Skip to content

Latest commit

 

History

History
312 lines (173 loc) · 16.9 KB

README.md

File metadata and controls

312 lines (173 loc) · 16.9 KB
Raw

Unit 13 Homework Assignment - The Power of the Cloud and Unsupervised Learning

Background

It is time to take what you have learned about unsupervised learning and the AWS services and apply it to new situations. For this assignment, you will need to complete one of two (not both) challenges. Which challenge you take on is your choice. Just be sure to give it your all -- as the skills you hone will become powerful tools in your FinTech tool belt.

Before You Begin

  1. Create a new repository for this project called unit13-challenge. Do not add this homework to an existing repository.

  2. Clone the new repository to your computer.

  3. Inside your local git repository, create a directory for the challenge assignment you choose. Use folder names corresponding to the challenges: RoboAdvisor or ClusteringCrypto.

  4. Add your solution files to this folder.

  5. Push the above changes to GitHub or GitLab.

Option 1: Robo Advisor for Retirement Plans

Robot

Photo by Alex Knight from Pexels | Free License

Background

You were hired as a digital transformation consultant by one of the most prominent retirement plan providers in the country; they want to increase their client portfolio, especially by engaging young people. Since machine learning and NLP are disrupting finance to improve customer experience, you decide to create a robo advisor that could be used by customers or potential new customers to get investment portfolio recommendations for retirement.

In this homework assignment, you will combine your new skills with Amazon Web Services with your already mastered Python superpowers, to create a bot that will recommend an investment portfolio for a retirement plan.

You are asked to accomplish the following main tasks:

  1. Initial Robo Advisor Configuration: Define an Amazon Lex bot with a single intent that establishes a conversation about the requirements to suggest an investment portfolio for retirement.

  2. Build and Test the Robo Advisor: Make sure that your bot is working and responding accurately along with the conversation with the user, by building and testing it.

  3. Enhance the Robo Advisor with an Amazon Lambda Function: Create an Amazon Lambda function that validates the user's input and returns the investment portfolio recommendation. This task includes testing the Amazon Lambda function and making the integration with the bot.

Files

Instructions

Initial Robo Advisor Configuration

In this section, you will create the RoboAdvisor bot and add an intent with its corresponding slots.

Sign in into your AWS Management Console and create a new custom Amazon Lex bot. Use the following parameters:

  • Bot name: RoboAdvisor
  • Output voice: Salli
  • Session timeout: 5 minutes
  • Sentiment analysis: No
  • COPPA: No

Create the RecommendPortfolio intent, and configure some sample utterances as follows (you can add more utterances at your own criteria):

  • I want to save money for my retirement
  • I'm ​{age}​ and I would like to invest for my retirement
  • I'm ​{age}​ and I want to invest for my retirement
  • I want the best option to invest for my retirement
  • I'm worried about my retirement
  • I want to invest for my retirement
  • I would like to invest for my retirement

This bot will use four slots, three using built-in types and one custom slot named riskLevel. Define the three initial slots as follows:

Name Slot Type Prompt
firstName AMAZON.US_FIRST_NAME Thank you for trusting on me to help, could you please give me your name?
age AMAZON.NUMBER How old are you?
investmentAmount AMAZON.NUMBER How much do you want to invest?

The riskLevel custom slot will be used to retrieve the risk level the user is willing to take on the investment portfolio; create this custom slot as follows:

  • Name: riskLevel
  • Prompt: What level of investment risk would you like to take?
  • Maximum number of retries: 2
  • Prompt response cards: 4

Configure the response cards for the riskLevel slot as is shown bellow:

Card 1 Card 2
Card 1 sample Card 2 sample
Card 3 Card 4
Card 3 sample Card 4 sample

Note: You can download free icons from this website or you can use the icons provided in the Icons directory.

Move to the Confirmation Prompt section, and set the following messages:

  • Confirm: Thanks, now I will look for the best investment portfolio for you.
  • Cancel: I will be pleased to assist you in the future.

Leave the error handling configuration for the RecommendPortfolio bot with the default values.

Error handling configuration

Build and Test the Robo Advisor

In this section, you will test your Robo Advisor. Build the bot and test it in the chatbot window. You should see a conversation like the one below.

Robo Advisor test

Enhance the Robo Advisor with an Amazon Lambda Function

In this section, you will create an Amazon Lambda function that will validate the data provided by the user on the Robo Advisor. Start by creating a new lambda function from scratch and name it recommendPortfolio. Select Python 3.7 as runtime.

Use the starter code provided on lambda_function.py and complete the recommend_portfolio() function by following these guidelines:

User Input Validation
  • The age should be greater than zero and less than 65.
  • the investment_amount should be equal to or greater than 5000.
Investment Portfolio Recommendation

Once the intent is fulfilled, the bot should response with an investment recommendation based on the selected risk level as follows:

  • none: "100% bonds (AGG), 0% equities (SPY)"
  • very low: "80% bonds (AGG), 20% equities (SPY)"
  • low: "60% bonds (AGG), 40% equities (SPY)"
  • medium: "40% bonds (AGG), 60% equities (SPY)"
  • high: "20% bonds (AGG), 80% equities (SPY)"
  • very high: "0% bonds (AGG), 100% equities (SPY)"

Be creative while coding your solution, you can have all the code on the recommend_portfolio() function, or you can split the functionality across different functions, put your Python coding skills in action!

Once you finish coding your lambda function, test it using the sample test cases provided for this homework.

After successfully testing your code, open the Amazon Lex Console and navigate to the RecommendPortfolio bot configuration, integrate your new lambda function by selecting it in the Lambda initialization and validation and Fulfillment sections. Build your bot, and you should have a conversation as follows.

Robo Advisor test with Lambda

Submission

You should create a brand new repository in GitHub and upload the following files to your repo.

  • A python script with your final lambda function.

  • From the Amazon Lex Console, export your bot, intent, and slot using Amazon Lex as the target platform, and upload the ZIP files to your repo.

  • Create a short video or animated GIF showing a demo of your Robo Advisor in action from the test window. Upload the video or animated GIF file to your repo.

Once you have uploaded all the files into the repo, post a link to your homework's repository in BootCamp Spot.

Hints

  • If you are using a Mac, you can create a screen-recording using the built-in QuickTime player. Follow this link to learn more.

  • If you are using Windows 10, you can create a screen-recording using the built-in Xbox Game Bar. Follow this link to learn more.

Option 2: Clustering Crypto

Cryptocurrencies coins Cryptocurrencies coins by Worldspectrum | Free License

Background

You are a Senior Manager at the Advisory Services team on a Big Four firm, and one of your most important clients, a prominent investment bank, is interested in offering a new cryptocurrencies investment portfolio for its customers, however, they are lost in the immense universe of cryptocurrencies, so they ask you to present a report of what cryptocurrencies are on the trading market and how cryptocurrencies could be grouped towards creating a classification for developing this new investment product.

In this homework assignment, you have the opportunity to put in action your new unsupervised learning and Amazon SageMaker skills to cluster cryptocurrencies and create some plots to present your results.

You are asked to accomplish the following main tasks:

Files

Instructions

Data Preprocessing

In this section, you have to load the information about cryptocurrencies from the provided CSV file and perform some data preprocessing tasks. The data was retrieved from CryptoCompare using this endpoint: https://min-api.cryptocompare.com/data/all/coinlist.

Start by loading the data in a Pandas DataFrame named crypto_df, and continue with the following data preprocessing tasks.

  1. Remove all cryptocurrencies that are not in trading.

  2. Remove all cryptocurrencies that do not have an algorithm defined.

  3. Remove the IsTrading column.

  4. Remove all cryptocurrencies with at least one null value.

  5. Remove all cryptocurrencies without coins mined.

  6. Store the names of all cryptocurrencies in a DataFrame named coins_name, use the crypto_df.index as the index for this new DataFrame.

  7. Remove the CoinName column.

  8. Create dummies variables for all the text features, store the resulting data on a DataFrame named X.

  9. Use the StandardScaler from sklearn to standardize all the data of the X DataFrame. Remember, this is important prior to using PCA and K-Means algorithms.

Reducing Data Dimensions Using PCA

Use the PCA algorithm from sklearn to reduce the dimensions of the X DataFrame down to three principal components.

Once you have reduced the data dimensions, create a DataFrame named pcs_df using as columns names "PC 1", "PC 2" and "PC 3"; use the crypto_df.index as the index for this new DataFrame.

You should have a DataFrame like the following.

pcs_df

Clustering Cryptocurrencies Using K-Means

In this section, you will use the KMeans algorithm from sklearn to cluster the cryptocurrencies using the PCA data.

Perform the following tasks:

  1. Create an Elbow Curve to find the best value for k, use the pcs_df DataFrame.

  2. Once you define the best value for k, run the Kmeans algorithm to predict the k clusters for the cryptocurrencies data. Use the pcs_df to run the KMeans algorithm.

  3. Create a new DataFrame named clustered_df, that includes the following columns "Algorithm", "ProofType", "TotalCoinsMined", "TotalCoinSupply", "PC 1", "PC 2", "PC 3", "CoinName", "Class". You should maintain the index of the crypto_df DataFrames as is shown bellow.

    clustered_df

Visualizing Results

In this section, you will create some data visualization to present the final results. Perform the following tasks:

  1. Create a 3D-Scatter using Plotly Express to plot the clusters using the clustered_df DataFrame. You should include the following parameters on the plot: hover_name="CoinName" and hover_data=["Algorithm"] to show this additional info on each data point.

  2. Use hvplot.table to create a data table with all the current tradable cryptocurrencies. The table should have the following columns: "CoinName", "Algorithm", "ProofType", "TotalCoinSupply", "TotalCoinsMined", "Class"

  3. Create a scatter plot using hvplot.scatter, to present the clustered data about cryptocurrencies having x="TotalCoinsMined" and y="TotalCoinSupply" to contrast the number of available coins versus the total number of mined coins. Use the hover_cols=["CoinName"] parameter to include the cryptocurrency name on each data point.

Optional Challenge

For the challenge section, you have to upload your Jupyter notebook to Amazon SageMaker and deploy it.

The hvplot and Plotly Express libraries are not included in the built-in anaconda environments, however you can install external libraries on Amazon SageMaker, there is still not full support for these libraries; so, for this challenge section, you should use the altair library instead.

Perform the following tasks:

  1. Upload your Jupyter notebook and rename it as crypto_clustering_sm.ipynb

  2. Select the conda_python3 environment.

  3. Install the altair library by running the following code before the initial imports.

!pip install -U altair

  1. Use the altair scatter plot to create the Elbow Curve.

  2. Use the altair scatter plot, instead of the 3D-Scatter from Plotly Express, to visualize the clusters. Since this is a 2D-Scatter, use x="PC 1" and y="PC 2" for the axes, and add the following columns as tool tips: "CoinName", "Algorithm", "TotalCoinsMined", "TotalCoinSupply".

  3. Use the altair scatter plot to visualize the tradable cryptocurrencies using x="TotalCoinsMined" and y="TotalCoinSupply" for the axes.

  4. Show the table of current tradable cryptocurrencies using the display() command.

  5. Remove all hvplot and Plotly Express references from your code.

Complementary Resources

Submission

  • Code your solution using the provided starter Jupyter notebook.

  • For the Challenge section, create a new Jupyter notebook named crypto_clustering_sm.ipynb and include the necessary code to import the additional required library.

  • Create and upload a repository with the above files to GitHub and post a link in BootCamp Spot.

© 2019 Trilogy Education Services, a 2U, Inc. brand. All Rights Reserved.