代写program、代做Java/Python程序设计
- 首页 >> C/C++编程 Project 1
Please remember to use complete sentences and explain all of your answers using Markdown. You may
work in groups of up to 2 people. The ffnal document must contain the names of everyone in the
group. You are expected to turn in a Jupyter Notebook ffle.
1. Overview
A sample of images of characters representing the numbers 1-10 are included, already processed. The goal is to
use various methods to categorize the images using at least 2 different methods OR create your own perceptron.
You must use 3 different levels of explained variance. Additionally, we want to compare and contrast these
methods in terms of time and how “good” the results. You must explain what you are doing in Markdown
before the code cells.
2. Introduction
Describe the goal, in your own words, what methods you are using, and why you believe they would be
appropriate and effective.
3. PCA
Create your covariance matrix, ffnd the eigenvalues and eigenvectors. I highly suggest saving these in two
different csv ffles rather than computing them repeatedly. Decide on what levels of explained variance you
want.
4. Two Supervised Learning Methods
Your two supervised learning methods must use binary (yes/no) classiffcations to classify any 51x51 pixel image
as a number between 1 and 10 with an unclassiffed category. Explain how you make each decision and show an
example with a completed model. You must use appropriate hyper-parameter tuning to decide the parameters
of the models that ffts well and avoids overfftting.
5. Results
Summarize the results, using a variety of metrics including accuracy to assess how these models perform.6. Grading
I. Organization
A. Are there appropriate headers?
B. Were complete sentences used?
C. Is the notebook clean looking with appropriate outputs in minimal locations?
D. Is the code commented?
II. Content
A. Were the guidelines followed?
B. Was hyper-parameter tuning done, with proper visualizations and computation?
C. Are explanations and descriptions proper and adequate?
D. Do results include a comparison between the models and levels of explained variance?
III. Style
A. Does the topic “ffow”?
B. Is there evidence of original thought?
C. Does it look rushed?
Please remember to use complete sentences and explain all of your answers using Markdown. You may
work in groups of up to 2 people. The ffnal document must contain the names of everyone in the
group. You are expected to turn in a Jupyter Notebook ffle.
1. Overview
A sample of images of characters representing the numbers 1-10 are included, already processed. The goal is to
use various methods to categorize the images using at least 2 different methods OR create your own perceptron.
You must use 3 different levels of explained variance. Additionally, we want to compare and contrast these
methods in terms of time and how “good” the results. You must explain what you are doing in Markdown
before the code cells.
2. Introduction
Describe the goal, in your own words, what methods you are using, and why you believe they would be
appropriate and effective.
3. PCA
Create your covariance matrix, ffnd the eigenvalues and eigenvectors. I highly suggest saving these in two
different csv ffles rather than computing them repeatedly. Decide on what levels of explained variance you
want.
4. Two Supervised Learning Methods
Your two supervised learning methods must use binary (yes/no) classiffcations to classify any 51x51 pixel image
as a number between 1 and 10 with an unclassiffed category. Explain how you make each decision and show an
example with a completed model. You must use appropriate hyper-parameter tuning to decide the parameters
of the models that ffts well and avoids overfftting.
5. Results
Summarize the results, using a variety of metrics including accuracy to assess how these models perform.6. Grading
I. Organization
A. Are there appropriate headers?
B. Were complete sentences used?
C. Is the notebook clean looking with appropriate outputs in minimal locations?
D. Is the code commented?
II. Content
A. Were the guidelines followed?
B. Was hyper-parameter tuning done, with proper visualizations and computation?
C. Are explanations and descriptions proper and adequate?
D. Do results include a comparison between the models and levels of explained variance?
III. Style
A. Does the topic “ffow”?
B. Is there evidence of original thought?
C. Does it look rushed?