'Day_#19': 'Instaloader',
             'Day_#20': 'Cryptocompare API',
-            'Day_#21': '',
+            'Day_#21': 'Streamlit Docs',
+            'Day_#22': '',
         },
     })
 

projects/day21.py

@@ -1,5 +1,9 @@
 import streamlit as st
 import pandas as pd
+from sklearn.preprocessing import LabelEncoder
+import matplotlib.pyplot as plt
+import seaborn as sns
+sns.set_theme(style="whitegrid", palette="rocket")
 
 @st.cache_data()
 def load_data():
@@ -20,4 +24,60 @@ def app():
 
     data = load_data()
 
-    st.markdown('### Line Chart')
+    st.markdown('### Data Overview')
+    st.info(f'Number of rows: {data.shape[0]} and Number of columns: {data.shape[1]}')
+    st.caption('Columns:')
+    st.json(data.columns.to_list())
+    st.caption('Data:')
+    st.dataframe(data.style.highlight_max(axis=0, color='green').highlight_min(axis=0, color='purple'))
+    st.caption('Data Types:')
+    st.json(data.dtypes.to_dict())
+    st.caption('Numeric Data Description:')
+    st.dataframe(data.describe().T.style.highlight_max(axis=0, color='green').highlight_min(axis=0, color='purple'))
+
+    st.markdown('### Data Visualization')
+
+    st.markdown('#### Salary Distribution')
+    plt.figure(figsize=(12, 6), dpi=300)
+    sns.histplot(data['salary_in_usd'], kde=True, bins=30)
+    plt.xlabel('Salary in USD')
+    st.pyplot()
+
+    st.markdown('#### Distribution of experience levels')
+    st.bar_chart(data['experience_level'].value_counts())
+
+    st.markdown('#### Distribution of employment types')
+    st.bar_chart(data['employment_type'].value_counts())
+
+    st.markdown('#### Top 10 Job Titles')
+    st.bar_chart(data['job_title'].value_counts().head(10))
+
+    st.markdown('#### Top 10 Locations for Data Scientists')
+    st.bar_chart(data['company_location'].value_counts().head(10))
+
+    st.markdown('#### Salary vs Experience Level')
+    plt.figure(figsize=(12, 6), dpi=300)
+    sns.boxplot(x='experience_level', y='salary_in_usd', data=data, order=data['experience_level'].value_counts().index)
+    plt.xlabel('Experience Level')
+    plt.ylabel('Salary in USD')
+    st.pyplot()
+
+    st.markdown('#### Salary vs Employment Type')
+    plt.figure(figsize=(12, 6), dpi=300)
+    sns.boxplot(x='employment_type', y='salary_in_usd', data=data, order=data['employment_type'].value_counts().index)
+    plt.xlabel('Employment Type')
+    plt.ylabel('Salary in USD')
+    st.pyplot()
+
+    def data_prepare(data):
+        le = LabelEncoder()
+        for col in ['experience_level', 'employment_type', 'salary_currency', 'job_title', 'employee_residence', 'company_location', 'company_size']:
+            data[col] = le.fit_transform(data[col])
+        return data
+
+    st.markdown('#### Correlation Matrix')
+    with st.spinner('Preparing data...'):
+        data = data_prepare(data)
+    plt.figure(figsize=(12, 6), dpi=300)
+    sns.heatmap(data.corr(), annot=True, cmap='rocket', fmt='.2f')
+    st.pyplot()