Merging files

README.md

@@ -4,21 +4,32 @@ Final Project for CAPP122
 
 ### Team
 
-Jackie Glasheen: jglasheen
+Jackie Glasheen, Harsh Vardhan Pachisia, Shwetha Srinivasan, and Jennifer Yeaton
 
-Harsh Vardhan Pachisia: harshpachisia
+### Description
 
-Shwetha Srinivasan: shwetha
+**42.63 %** of natural disasters in the United States since 1980 have taken place over the last 12 years. It is evident that climate change and global warming are causing extreme weather disasters to occur with more frequency. The increase in natural disaster declarations over the past several decades has led to the federal government spending more money on disaster relief.  Due to the increase in frequency and cost of such disasters,  a Congressional Research Service [paper](https://sgp.fas.org/crs/homesec/R45484.pdf) from January 2022 suggests that Congress may consider limiting federal disaster relief spending. 
 
-Jennifer Yeaton: jkyeaton
+Disasters impact different geographic regions, affecting specific states and communities disproportionately. Through this dashboard, we hope to shed light on the increase in disaster frequency and cost nationwide. We highlight the specific situation in each state, where its political stance is on climate, and showcase the top ten counties that are at most risk based on socio-economic factors and require long term policy action. 
 
-### Description
-- Why this problem matters. What did we do about it? 
--  Major insights. 
-- How can this be helpful? 
+This dashboard was built using Python and Dash, using [climate disaster](https://www.fema.gov/openfema-data-page/disaster-declarations-summaries-v2) and [public assistance](https://www.fema.gov/openfema-data-page/public-assistance-funded-project-summaries-v1) data from the Federal Emergency Management Agency, scraped data of the voting patterns on the Climate IRA bill, and the [Census API](https://www.census.gov/data/developers/data-sets.html) for county-level socio-economic breakdown. 
 
 ### Runing the project
-- Add steps that the person needes to follow to run the entire project.
 
-### Example Insights (Add charts and screenshots of graphics). 
-- 
+This project can be run using Poetry. The steps are outlined below.
+
+1. Make a clone of the project repository
+
+2. Go to the project directory: `cd 30122-project-snow-lm`
+
+3. From the directory install virtual environment and dependencies: `poetry install`
+
+4. Activate the virtual environment: `poetry shell`
+
+5. Create an environment variable for the U.S. Census API key: `export CENSUS_API_KEY=YOUR_KEY_HERE`
+
+6. Run the project: `python3 -m snowlm`
+
+It will ask you to open the dash application in the browser to see the interactive dashboard.
+
+Pressing Control+C in your terminal will close the application. 

pyproject.toml

@@ -4,14 +4,24 @@ version = "0.1.0"
 description = ""
 authors = ["Harsh Vardhan Pachisia <[email protected]>"]
 readme = "README.md"
-packages = [{include = "30122_project_snow_lm"}]
+packages = [{include = "snowlm"}]
 
 [tool.poetry.dependencies]
 python = "^3.8"
 pandas = "^1.5.3"
 requests = "^2.28.2"
 lxml = "^4.9.2"
 plotly-express = "^0.4.1"
+dash = "^2.8.1"
+plotly-geo = "^1.0.0"
+geopandas = "^0.12.2"
+pyshp = "^2.3.1"
+dash-bootstrap-components = "^1.4.0"
+cssselect = "^1.2.0"
+dash-table = "^5.0.0"
+dash-bootstrap-templates = "^1.0.8"
+black = "^23.1.0"
+pylint = "^2.16.4"
 
 
 [build-system]

snowlm/EDA/census_disaster_data_merged_eda.py

@@ -0,0 +1,50 @@
+'''
+CAPP 30122
+Team: Snow Laughing Matter
+Primary Author: Jennifer Yeaton
+
+Purpose: Exploratory data analysis of county list with disaster data and census data.
+    Merging list of counties with disaster data into census dataset.
+'''
+
+import pandas as pd
+import pprint
+from snowlm.scrape_api.census_api_query import api_query
+from snowlm.data.climate import get_cleaned_data, counties_affected_by_disasters
+
+
+def merge_data_census_disaster():
+    '''
+    Merge list of counties with disaster data into census dataset.
+
+    Inputs: 
+        None
+
+    Returns: 
+        Pandas dataframe with census and disaster data merged together.
+    '''
+
+    full_census_df = api_query()
+    cleaned_data_df = get_cleaned_data("disaster_declarations.csv")
+
+    counties_with_disaster_list = counties_affected_by_disasters(get_cleaned_data("disaster_declarations.csv", only_2000_onwards = True))
+    column_name = ["state_and_county_code"]
+    counties_with_disaster_df = pd.DataFrame(columns = column_name, data = counties_with_disaster_list)
+
+    #Check data types for county_code:
+    print(full_census_df.dtypes)
+    print(counties_with_disaster_df.dtypes)
+
+    #Look at head of each dataset:
+    print(full_census_df.head)
+    print(counties_with_disaster_df.head)
+
+    #Add binary variable for whether disaster data exists:
+    counties_with_disaster_df["county_has_disaster_data"] = 1
+
+    #Merge datasets
+    census_data_w_disaster_counties_df = full_census_df.merge(counties_with_disaster_df, 
+        how = "left", on = "state_and_county_code")
+
+    #Return counties_with_disaster_df, full_census_df
+    return census_data_w_disaster_counties_df