Skip to content

Commit ed59fbc

Browse files
clizarraga-UAD7clizarraga-UAD7
committed
Updated PromptEng
1 parent 3cbb35d commit ed59fbc

File tree

3 files changed

+393
-0
lines changed

3 files changed

+393
-0
lines changed

‎docs/#mlpaths.md#‎

Lines changed: 389 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,389 @@
1+
2+
## Data Science Learning Path
3+
4+
We present 12 topics in the data science learning path, providing learning objectives, related skills, subtopics, and references/resources for each. The goal is to give graduate students a structured and comprehensive program to acquire data science expertise, including hands-on experience with real-world open-source tools and libraries.
5+
6+
```mermaid
7+
timeline
8+
title Machine Learning Learning Path
9+
A. General Data Science : 1. Introduction to Data Science and Machine Learning
10+
: 2. Python for Data Science
11+
: 3. Ethical Considerations in Data Science
12+
B. Statistics : 4. Statistical Learning and Regression Models
13+
C. Classical Machine Learning : 5. Classification Algorithms
14+
: 6. Ensemble Methods
15+
: 7. Unsupervised Learning
16+
D. Deep Learning : 8. Introduction to Deep Learning
17+
: 9. Recurrent Neural Networks and Sequence Models
18+
: 10. Generative Models
19+
: 11. Transfer Learning and Fine-tuning
20+
E. Continuous Integration / Continuous Deployment : 12. Model Deployment and Productionization
21+
22+
```
23+
24+
25+
### A: General Data Science
26+
27+
#### 1. Introduction to Data Science and Machine Learning
28+
29+
??? note "Topic description"
30+
31+
**Learning Objective**: Understand the fundamental concepts of data science and machine learning, and their real-world applications.
32+
33+
**Related Skills**:
34+
35+
- Defining and framing data science problems
36+
- Identifying appropriate machine learning techniques for different tasks
37+
- Distinguishing between supervised and unsupervised learning
38+
39+
**Subtopics**:
40+
41+
- Definition and scope of data science: [Lies, Damned Lies, and Data Science](https://beabytes.com/data-science-lies/). Béatrice Moissinac.
42+
- Overview of machine learning algorithms (regression, classification, clustering): [Introduction to Machine Learning](https://developers.google.com/machine-learning/intro-to-ml). Developers Google.
43+
- Applications of data science in various industries (e.g., healthcare, finance, marketing): [Data Science Applications Across 10 Different Industries](https://csweb.rice.edu/academics/graduate-programs/online-mds/blog/data-science-industry-applications). Rice University.
44+
- Ethical considerations in data science: [A Guide for Ethical Data Science](https://rss.org.uk/RSS/media/News-and-publications/Publications/Reports%20and%20guides/A-Guide-for-Ethical-Data-Science-Final-Oct-2019.pdf). Royal Statistical Society.
45+
- Hands-on introduction to machine learning using Python and scikit-learn: [Machine Learning Crash Course](https://developers.google.com/machine-learning/crash-course). Google Developers.
46+
47+
**References and Resources**:
48+
49+
- [Data Science; Concepts and Practice](https://asolanki.co.in/wp-content/uploads/2019/04/Data-Science-Concepts-and-Practice-2nd-Edition-3.pdf). V. Kotu and B. Deshpande.
50+
- [Data Science for Beginners - A curriculum](https://github.com/microsoft/Data-Science-For-Beginners/blob/main/README.md). Microsoft 10-week, 20-lesson curriculum all about Data Science.
51+
- [General Data Science Learning Resources](https://github.com/ua-data7/LearningResources/wiki/General-Data-Science). Data Science Institute, University of Arizona.
52+
53+
54+
55+
#### 2. Python for Data Science
56+
57+
??? note "Topic description"
58+
59+
**Learning Objective**: Develop proficiency in using Python for data manipulation, analysis, and visualization.
60+
61+
**Related Skills**:
62+
63+
- Mastering Python syntax and data structures
64+
- Utilizing NumPy for efficient numerical operations
65+
- Applying Pandas for data ingestion, cleaning, and transformation
66+
67+
**Subtopics**:
68+
69+
- Python programming basics (variables, data types, control structures, functions): [Chap 2.](https://wesmckinney.com/book/python-basics), and [Chap 3, McKinney](https://wesmckinney.com/book/python-builtin).
70+
- NumPy arrays and universal functions: [Chap 4. McKinney](https://wesmckinney.com/book/numpy-basics)
71+
- Pandas DataFrames and Series for data manipulation: [Chap 5.](https://wesmckinney.com/book/pandas-basics), [Chap 6.](https://wesmckinney.com/book/accessing-data), and [Chap 7., McKinney](https://wesmckinney.com/book/data-cleaning)
72+
- Data visualization with Matplotlib and Seaborn: [Matplotlib tutorials](https://matplotlib.org/stable/tutorials/index.html), and [Seaborn tutorials](https://seaborn.pydata.org/tutorial.html).
73+
- Integrating Python with data science libraries (scikit-learn, TensorFlow, PyTorch)
74+
75+
**References and Resources**:
76+
77+
- [Python for Data Analysis, 3E](https://wesmckinney.com/book/). Wes McKinney.
78+
- [Python Data Science Handbook](https://jakevdp.github.io/PythonDataScienceHandbook/). Jake VanderPlas.
79+
- [Data Visualization: A practical introduction](https://socviz.co/index.html#preface). Kieran Healy.
80+
- [Fundamentals of Data Visualization](https://clauswilke.com/dataviz/). Claus O. Wilke.
81+
- [Python Programming Language Learning Resources](https://github.com/ua-data7/LearningResources/wiki/Python-Programming-Language). Data Science Institute, University of Arizona.
82+
83+
84+
85+
#### 3. Ethical Considerations in Data Science
86+
87+
??? note "Topic description"
88+
89+
**Learning Objective**: Develop an understanding of the ethical implications and responsible practices in data science.
90+
91+
**Related Skills**:
92+
93+
- Identifying and mitigating bias in data and models
94+
- Ensuring fair and equitable decision-making
95+
- Protecting privacy and data security
96+
97+
**Subtopics**:
98+
99+
- Bias and fairness in machine learning
100+
- Interpretability and explainability of models
101+
- Privacy-preserving techniques (differential privacy, federated learning)
102+
- Data provenance and provenance tracking
103+
- Responsible AI principles and guidelines
104+
105+
**References and Resources**:
106+
107+
- "Ethical Algorithms" by Michael Kearns and Aaron Roth
108+
- "Artificial Intelligence: A Modern Approach" by Stuart Russell and Peter Norvig
109+
- Coursera course "AI Ethics" by DeepLearning.AI
110+
111+
112+
### B: Statistics
113+
114+
#### 4. Statistical Learning and Regression Models
115+
116+
??? note "Topic description"
117+
118+
**Learning Objective**: Understand and apply statistical learning techniques, with a focus on regression models.
119+
120+
**Related Skills**:
121+
122+
- Fitting and evaluating linear regression models
123+
- Applying logistic regression for classification tasks
124+
- Interpreting model coefficients and making predictions
125+
126+
**Subtopics**:
127+
128+
- Simple and multiple linear regression
129+
- Assumptions and diagnostics of linear regression
130+
- Logistic regression for binary classification
131+
- Evaluating model performance (R-squared, accuracy, precision, recall, F1-score)
132+
- Regularization techniques (Ridge, Lasso, Elastic Net)
133+
134+
**References and Resources**:
135+
136+
- "An Introduction to Statistical Learning" by Gareth James et al.
137+
- "Pattern Recognition and Machine Learning" by Christopher Bishop
138+
- Coursera course "Machine Learning" by Andrew Ng
139+
140+
141+
### C: Classical Machine Learning
142+
143+
#### 5. Classification Algorithms
144+
145+
??? note "Topic description"
146+
147+
**Learning Objective**: Acquire knowledge of various classification algorithms and their application in real-world problems.
148+
149+
**Related Skills**:
150+
151+
- Implementing and evaluating decision tree classifiers
152+
- Applying k-nearest neighbors for classification
153+
- Understanding the principles of support vector machines
154+
155+
**Subtopics**:
156+
157+
- Decision tree classification
158+
- K-nearest neighbors (KNN) algorithm
159+
- Support vector machines (SVMs)
160+
- Evaluating classification models (accuracy, precision, recall, F1-score, ROC-AUC)
161+
- Handling class imbalance (oversampling, undersampling, SMOTE)
162+
163+
**References and Resources**:
164+
165+
- "Pattern Recognition and Machine Learning" by Christopher Bishop
166+
- "Hands-On Machine Learning with Scikit-Learn and TensorFlow" by Aurélien Géron
167+
- Udacity course "Intro to Machine Learning"
168+
169+
170+
#### 6. Ensemble Methods
171+
172+
??? note "Topic description"
173+
174+
**Learning Objective**: Explore ensemble techniques for improving the performance of machine learning models.
175+
176+
**Related Skills**:
177+
178+
- Implementing random forest algorithms
179+
- Understanding the principles of gradient boosting
180+
- Applying bagging and boosting techniques to enhance model accuracy
181+
182+
**Subtopics**:
183+
184+
- Random forest classification and regression
185+
- Gradient boosting with XGBoost and LightGBM
186+
- Bagging and boosting (AdaBoost, Gradient Boosting)
187+
- Hyperparameter tuning for ensemble methods
188+
- Feature importance and interpretation in ensemble models
189+
190+
**References and Resources**:
191+
192+
- "Hands-On Machine Learning with Scikit-Learn and TensorFlow" by Aurélien Géron
193+
- "Introduction to Statistical Learning" by Gareth James et al.
194+
- Kaggle micro-course on "Advanced Ensembling"
195+
196+
197+
#### 7. Unsupervised Learning
198+
199+
??? note "Topic description"
200+
201+
**Learning Objective**: Gain proficiency in unsupervised learning techniques for data exploration and pattern discovery.
202+
203+
**Related Skills**:
204+
205+
- Implementing K-means clustering algorithms
206+
- Applying principal component analysis (PCA) for dimensionality reduction
207+
- Identifying anomalies and outliers in data
208+
209+
**Subtopics**:
210+
211+
- K-means clustering
212+
- Hierarchical clustering
213+
- Principal component analysis (PCA)
214+
- Anomaly detection techniques (Isolation Forest, One-Class SVM)
215+
- Dimensionality reduction methods (t-SNE, UMAP)
216+
217+
**References and Resources**:
218+
219+
- "Pattern Recognition and Machine Learning" by Christopher Bishop
220+
- "Hands-On Unsupervised Learning Using Python" by Ankur Patel
221+
- Coursera course "Cluster Analysis in Data Mining" by University of Illinois
222+
223+
224+
### D: Deep Learning
225+
226+
#### 8. Introduction to Deep Learning
227+
228+
??? note "Topic description"
229+
230+
**Learning Objective**: Develop an understanding of the fundamental concepts and architectures of deep neural networks.
231+
232+
**Related Skills**:
233+
234+
- Constructing and training feedforward neural networks
235+
- Applying convolutional neural networks for image-related tasks
236+
- Selecting appropriate activation functions and optimization techniques
237+
238+
**Subtopics**:
239+
240+
- Artificial neural networks (ANNs) and their structure
241+
- Activation functions (sigmoid, ReLU, tanh)
242+
- Feedforward neural networks and their training
243+
- Convolutional neural networks (CNNs) for image recognition
244+
- Hyperparameter tuning and optimization techniques
245+
246+
**References and Resources**:
247+
248+
- "Deep Learning" by Ian Goodfellow, Yoshua Bengio, and Aaron Courville
249+
- "Deep Learning with Python" by François Chollet
250+
- Coursera course "Deep Learning Specialization" by deeplearning.ai
251+
252+
253+
#### 9. Recurrent Neural Networks and Sequence Models
254+
255+
??? note "Topic description"
256+
257+
**Learning Objective**: Understand the principles of recurrent neural networks and their applications in sequence-to-sequence problems.
258+
259+
**Related Skills**:
260+
261+
- Implementing LSTM and GRU models for sequence modeling
262+
- Applying recurrent neural networks for time series forecasting
263+
- Generating text and other sequential data using RNNs
264+
265+
**Subtopics**:
266+
267+
- Recurrent neural networks (RNNs)
268+
- Long short-term memory (LSTMs)
269+
- Gated recurrent units (GRUs)
270+
- Sequence-to-sequence modeling
271+
- Time series forecasting with RNNs
272+
273+
**References and Resources**:
274+
275+
- "Deep Learning for Time Series Forecasting" by Jason Brownlee
276+
- "Natural Language Processing with Python" by Steven Bird et al.
277+
- Coursera course "Sequence Models" by deeplearning.ai
278+
279+
280+
#### 10. Generative Models
281+
282+
??? note "Topic description"
283+
284+
**Learning Objective**: Explore generative models and their applications in synthesizing new data.
285+
286+
**Related Skills**:
287+
288+
- Implementing generative adversarial networks (GANs)
289+
- Applying variational autoencoders (VAEs) for image and text generation
290+
- Evaluating the performance of generative models
291+
292+
**Subtopics**:
293+
294+
- Generative adversarial networks (GANs)
295+
- Variational autoencoders (VAEs)
296+
- Generative modeling for images, text, and other data types
297+
- Evaluating generative models (Inception Score, FID, BLEU)
298+
- Applications of generative models (data augmentation, creative generation)
299+
300+
**References and Resources**:
301+
302+
- "Generative Adversarial Networks" by Ian Goodfellow et al.
303+
- "Variational Autoencoders" by Diederik Kingma and Max Welling
304+
- Deeplearning.ai course "Generative Adversarial Networks (GANs)"
305+
306+
307+
308+
#### 11. Transfer Learning and Fine-tuning
309+
310+
??? note "Topic description"
311+
312+
**Learning Objective**: Understand the principles of transfer learning and how to leverage pre-trained models for various tasks.
313+
314+
**Related Skills**:
315+
316+
- Applying feature extraction with pre-trained models
317+
- Finetuning pre-trained models for domain-specific tasks
318+
- Evaluating the performance of transfer learning approaches
319+
320+
**Subtopics**:
321+
322+
- Concept of transfer learning
323+
- Feature extraction using pre-trained models (e.g., VGG, ResNet, BERT)
324+
- Finetuning pre-trained models for specific applications
325+
- Domain adaptation and dataset shift
326+
- Evaluating transfer learning performance
327+
328+
**References and Resources**:
329+
330+
- "Transfer Learning with Deep Learning" by Sebastian Ruder
331+
- "Practical Deep Learning for Cloud, Mobile, and Edge" by Anirudh Koul et al.
332+
- Coursera course "Convolutional Neural Networks" by deeplearning.ai
333+
334+
335+
### E: Continuous Integration / Continuous Deployment
336+
337+
#### 12. Model Deployment and Productionization
338+
339+
??? note "Topic description"
340+
341+
**Learning Objective**: Gain knowledge on how to deploy and maintain machine learning models in production environments.
342+
343+
**Related Skills**:
344+
345+
- Containerizing models using Docker
346+
- Deploying models on cloud platforms (e.g., AWS, GCP, Azure)
347+
- Monitoring and maintaining production models
348+
349+
**Subtopics**:
350+
351+
- Containerization with Docker
352+
- Cloud deployment on AWS, GCP, and Azure
353+
- Serving models with Flask, FastAPI, or Streamlit
354+
- Model monitoring and logging
355+
- Continuous integration and deployment (CI/CD) pipelines
356+
357+
**References and Resources**:
358+
359+
- "Deploying Machine Learning Models" by Abhishek Thakur
360+
- "Kubernetes in Action" by Marko Lukša
361+
- Coursera course "Machine Learning Engineering for Production (MLOps)" by deeplearning.ai
362+
363+
364+
***
365+
366+
367+
## Working with different data types.
368+
369+
Next you will find five specialized data science learning paths that branch off from the core topics in the previous section. Each specialized path includes a learning objective, related skills, subtopics, and references/resources.
370+
371+
372+
```mermaid
373+
flowchart LR;
374+
A@{shape:processes, label: "Data types"}-->B["`**Working with Numeric and Categorical Data**`"];
375+
A-->C["`**Computer Vision and Image-based Learning**`"];
376+
A-->D["`**Time Series Analysis and Forecasting**`"];
377+
A-->E["`**Natural Language Processing**`"];
378+
A-->F["`**Speech and Audio Processing**`"];
379+
click B href "https://github.com/ua-datalab/mlpaths/wiki/Working-with-Numeric-and-Categorical-Data" "Open this in a new tab" _blank
380+
click C href "https://github.com/ua-datalab/mlpaths/wiki/Computer-Vision-and-Image%E2%80%90based-Learning" "Open this in a new tab" _blank
381+
click D href "https://github.com/ua-datalab/mlpaths/wiki/Time-Series-Analysis-and-Forecasting" "Open this in a new tab" _blank
382+
click E href "https://github.com/ua-datalab/mlpaths/wiki/Natural-Language-Processing" "Open this in a new tab" _blank
383+
click F href "https://github.com/ua-datalab/mlpaths/wiki/Computer-Vision-and-Image%E2%80%90based-Learning" "Open this in a new tab" _blank
384+
385+
386+
```
387+
388+
389+
:bookmark: [Prompt Engineering](PromptEng/prompteng.md)

‎docs/.#mlpaths.md‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1 @@
1+
carlos@UAD7-MacBook.local.12601:1747769125

0 commit comments

Comments
 (0)