Skip to content

Commit 7afe5d3

Browse files
committed
audit: fix lambda, iam notes
1 parent 49a32a4 commit 7afe5d3

62 files changed

Lines changed: 458 additions & 166 deletions

File tree

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.
-482 KB
Binary file not shown.
-1.6 MB
Binary file not shown.
Lines changed: 48 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,48 @@
1+
---
2+
tags:
3+
type:
4+
date: 2025-12-11
5+
done: false
6+
---
7+
A Shift from Rule based Systems with input and output to be consistent towards ML systems. These systems doesn't require rules, as they are probabilistic, based on patterns, public datasets.
8+
9+
**AI** in general is whatever systems that's built to assist human effort, it even includes basic calculators, rule based systems to self driving cars and robotic automations
10+
11+
**Machine Learning** - Enable computers to learn from data patterns through various algorithms, decision trees, vector machine, clustering. Usage includes spam detection, fraud detections, predictive maintenance. Ideal for simple predictive analysis, such as "will this person buy this"
12+
13+
**Deep learning** - Learn representation of data across various level of abstractions. Suited for large scale system that requires precise predictions.
14+
15+
# Inferencing
16+
The process of trained AI systems that is making predictions or classification based on the input
17+
18+
real time inferencing - chatbots, emails, autonomous driving systems. **AWS Sage Maker*** provides endpoints for deploying models, that performs real ṇime inferencing
19+
20+
batch inferencing - Sentiment analysing where real time analysis isn't required. **AWS Sage Maker** transforms jobs for applying models for datasets in S3
21+
22+
# Data Types in AI
23+
## Numerical
24+
AWS SageMaker provides us with various integrations for S3 and Redshift to integrate and process data. Numerical data - often integers, floating point numbers - represents mostly on machine learning models where regression analysis is performed to make a prediction
25+
## Categorical
26+
With techniques such as [[one hot encoding]], [[label encoding]] data can be transformed into numberical categories, as AI models are not gonna handle it itself
27+
## unstructured (Text, Images)
28+
Unstructured data often require a lot of complex pattern matching to convert as they have lack of predefined models.
29+
30+
High quality pre-processed data produces better performance
31+
32+
Things like book and conversations need to be pre-processed with techniques like tokenisation and [[stop word removal]] which prepares text data for model training. so it can get to NLP model and sentimental analysis.
33+
34+
As for image data, pre processing techinques such as normalisation, augmentation for preprocessing prepares the data for NLP models. **AWS Rekognition** the service for working with image data
35+
36+
As for audio data which is unlike text and image data which are monotonic, audio data has variying ambiguity. **AWS Transcribe** is the service that performs operations on audio data
37+
38+
# Data Preprocessing
39+
A clean structured data means accurate and performant models. So raw data are to be processed with various processes such as encoding, scaling and cleaning.
40+
41+
# Labelled vs Unlabelled
42+
Supervised learning will use labelled data for training. Great for tasks like classification
43+
Unsupervised learning will use unlableled data for training
44+
# Time series data
45+
> Data over time
46+
47+
Tools like **AWS Forecast** will use [forecasting algorithms](https://docs.aws.amazon.com/forecast/latest/dg/aws-forecast-recipe-prophet.html) on top of timeseries data to make predictions
48+

aws/Certification Learning Path.md renamed to aws/certifications/Certification Learning Path.md

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -5,5 +5,4 @@ type:
55
date: 19th July 2025
66
done: false
77
---
8-
98
[Path](https://d1.awsstatic.com/onedam/marketing-channels/website/aws/en_US/certification/approved/pdfs/AWS_certification_paths.pdf)

aws/cloud-formation/Helper scripts.md

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -5,7 +5,6 @@ type:
55
date: 19th July 2025
66
done: false
77
---
8-
98
# Cloudformation helper scripts
109
- How do we know if user data script ran successfully?
1110
- How do we make user data readable

aws/cloud-formation/Stack policy.md

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -5,5 +5,4 @@ type:
55
date: 19th July 2025
66
done: false
77
---
8-
98
> Prevents stack RESOURCES updated or deleted accidentally

aws/cloud-formation/cloudformation lore.md

Lines changed: 0 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -5,7 +5,6 @@ type:
55
date: 19th July 2025
66
done: false
77
---
8-
98
### 🔧 Core Functions:
109
- `!Ref` → get value of param or resource
1110
- `!GetAtt` → get attribute (e.g. `!GetAtt EC2Instance.PublicIp`)
@@ -33,5 +32,3 @@ done: false
3332
- `Resources` → actual AWS stuff
3433
- `Outputs` → expose values
3534
- `Mappings`, `Conditions`, `Metadata`, `Rules`
36-
37-
Master these, and you're 90% CloudFormation-ready.

aws/cloud-formation/cloudformation-custom-resources.md

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -5,7 +5,6 @@ type:
55
date: 19th July 2025
66
done: false
77
---
8-
98
There are some niche services that might not be covered by cloudformation resources.
109

1110
These custom resources are the one that we use as workaround, more like terraform providers

aws/cloud-formation/deletion policy, stack policy, termination policy.md

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -5,7 +5,6 @@ type:
55
date: 19th July 2025
66
done: false
77
---
8-
98
# Deletion Policy
109
- Delete
1110
- Snapshot (Supported services only)

aws/cloud-formation/mappings.md

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -5,7 +5,6 @@ type:
55
date: 19th July 2025
66
done: false
77
---
8-
98
The locals for cfn, hardcoded
109

1110
```yaml

0 commit comments

Comments
 (0)