讲解 Project 5: Simulating Elephant Population Management讲解 Python编程

CS152

Project 5: Simulating Elephant Population Management

As noted in the lab, this is the first of a two-part project where we'll be simulating the elephant population in Kruger National Park, South Africa.

The carrying capacity of the park is approximately 7000 elephants (1 elephant per square mile of park). Previous efforts to manage the population involved culling approximately 400 animals per year. After the development of an elephant contraceptive, the current effort to manage the population involves using a contraceptive dart on adult female elephants to limit the birth rate.

The conceptual goal of this week's simulation is twofold.

First, identify differences in the population distributions between the two methods of controlling the population.

Second, identify the percentage of adult females that need to be darted with the contraceptive each year in order to maintain a stable population without culling.

Following the overall design below, continue to develop the simulation step by step creating one function at a time and testing it with the provided “test” scripts. By the end of the project you will  have a program which will allow you to test the effect of different percentages of darting on the total elephant population. For this project we will employ “trial and error” to get the best darting probability. In the next project, we will utilize optimization to programmatically find the optimum darting percentage.

Project Tasks (1-9)

Task 1: Write a function to calculate which elephants survive a year

In the lab, you created the newElephant,  initPopulation and incrementAge functions. The next function to create is calcSurvival. The calcSurvival function processes the elephant population and determines whether each individual elephant survives to the next year.

The function takes the parameter list and population list as arguments. The function uses the parameters max age and the three survival probabilities (calf, adult, senior).

The function should loop over the existing population list and add each elephant to a new population list if it survives, using the elephant's age to determine which probability to use to determine survival. It should return the new population list. To do this:

1.   Create a new empty list, new_population.

2.   Loop over the existing population list. Use the age of the elephant to determine which survival probability applies, then use the appropriate survival probability to see if the elephant should be added to the new_population list. Do this by testing if a call to random.random() is less than the survival probability.

3.  After the loop, return the new_population list.

4.   Test your function by downloading test_calcSurvival.py, reading the code to make sure you understand it. Read the comments to make sure you know what output to expect.

Run it, examine the output, and make any necessary changes to your code.

Task 2: Write a function to randomly dart female elephants

Write a function, dartElephants, that goes through the adult females and randomly selects individuals for darting based on the comparison of random.random() and the darting probability parameter. Make sure that you generate a new random number for each comparison. The function takes in the parameter list and the population list as arguments. It returns the updated population list. The function makes use of the probability of darting, the juvenile age, and the maximum age.

1.   The function should loop over each elephant in the population list. Inside the loop, it

should test if the elephant is female and is older than the juvenile age and is less than or equal to the maximum reproductive age.

2.   If the elephant is an appropriate age, then determine if the elephant should be darted by testing if a call to random.random() is less than the probability of darting.

3.   If the animal should be darted, set its pregnancy field to 0 and set the months of contraceptive left field to 22.

4.  After the loop, return the population list.

5.   Test your dartElephants function to make sure some fraction of the adult females are being modified correctly. You may use test_dartElephants.py

Task 3: Write a function to cull elephants from the population

Write a function, cullElephants, that checks if there are more elephants than the carrying capacity. If there are too many elephants, the function should remove enough randomly chosen elephants from the population so there are as many elephants as the carrying capacity. The function takes in the parameter list and a population list as arguments. It should return a tuple with the first element being the new population list and second element being the number of elephants culled. The function makes use of the carrying capacity of the population.

1.   Determine the number of animals that need to be culled (that is the total number of animals in the population minus the carrying capacity).

2.   If there are too many elephants, then randomly shuffle the population list (use random.shuffle()). Use the appropriate list-slice to keep the first carryingCapacity number of animals.

3.   Return the tuple  like this:

return (newPopulation, numCulled)

4.   Use the  test_cullElephants.py file to test your cullElephants function.

Task 4: Write a function to manage how to control the population

Write a function, controlPopulation, that will determine whether the population should be darted or culled and call the appropriate function. It then returns the new population list and the number culled (which will be zero, if the elephants were darted) as a tuple.

The following is the structure of the controlPopulation function:

Test your controlPopulation function to make sure the list is being modified correctly.

This is just a repeat of the individual tests you did on cullElephants and dartElephants making sure that the logic of controlPopulation is working correctly. You may use test_controlPopulation.py

Task 5: Write a function to simulate a month

Note: This is the number one function that causes problems for students, so think carefully and methodically about your logic. Using a case by case analysis will go a long way in combating the complexity of this task.

Also, as a general practice you should NOT add or remove elements from a list that you  are iterating over (Remember the homework program on lists?) You should create a new population list and add all elephants to it AND add any new calves.

The function simulateMonth moves the simulation forward by one month. It modifies only the adult females in the population, and it adds a new calf to the population if the female has reached the end of her gestation. In the other cases, the function will either add an additional month to the number months pregnant variable, or reduce by one the number of months contraception remaining.

The function takes in the parameter list and the population list. It returns a new population list. The function uses the calving interval, juvenile age, and maximum age parameters.

1.   Loop over the population list. The algorithm inside the loop is given below. After the loop is complete, return the population list.

2.   On a month to month basis the probability of a female elephant (not already pregnant and not on contraceptive) becoming pregnant is 1.0 / (3.1*12 - 22) where 3.1 is the     calving interval, and 22 is the length of gestation in months.

3.   Test your function using test_simulateMonth.py

Task 6:  Write a function to simulate a year

The simulateYear function takes in the parameter list and population list. It calls

calcSurvival, then it calls incrementAge, then it loops twelve times calling

simulateMonth. Finally, it returns the population list. When calling each of the helper functions, pass in the population list and assign the result of the function back to the population list.

For example:

population = calc Survival ( parameters, population )

Test your function using test_simulateYear.py It will ensure that the basic functionality is present. We will test it more thoroughly once more functions have been written.

Task 7: Write a function to calculate statistics on the simulation

The calcResults function calculates how many calves, juveniles, adult males, adult females,  and seniors are in the population. It then returns a list with those values in it, along with the total number in the population, and the number culled from the population that year. It takes as input  the parameters, the population list, and the number of animals that were just culled from the population.

1.   Get the juvenile age and max age parameters. Initialize variables to hold the number of calves, juveniles, adult males, adult females, and seniors.

2.   Then loop over the population list. For each elephant, increment the appropriate variable (use if statements to figure out the category for each elephant).

3.  When the loop is complete, return a list with the following items: the total population size, the number of calves, the number of juveniles, the number of adult males, the number of adult females, the number of seniors, and the number of animals culled.

4.   Test your function using test_calcResults.py.

Task 8: Run a simulation

The runSimulation function takes one parameter–the parameter list. Then it creates the new population, applies any control procedures (e.g. darts them), loops over N years, simulating the year, and it keeps track of the demographics for each year by appending them to a list. Use  the following code as a template and add the function to your elephant.py script.

Note: you may need to add a parameter to your parameter list (and indexes) that specifies the number of years to run the simulation. The default number of years to simulate is 200. Use

IDXNumYears for the index.

Task 9: Write your main function

Setup your main function and conditional statement to run elephant. main(argv) should take argv as an argument and the call to main should pass in sys.argv. The only command line   argument you will need is the darting probability so the command line will look like this:

%python3 elephant.py .5

The first part of your main function should be a usage statement which stops program

execution if there are less than two command line arguments. Next, cast and assign the darting probability from the command line parameter float(argv[1]).

Next, set up the rest of the parameters and make the parameter list just as you did in the test    function. Next, call runSimulation once and store the return value in a list. Make sure you include the probDart parameter based on the command line argument.

Test your function and print out the last item in the results list. You may want to edit your

runSimulation function so it prints out the total population value each year of the simulation as a sanity check.

Finish your main function by calculating the average results (average total population, average number of calves, average number of juveniles, and so on). Print out those results nicely at the   end of your main function.

Required Element 1: Screenshot of console output of your program displaying the current darting probability and the results that were generated.

Follow-up Questions

1.   What is the difference between a tuple and a list?

2.  What are the benefits of using a naming scheme like IDXNumYears instead of just using an index number?

3.   Discuss where errors might impact the simulation? How would you ensure the final simulation model would produce reliable results to support well-informed wildlife

conservation decisions about this elephant population?

Extensions: Extensions are your opportunity to customize your project, learn

something else of interest to you, and improve your grade. The following are some suggested extensions, but you are free to choose your own. Be sure to describe any extensions you complete in your report.

1.   Run simulations and compare results. Run simulations that use the culling

population-control strategy (set the darted probability parameter to 0.0). Determine how the population demographics are different in this scenario compared to using contraception. Describe those differences. What are the average numbers of calves, juveniles, adult males, adult females, and seniors? Where are the most obvious differences in demographics between the two groups (think like a tourist). Does the percent to be darted change with different carrying capacities?

2.   How sensitive is the system to changes in various variables? For example, modify the  calf survival rate to 80% or up to 90% and see how that modifies the culling rate or the percent darted rate.

3.   Use your stats.py file from Project 3 to compute the average values of each

demographic. The list of results that you accrue when running multiple simulations can be thought of as the contents of a spreadsheet. Each list is a row, and each element in the list belongs in a column. The first column is for the total population, the second

column is for the count of the calf population, etc. You may want to add a function that rearranges data from lists of rows to lists of columns, then you will be able to use your existing functions in stats.py to compute the average value of a given column.

4.  Write the population data to a CSV file and generate plots of total population--or demographic subsets--under different scenarios.

5.   Expand on the number of command line arguments that your program can accept.

Write your project report

Please read the grading rubric so that you know exactly what criteria your code and report will be evaluated upon. If you have any questions, ask!

Reports are not included in the compressed code archive file! Please don’t make the graders hunt for your report.

You can write your report in any word processor you like and submit a PDF document in the Google Classroom assignment folder. Or use a Google Document format.

Review the Writeup Guidelines document in the Labs and Projects folder.

Your intended audience for your report is your peers who are not taking CS classes.

From week to week, you can assume your audience has read your prior reports. Your goal should be to explain to peers what you accomplished in the project and to give

them a sense of how you did it. The following is a list and description of the mandatory sections you must include in your report. Do not include the descriptions in your report,  but use them as a guide in writing your report.

●   Abstract

A summary of the project, in your own words. This should be no more than a few sentences. Give the reader context and identify the key purpose of the assignment. An abstract should define the project's key lecture concepts in your   own words for a general, non-CS audience. It should also describe the program's context and output, highlighting a couple of important algorithmic and/or scientific details. Writing an effective abstract is an important skill. Consider the following questions while writing it.

○   Does it describe the CS concepts of the project (e.g. writing well-organized and efficient code)?

○   Does it describe the specific project application (e.g. generating data)?

○   Does it describe your solution and how it was developed (e.g. what code did you write)?

○   Does it describe the results or outputs (e.g. did your code work as expected and what did the results tell you)?

○   Is it concise?

○   Are all of the terms well-defined?

○   Does it read logically and in the proper order?

●   Methods

The method section should describe in clear sentences (without pasting any

code) at least one example of your own computational thinking that helped you complete your project. This could involve illustrating how a key lecture concept was applied to creating an image, how you solved a challenging problem, or explaining an algorithmic feature that is essential to your program as well as why it is so essential. The explanation should be suitable for a general audience who  does not know Python.

●   Results

Present your results in a clear manner using human-friendly images or graphs labeled with captions and interpreted for a general audience such as your peers not in the course. Explain, for a general, non-CS audience, what your output means and whether it makes sense.

●   Reflection and Follow-up questions

Draw connections between lecture concepts utilized in this project and real-world problems that interest you. How else could these concepts apply to our everyday lives? What are some specific things you had to learn or discover in order to complete the project? Look for a set of short answer questions in this section of the report template.

●   Extensions (Required even if you did not do any)

A description of any extensions you undertook, including text output or images demonstrating those extensions. If you added any modules, functions, or other design components, note their structure and the algorithms you used.

●   References/Acknowledgements  (Required even if there are none)

Identify your collaborators, including TAs and professors. Include in that list anyone whose code you may have seen, such as those of friends who have taken the course in a previous semester. Cite any other sources, imported    libraries, or tutorials you used to complete the project.

Submit your Code

Turn in your code by zipping the file and uploading it to Google Classroom assignment folder. When submitting your code, double check the following.

1.   Is your name at the top of each Python file?

2.   Does every function have a docstring (‘’’ ‘’’) specifying what it does?

3.   Have you checked to make sure you have included all required elements and outputs in your project report?

4.   If you have done an extension, have you included this information in your report under  the Extension heading? Even if you have not done any extensions, include a section in your report where you state this.

5.   Have you acknowledged any help you may have received from classmates, your

instructor, the TAs, or outside sources (internet, books, videos, etc.)? If you received no help at all, have you indicated that under the Sources heading of the report?