Learning and Modifying Codes with ChatGPT

Language models like ChatGPT have become an integral part of my daily life, both personally and professionally, ever since I started using them. One particular area where I’ve seen a significant improvement is in my prompt engineering skills. Here is one of the ways I have been using ChatGPT to save time.

As a PhD student, I often encounter situations where I need to learn new concepts or techniques, especially when working on projects involving R or Python programming. Whether it’s writing a script or tackling a small-to-medium-scale task, my approach used to involve searching the internet for existing code snippets or examples. I preferred modifying an existing code to ensure efficiency, and in case modifications were needed, I relied on package examples rather than diving deep into documentation (which I would read later for more detailed understanding). I simply ask ChatGPT to provide or create some examples for me. Not only does this save me valuable time that would have been spent on extensive googling, but it also allows me to seek ChatGPT’s assistance in optimizing and modifying my code.

Even if I don’t end up using the exact codes generated by ChatGPT, they still serve as valuable learning resources. I can utilize them to gain insights into Python or R packages and their functionalities. However, I most note that these codes often do not work and they are bugged. You need to make further prompts to make ChatGPT to correct it, or correct them by yourself. In most cases, you need to the latter.

Here is an example prompt and ChatGPT’s response below. You will see the answer is super clear and well explained.

Prompt: I have two shapefilesThehe first one is for the postcodes. The second one is the shapefile of parks in the city. I would like to find the centroid of each postcode and calculate the park density and the total number of parks within 1km distance. How can I do this in python?

Answer: To calculate the park density and the total number of parks within a 1km distance of each postcode centroid using shapefiles in Python, you can follow these steps:

1. Install the required packages:
   • geopandas: for handling shapefiles and spatial operations
   • shapely: for geometric calculations

   You can install these packages using pip:

   pip install geopandas shapely
   

2. Load the shapefiles:

   import geopandas as gpd
   
   # Load postcode shapefile
   postcodes = gpd.read_file('postcodes.shp')
   
   # Load parks shapefile
   parks = gpd.read_file('parks.shp')
   

3. Calculate the centroid of each postcode:

   postcodes['centroid'] = postcodes['geometry'].centroid
   

4. Iterate over each postcode and calculate park density and the total number of parks within a 1km distance:

   from shapely.geometry import Point
   
   # Create a buffer of 1km around each postcode centroid
   postcodes['buffer'] = postcodes['centroid'].buffer(1000)
   
   # Initialize lists to store results
   park_density = []
   total_parks = []
   
   for index, row in postcodes.iterrows():
       # Filter parks within the buffer
       parks_within_buffer = parks[parks.geometry.within(row['buffer'])]
   
       # Calculate park density (parks per square kilometer)
       park_area = parks_within_buffer.geometry.area.sum()
       postcode_area = row['geometry'].area
       density = len(parks_within_buffer) / (park_area / postcode_area)
   
       # Calculate the total number of parks within 1km distance
       num_parks = len(parks_within_buffer)
   
       # Append results to the lists
       park_density.append(density)
       total_parks.append(num_parks)
   
   # Add the results as new columns in the postcodes dataframe
   postcodes['park_density'] = park_density
   postcodes['total_parks'] = total_parks
   

5. You can now access the results in the postcodes dataframe, which will contain the original postcode shapefile with additional columns for park density and the total number of parks within a 1km distance.

Remember to replace 'postcodes.shp' and 'parks.shp' with the actual paths to your shapefiles.

P.S.: Of course, I got help from ChatGPT to write this post. What were you expecting?