Merge branch 'iss183-adding-final-colabs'

contents/ai_for_good/ai_for_good.qmd

@@ -78,6 +78,15 @@ Widespread TinyML applications can help digitize smallholder farms to increase p
 
 With greater investment and integration into rural advisory services, TinyML could transform small-scale agriculture and improve livelihoods for farmers worldwide. The technology effectively brings the benefits of precision agriculture to disconnected regions most in need.
 
+:::{#exr-agri .callout-exercise collapse="true"}
+
+### Crop Yield Modeling
+
+This exercise teaches you how to predict crop yields in Nepal by combining satellite data (Sentinel-2), climate data (WorldClim), and on-the-ground measurements. You'll use a machine learning algorithm called XGBoost Regressor to build a model, split the data for training and testing, and fine-tune the model parameters for the best performance. This notebook lays the foundation to implement TinyML in the agriculture domain  – consider how you could adapt this process for smaller datasets, fewer features, and simplified models to make it compatible with the power and memory constraints of TinyML devices.  
+
+[![](https://colab.research.google.com/assets/colab-badge.png)](https://colab.research.google.com/github/developmentseed/sat-ml-training/blob/main/_notebooks/2020-07-29-Crop_yield_modeling_with_XGBoost.ipynb#scrollTo=GQd7ELsRWkBI)
+:::
+
 ## Healthcare
 
 ### Expanding Access
@@ -116,6 +125,15 @@ An on-device algorithm for early and timely life-threatening VA detection will i
 
 The champion, GaTech EIC Lab, obtained 0.972 in $F_\beta$ (F1 score with a higher weight to recall), 1.747 ms in latency and 26.39 kB in memory footprint with a deep neural network. An ICD with an on-device VA detection algorithm was [implanted in a clinical trial](https://youtu.be/vx2gWzAr85A?t=2359).
 
+:::{#exr-hc .callout-exercise collapse="true"}
+
+### Clinical Data:  Unlocking Insights with Named Entity Recognition
+
+In this exercise, you'll learn about Named Entity Recognition (NER), a powerful tool for extracting valuable information from clinical text. Using Spark NLP, a specialized library for healthcare NLP, we'll explore how NER models like BiLSTM-CNN-Char and BERT can automatically identify important medical entities such as diagnoses, medications, test results, and more. You'll get hands-on experience applying these techniques with a special focus on oncology-related data extraction, helping you unlock insights about cancer types and treatment details from patient records.  
+
+[![](https://colab.research.google.com/assets/colab-badge.png)](https://colab.research.google.com/github/JohnSnowLabs/spark-nlp-workshop/blob/master/tutorials/Certification_Trainings/Healthcare/1.Clinical_Named_Entity_Recognition_Model.ipynb#scrollTo=I08sFJYCxR0Z)
+:::
+
 ## Science
 
 In many scientific fields, researchers are limited by the quality and resolution of data they can collect. They often must infer the true parameters of interest indirectly, using approximate correlations and models built on sparse data points. This constrains the accuracy of scientific understanding and predictions.
@@ -232,9 +250,9 @@ These slides serve as a valuable tool for instructors to deliver lectures and fo
 :::{.callout-exercise collapse="false"}
 # Exercises
 
-To reinforce the concepts covered in this chapter, we have curated a set of exercises that challenge students to apply their knowledge and deepen their understanding. 
+- @exr-agri
 
-Coming soon.
+- @exr-hc
 :::
 
 :::{.callout-lab collapse="false"}

contents/benchmarking/benchmarking.qmd

@@ -141,7 +141,7 @@ Example: [DeepBench](https://github.com/baidu-research/DeepBench), introduced by
 
 Ever wondered how your image filters get so fast? Special libraries like cuDNN supercharge those calculations on certain hardware. In this Colab, we're gonna use cuDNN with PyTorch to speed up image filtering. Think of it like a tiny benchmark, showing how the right software can unlock your GPU's power!  
 
-[![](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/RyanHartzell/cudnn-image-filtering/blob/master/notebooks/CuDNN%20Image%20Filtering%20Tutorial%20Using%20PyTorch.ipynb#scrollTo=1sWeXdYsATrr)
+[![](https://colab.research.google.com/assets/colab-badge.png)](https://colab.research.google.com/github/RyanHartzell/cudnn-image-filtering/blob/master/notebooks/CuDNN%20Image%20Filtering%20Tutorial%20Using%20PyTorch.ipynb#scrollTo=1sWeXdYsATrr)
 
 :::
 
@@ -455,7 +455,7 @@ By measuring these metrics, we can assess the performance of the object detectio
 
 Get ready to put your AI models to the ultimate test! MLPerf is like the Olympics for machine learning performance. In this Colab, we'll use a toolkit called CK to run official MLPerf benchmarks, measure how fast and accurate your model is, and even use TVM to give it a super speed boost. Are you ready to see your model earn its medal?  
 
-[![](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1aywGlyD1ZRDtQTrQARVgL1882JcvmFK-?usp=sharing#scrollTo=tnyHAdErL72u)
+[![](https://colab.research.google.com/assets/colab-badge.png)](https://colab.research.google.com/drive/1aywGlyD1ZRDtQTrQARVgL1882JcvmFK-?usp=sharing#scrollTo=tnyHAdErL72u)
 
 :::
 

contents/data_engineering/data_engineering.qmd

@@ -127,7 +127,7 @@ In this context, using KWS as an example, we can break each of the steps out as
 
 Explore a hands-on guide for building and deploying Keyword Spotting (KWS) systems using TensorFlow Lite Micro. Follow steps from data collection to model training and finally deployment to microcontrollers. Learn to create efficient KWS models that recognize specific keywords amidst background noise. Perfect for those interested in machine learning on embedded systems. Unlock the potential of voice-enabled devices with TensorFlow Lite Micro!  
 
-[![](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/17I7GL8WTieGzXYKRtQM2FrFi3eLQIrOM)
+[![](https://colab.research.google.com/assets/colab-badge.png)](https://colab.research.google.com/drive/17I7GL8WTieGzXYKRtQM2FrFi3eLQIrOM)
 :::
 
 The current chapter underscores the essential role of data quality in ML, using Keyword Spotting (KWS) systems as an example. It outlines key steps, from problem definition to stakeholder engagement, emphasizing iterative feedback. The forthcoming chapter will delve deeper into data quality management, discussing its consequences and future trends, focusing on the importance of high-quality, diverse data in AI system development, addressing ethical considerations and data sourcing methods.
@@ -178,7 +178,7 @@ Web scraping can yield inconsistent or inaccurate data. For example, the photo i
 
 Discover the power of web scraping with Python using libraries like Beautiful Soup and Pandas. In this exercise, we'll scrape Python documentation for function names and descriptions, and explore NBA player stats. By the end, you'll have the skills to extract and analyze data from real-world websites. Ready to dive in? Access the Google Colab notebook below and start practicing!  
 
-[![](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Andy-Pham-72/Web-Scraping-with-BeautifulSoup-and-Pandas/blob/master/Web_scraping_with_beautiful_soup_and_pandas_complete.ipynb)
+[![](https://colab.research.google.com/assets/colab-badge.png)](https://colab.research.google.com/github/Andy-Pham-72/Web-Scraping-with-BeautifulSoup-and-Pandas/blob/master/Web_scraping_with_beautiful_soup_and_pandas_complete.ipynb)
 ::: 
 
 ### Crowdsourcing
@@ -222,7 +222,7 @@ While synthetic data offers numerous advantages, it is essential to use it judic
 ### Synthetic Data
 Let us learn about synthetic data generation using Generative Adversarial Networks (GANs) on tabular data. We'll take a hands-on approach, diving into the workings of the CTGAN model and applying it to the Synthea dataset from the healthcare domain. From data preprocessing to model training and evaluation, we'll go step-by-step, learning how to create synthetic data and assess its quality and unlock the potential of GANs for data augmentation and real-world applications.  
 
-[![](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1nwbvkg32sOUC69zATCfXOygFUBeo0dsx?usp=sharing#scrollTo=TkwYknr44eFn)
+[![](https://colab.research.google.com/assets/colab-badge.png)](https://colab.research.google.com/drive/1nwbvkg32sOUC69zATCfXOygFUBeo0dsx?usp=sharing#scrollTo=TkwYknr44eFn)
 :::
 
 ## Data Storage
@@ -320,7 +320,7 @@ There is a boom of data processing pipelines, these are commonly found in ML ope
 
 Let us explore two significant projects in speech data processing and machine learning. The Multilingual Spoken Words Corpus (MSWC) is a vast audio dataset with over 340,000 keywords and 23.4 million 1-second spoken examples. It's used in various applications like voice-enabled devices and call center automation. The Few-Shot Keyword Spotting project introduces a new approach for keyword spotting across different languages, achieving impressive results with minimal training data. We'll delve into the MSWC dataset, learn how to structure it effectively, and then train a few-shot keyword-spotting model. Let's get started!  
 
-[![](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/harvard-edge/multilingual_kws/blob/main/multilingual_kws_intro_tutorial.ipynb#scrollTo=ApnPyIlYNFYD)
+[![](https://colab.research.google.com/assets/colab-badge.png)](https://colab.research.google.com/github/harvard-edge/multilingual_kws/blob/main/multilingual_kws_intro_tutorial.ipynb#scrollTo=ApnPyIlYNFYD)
 :::
 
 ## Data Labeling

contents/dl_primer/dl_primer.qmd

@@ -116,13 +116,13 @@ In embedded AI systems, MLPs can function as compact models for simpler tasks li
 Get ready to dive into the exciting world of deep learning and TinyML! We've just covered the core building blocks of neural networks, from simple perceptrons to complex architectures. Now, you'll get to apply these concepts in practical examples.  In the provided Colab notebooks, you'll explore:
 
 **Predicting house prices:** Learn how neural networks can analyze housing data to estimate property values.  
-[![](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Mjrovai/UNIFEI-IESTI01-TinyML-2022.1/blob/main/00_Curse_Folder/1_Fundamentals/Class_07/TF_Boston_Housing_Regression.ipynb)
+[![](https://colab.research.google.com/assets/colab-badge.png)](https://colab.research.google.com/github/Mjrovai/UNIFEI-IESTI01-TinyML-2022.1/blob/main/00_Curse_Folder/1_Fundamentals/Class_07/TF_Boston_Housing_Regression.ipynb)
 
 **Image Classification:** Discover how to build a network capable of understanding the famous MNIST handwritten digit dataset.  
-[![](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Mjrovai/UNIFEI-IESTI01-TinyML-2022.1/blob/main/00_Curse_Folder/1_Fundamentals/Class_09/TF_MNIST_Classification_v2.ipynb)
+[![](https://colab.research.google.com/assets/colab-badge.png)](https://colab.research.google.com/github/Mjrovai/UNIFEI-IESTI01-TinyML-2022.1/blob/main/00_Curse_Folder/1_Fundamentals/Class_09/TF_MNIST_Classification_v2.ipynb)
 
 **Real-world medical diagnosis:** Use deep learning to tackle the important task of breast cancer classification.  
-[![](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Mjrovai/UNIFEI-IESTI01-TinyML-2022.1/blob/main/00_Curse_Folder/1_Fundamentals/Class_13/docs/WDBC_Project/Breast_Cancer_Classification.ipynb)
+[![](https://colab.research.google.com/assets/colab-badge.png)](https://colab.research.google.com/github/Mjrovai/UNIFEI-IESTI01-TinyML-2022.1/blob/main/00_Curse_Folder/1_Fundamentals/Class_13/docs/WDBC_Project/Breast_Cancer_Classification.ipynb)
 
 Are you excited to start building? Let's go!  
 
@@ -140,7 +140,7 @@ In embedded AI, CNNs are crucial for image and video recognition tasks, where re
 
 We discussed that CNNs excel at identifying features in images, making them ideal for tasks like object classification. Now, you'll get to put this knowledge into action! This Colab notebook focuses on building a CNN to classify images from the CIFAR-10 dataset, which includes objects like airplanes, cars, and animals. You'll learn about the key differences between CIFAR-10 and the MNIST dataset we explored earlier, and how these differences influence model choice. By the end of this notebook, you'll have a grasp of CNNs for image recognition and be well on your way to becoming a TinyML expert!  
 
-[![](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Mjrovai/UNIFEI-IESTI01-TinyML-2022.1/blob/main/00_Curse_Folder/1_Fundamentals/Class_11/CNN_Cifar_10.ipynb)
+[![](https://colab.research.google.com/assets/colab-badge.png)](https://colab.research.google.com/github/Mjrovai/UNIFEI-IESTI01-TinyML-2022.1/blob/main/00_Curse_Folder/1_Fundamentals/Class_11/CNN_Cifar_10.ipynb)
 
 :::
 

contents/frameworks/frameworks.qmd

@@ -126,7 +126,7 @@ The Parameter Server (PS) architecture is a popular design for distributing the
 
 Let's get a comprehensive understanding of core machine learning algorithms using TensorFlow and their practical applications in data analysis and predictive modeling. We will start with linear regression to predict survival rates from the Titanic dataset. Then using TensorFlow, we will construct classifiers to identify different species of flowers based on their attributes. Next, we will use K-Means algorithm and its application in segmenting datasets into cohesive clusters. Finally, we will apply hidden markov models (HMM), in order to foresee weather patterns.  
 
-[![](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/15Cyy2H7nT40sGR7TBN5wBvgTd57mVKay#scrollTo=IEeIRxlbx0wY)
+[![](https://colab.research.google.com/assets/colab-badge.png)](https://colab.research.google.com/drive/15Cyy2H7nT40sGR7TBN5wBvgTd57mVKay#scrollTo=IEeIRxlbx0wY)
 :::
 
 :::{#exr-tfl .callout-exercise collapse="true"}
@@ -136,7 +136,7 @@ Let's get a comprehensive understanding of core machine learning algorithms usin
 Here we will see how to build a miniature machine learning model for microcontrollers. We will get to build a mini neural network, streamlined to learn from data, even with limited resources and optimize for deployment by shrinking our model for efficient use on microcontrollers.
 TensorFlow Lite, a powerful technology derived from TensorFlow, shrinks models for tiny devices and helps enable on-device features like image recognition in smart devices and is used in edge computing to allow for faster analysis and decisions in devices processing data locally.  
 
-[![](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Mjrovai/UNIFEI-IESTI01-TinyML-2022.1/blob/main/00_Curse_Folder/2_Applications_Deploy/Class_16/TFLite-Micro-Hello-World/train_TFL_Micro_hello_world_model.ipynb)
+[![](https://colab.research.google.com/assets/colab-badge.png)](https://colab.research.google.com/github/Mjrovai/UNIFEI-IESTI01-TinyML-2022.1/blob/main/00_Curse_Folder/2_Applications_Deploy/Class_16/TFLite-Micro-Hello-World/train_TFL_Micro_hello_world_model.ipynb)
 :::
 
 DistBelief and its architecture defined above were crucial in enabling distributed deep learning at Google but also introduced limitations that motivated the development of TensorFlow:
@@ -172,7 +172,7 @@ Additionally, one of TensorFlow's biggest advantages is its integration with Ker
 
 Here, we'll learn how to use Keras, a high-level neural networks API, for model development and training. We will explore the functional API for concise model building, understand loss and metric classes for model evaluation, and use built-in optimizers to update model parameters during training. Additionally, we'll discover how to define custom layers and metrics tailored to our needs. Lastly, we'll delve into Keras' training loops to streamline the process of training neural networks on large datasets. This knowledge will empower us to build and optimize neural network models across various applications in machine learning and artificial intelligence.  
 
-[![](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1UCJt8EYjlzCs1H1d1X0iDGYJsHKwu-NO#scrollTo=fxINLLGitX_n)
+[![](https://colab.research.google.com/assets/colab-badge.png)](https://colab.research.google.com/drive/1UCJt8EYjlzCs1H1d1X0iDGYJsHKwu-NO#scrollTo=fxINLLGitX_n)
 :::
 
 ### Limitations and Challenges

contents/hw_acceleration/hw_acceleration.qmd

@@ -601,7 +601,7 @@ Key challenges include expressing parallelism, managing memory across devices, a
 
 We've learned that fancy AI hardware needs special software to work its magic. TVM is like a super-smart translator, turning your code into instructions that accelerators understand. In this Colab, we'll use TVM to make a pretend accelerator called VTA do matrix multiplication super fast. Ready to see how software powers up hardware?  
 
-[![](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/uwsampl/tutorial/blob/master/notebook/04a_TVM_Tutorial_VTA_Mat_Mult.ipynb)
+[![](https://colab.research.google.com/assets/colab-badge.png)](https://colab.research.google.com/github/uwsampl/tutorial/blob/master/notebook/04a_TVM_Tutorial_VTA_Mat_Mult.ipynb)
 
 :::
 

contents/ml_systems/ml_systems.qmd

@@ -162,6 +162,16 @@ TinyML excels in low-power and resource-constrained settings. These environments
 
 ![Tiny ML Example: (Left) A TinyML kit that includes Arduino Nano 33 BLE Sense, an OV7675 camera module, and TinyML shield. (Right) The Nano 33 BLE includes a host of onboard integrated sensors, a Bluetooth Low Energy module, and an Arm Cortex-M microcontroller that can run neural-network models using TensorFlow Lite for Microcontrollers. (Source: [Widening Access to Applied Machine Learning with TinyML](https://arxiv.org/pdf/2106.04008.pdf)))](images/jpg/tiny_ml.jpg){#fig-tinyml-example}
 
+:::{#exr-tinyml .callout-exercise collapse="true"}
+
+### TinyML with Arduino
+
+Get ready to bring machine learning to the smallest of devices!  In the embedded machine learning world, TinyML is where resource constraints meet ingenuity. This Colab notebook will walk you through building a gesture recognition model specifically designed to run on an Arduino board. You'll learn how to train a small but effective neural network, optimize it for minimal memory usage, and actually deploy it to your microcontroller. If you're excited about making everyday objects smarter, this is where it begins!  
+
+[![](https://colab.research.google.com/assets/colab-badge.png)](https://colab.research.google.com/github/arduino/ArduinoTensorFlowLiteTutorials/blob/master/GestureToEmoji/arduino_tinyml_workshop.ipynb)
+
+:::
+
 ### Benefits
 
 **Extremely Low Latency**