-
Notifications
You must be signed in to change notification settings - Fork 0
/
course.qmd
114 lines (86 loc) · 8.19 KB
/
course.qmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
---
title: "Course"
bibliography: references.bib
csl: ieee.csl
link-citations: true
---
You will find below course syllabus and schedule for _AI-839: ML Production Engineering_ course offered at [IIIT-B](https://www.iiitb.ac.in/) in the Fall of 2024. This will evolve as the course progress.
### Syllabus & Schedule
::: {style="font-size: 80%;"}
| Wk | Dt | Topics | Resources |
| :- | :- | :------- | :------- |
| 01 | 01-Aug | **Introduction** <br> 1. Course Intro <br> 2. ML in Production - Intro <br> | [W01-L01](./lectures/w01-l01.qmd) |
| | @sec-hw-01 | Due Tue, 13th, August, 2024, 11.59pm IST |
| 02 | 06-Aug <br> 09-Aug | **Discovery** <br> 1. Contd. _ML in Production - Intro_ <br> 2. Software 1.0 vs 2.0 <br> 3. Design Thinking, Project Cards <br> 4. MLOps Canvas & Tool Landscape | [W02-L01](./lectures/w02-l01.qmd) <br> [W02-L02](./lectures/w02-l02.qmd) |
| | @sec-hw-02 | Due Fri, 23rd, August, 2024, 11.59pm IST |
| 03 | 13-Aug <br> 16-Aug | **Models for Modeling** <br> 1. Kedro <br> 2. DevOps | [W03-L01](./lectures/w03-l01.qmd) <br> [W03-L02](./lectures/w03-l02.qmd) |
| Note:| 16-Aug | DevOps guest lecture by [Ravi](https://www.linkedin.com/in/ravismula/) |
| 04 | 20-Aug <br> 23-Aug | **Dev Setup & Data Monitoring** <br> 1. Kedro, PyTest, Ruff, Quartodoc <br> 2. Data Quality | [W04-L01](./lectures/w04-l01.qmd) <br> [W04-L02](./lectures/w04-l02.qmd) |
| | @sec-hw-03 <br> @sec-hw-04 | 11.59PM IST, Friday, 6th Sep, 2024. <br> 11.59PM IST, Friday, 16th Sep, 2024. |
| 05 | 27-Aug <br> 30-Aug | **Model Monitoring and Deployment** <br> 1. Monitoring Metrics, Tests, Label Generation <br> 2. Deployment with MLFlow | [W05-L01](./lectures/w05-l01.qmd) <br> [W05-L02](./lectures/w05-l02.qmd) |
| | | |
| 06 | 03-Sep <br> 06-Sep | **Evaluation and Governance** <br> 1. Hypothesis Tets, DoEs, Model Comparison <br> 2. The R4 Framework | [W06-L01](./lectures/w06-l01.qmd) <br> [W06-L02](./lectures/w06-l02.qmd) |
| | @sec-hw-05 <br> @sec-hw-06 | Due 11.59PM IST, Tuesday, 24th Sep, 2024 <br> Due 11.59PM IST, Friday, 18th Oct, 2024 |
: **Part-1: Essentials** {.striped}
:::
::: {style="font-size: 80%;"}
| Wk | Dt | Topics | Resources |
| :- | :- | :------- | :------- |
| 07 | 10-Sep <br> 13-Sep | **Scaling Laws, Sample Hardness** <br> 1. Sample Sizes, Active Learning, Scaling Laws <br> 2. Sample Hardness | [W07-L01](./lectures/w07-l01.qmd) <br> [W07-L02](./lectures/w07-l02.qmd) |
| | @sec-hw-minor | Due 11.59PM IST, Friday, 25th Oct, 2024 |
| 08 | 17-Sep <br> 20-Sep | **Sample Fitness, Guest Lecture** <br> 1. Likelihood Ratio, $\nu\text{-information}$ <br> 2. ML Platform | [W08-L01](./lectures/w08-l01.qmd) <br> Talk by Abhishek |
| | | |
| 09 | 24-Sep <br> 27-Sep | **UQ, Midterm** <br> 1. Uncertainty Quantification, Conformal prediction <br> 2. Midterm | [W09-L01](./lectures/w09-l01.qmd) <br> NA |
| | Midterm | in-class |
| 10 | 15-Oct <br> 18-Oct | **Edge Deployment, ML Platforms** <br> 1. Model Compression, Quantization <br> 2. Distributed Systems, Solving ML Eng. issues with Keras and TensorFlow | [Talk](./talks/EdgeDeployment_SrinivasRana.pdf) by Dr. Srinivas <br> [Talk](./talks/MLOps_Kalyan.pdf) by Kalyan |
| | | |
| 11 | 22-Oct <br> 25-Oct | **Causal ML, Robustness** <br> 1. DoWhy and DICE <br> 2. Gradients is all you need | [Talk](./talks/CausalML_dowhy_dice_Amit.pdf) by Dr. Amit <br> [W11-L02](./lectures/w11-l02.qmd) |
| | | |
| 12 | 29-Oct <br> 01-Nov | **Robustness** <br> 1. Securing ML <br> 2. Holiday | [Talk](./talks/AI_Security_Manoj_29102024.pdf) by Manoj Parmar <br> [Text Attacks](./talks/Text_Atttacks_Manoj.ipynb) <br> [Adversarial Attacks](./talks/Adversarial_Sample_Using_ART_Manoj.ipynb) |
| | | |
| 13 | 07-Nov <br> 08-Nov | **LLMs** <br> 1. LLMs Intro <br> 2. LLMOps <br> 3. Fullstack LLMs | [W13-L01](./lectures/w13-l01.qmd) <br> [W13-L02](./lectures/w13-l01.qmd) <br> [W13-L03](./lectures/w13-l03.qmd) |
| | | |
| 14 | 12-Nov <br> 15-Nov | **Talks** <br> 1. Lessons Learnt in Production <br> 2. Building datasets | [Talk](./talks/Venkata_ScribbleData_Nov_2024.pdf) by Dr. Venkata <br> [Talk](./talks/Puneet_DataLabeling.pdf) by Puneet |
| | | |
| 15 | 19-Nov <br> 22-Nov | **Meta Learning** <br> No Class | [W15-L01](./lectures/w15-l01.qmd) |
| | | |
| 16 | Last Week | **Student Presentations** | |
| 26-Nov <br> <br> 29-Nov <br> 30-Nov | Fairness & Bias <br> Machine Unlearning <br> In-class final presentations-1 <br> In-class final presentations-2 | [W16-L01](./lectures/w16-l01.qmd) <br> [W16-L02](./lectures/w16-l02.qmd) |
: **Part-2: Full Stack ML** {.striped}
:::
::: {style="font-size: 80%;"}
| Wk | Dt | Topics | Resources |
| :- | :- | :------- | :------- |
| 03 | 16-Aug | Intro to DevOps <br> [ML Engineering] | [Ravi Mula](https://www.linkedin.com/in/ravismula/) <br> Architect: [Sanketika](https://sanketika.in/) |
| 08 | 20-Sep | ML Platforms <br> [ML Engineering] | [Abhishek Choudhary](https://www.linkedin.com/in/abhishekch123/) <br> Co-Founder & CTO: [TrueFoundry](https://www.truefoundry.com/) |
| 10 | 15-Oct | Deploying on Edges <br> [ML Engineering] <br> [Deck](./talks/EdgeDeployment_SrinivasRana.pdf) | [Dr. Srinivas Rana](https://www.linkedin.com/in/srinivasrana/) <br> Sr ML Scientist: [Wadhwani AI](https://www.wadhwaniai.org/) |
| 10 | 18-Oct | MLOps @ Scale <br> [ML Engineering] <br> [Deck](./talks/MLOps_Kalyan.pdf) | [Kalyan Deepak](https://www.linkedin.com/in/kalyan-deepak-2304622a/) <br> Sr Staff Engineer, [LinkedIn](https://www.linkedin.com/) |
| 11 | 22-Oct | DoWhy, DiCE <br> [Causal ML] <br> [Deck](./talks/CausalML_dowhy_dice_Amit.pdf) | [Dr. Amit Sharma](https://www.amitsharma.in/) <br> Principal Researcher, [MSR](https://www.microsoft.com/en-us/research/lab/microsoft-research-india/) |
| 12 | 29-Oct | Securing AI <br> [Adversarial ML] | [Manojkumar Parmar](https://www.linkedin.com/in/manojkumarparmar/) <br> CEO & CTO AI Shield [Bosch AI Shield](https://www.linkedin.com/company/bosch-aishield/) |
| 14 | 12-Nov | Building LLMs in Production in Regulated Industry <br> [System Design] | [Dr. Venkata Pingali](https://www.linkedin.com/in/pingali/) <br> Co-Founder: [Scribble Data](https://www.scribbledata.io/) |
| 14 | 15-Nov | Problems and Solutions in Data Labelling <br> [Data Collection] | [Puneet Jindal](https://www.linkedin.com/in/puneetjindalisb/) <br> CEO, [Labellerr](https://www.linkedin.com/company/labellerr/)
: **Guest Lectures** {.striped}
:::
## Discussions
We will use WhatsApp group for (informal)discussions and alerts.
## References
1. \[book\] [ML Engineering](https://www.mlebook.com/wiki/doku.php), Andiry Burkov, 2019, LeanPub
2. \[book\] [Effective Data Science Infrastructure](https://www.manning.com/books/effective-data-science-infrastructure), Vile Tuulos, 2023, Manning
3. \[book\] [ML System Design](https://www.oreilly.com/library/view/designing-machine-learning/9781098107956/), Chip Huyen, 2023, O'Reilly
4. \[course\] [CS329S @ Stanford: ML Systems Design](https://stanford-cs329s.github.io/), Chip Huyen, 2022
5. \[course\] [MLOps](https://github.com/GokuMohandas/mlops-course), Chip Huyen, 2024
## Grading
- 40%: Six assignments
- 15%: Midterm mini project
- 20%: In-class midterm
- 25%: Capstone project
## Policies
- _Late Submissions_: \ All deadlines are due at on the date and time indicated on the course page. The penalties for late submission
are as follows:
- Late submissions not allowed (incur a zero) - except with prior approval or in valid exceptional cases
- _Make-up Exam/Submission Policy_: As per institute [policy](https://www.iiitb.ac.in/includefiles/userfiles/images/pdf/code-of-coduct.pdf)
- _Citation Policy for Papers_: Always mention the source, give full attribution and credits to citations, and as per institute [policy](https://www.iiitb.ac.in/includefiles/userfiles/images/pdf/code-of-coduct.pdf)
- _Academic Dishonesty/Plagiarism_: As per institute [policy](https://www.iiitb.ac.in/includefiles/userfiles/images/pdf/code-of-coduct.pdf)
- _Accommodation of Divyangs_: As per institute [policy](https://www.iiitb.ac.in/includefiles/userfiles/images/pdf/code-of-coduct.pdf)
[Soma S Dhavala](https://www.linkedin.com/in/somasdhavala/) \
Course Instructor