Author: David Lowe

Chief Technology Officer, Daines Analytics

Multi-Class Image Classification Model for Reptiles and Amphibians Image Using TensorFlow Take 2

Template Credit: Adapted from a template made available by Dr. Jason Brownlee of Machine Learning Mastery.

SUMMARY: This project aims to construct a predictive model using a TensorFlow convolutional neural network (CNN) and document the end-to-end steps using a template. The Reptiles and Amphibians Image dataset is a multi-class classification situation where we attempt to predict one of several (more than two) possible outcomes.

INTRODUCTION: Reptiles and amphibians are crucial components of the ecosystems in which they reside. They are also useful indicators of the condition of the environment. The presence of a variety of amphibians and reptiles in an area suggests that it is stable and capable of supporting plant and animal life.

The dataset contains ten different classes of reptiles and amphibians images. The research team have resized the images to either (300px, n) or (n,300px) where n is a pixel size less than 300px.

ANALYSIS: The InceptionV3 model’s performance achieved an accuracy score of 87.80% after ten epochs using the training dataset. When we applied the model to the validation dataset, the model achieved an accuracy score of 83.83%.

CONCLUSION: In this iteration, the TensorFlow InceptionV3 CNN model appeared suitable for modeling this dataset.

Dataset ML Model: Multi-Class classification with numerical features

Dataset Used: Reptiles and Amphibians Image Dataset

Dataset Reference: https://www.kaggle.com/datasets/vencerlanz09/reptiles-and-amphibians-image-dataset

One source of potential performance benchmarks: https://www.kaggle.com/code/vencerlanz09/reptiles-and-amphibians-classification-using-cnn/notebook

The HTML formatted report can be found here on GitHub.

Multi-Class Image Classification Model for Reptiles and Amphibians Image Using TensorFlow Take 1

Template Credit: Adapted from a template made available by Dr. Jason Brownlee of Machine Learning Mastery.

SUMMARY: This project aims to construct a predictive model using a TensorFlow convolutional neural network (CNN) and document the end-to-end steps using a template. The Reptiles and Amphibians Image dataset is a multi-class classification situation where we attempt to predict one of several (more than two) possible outcomes.

INTRODUCTION: Reptiles and amphibians are crucial components of the ecosystems in which they reside. They are also useful indicators of the condition of the environment. The presence of a variety of amphibians and reptiles in an area suggests that it is stable and capable of supporting plant and animal life.

The dataset contains ten different classes of reptiles and amphibians images. The research team have resized the images to either (300px, n) or (n,300px) where n is a pixel size less than 300px.

ANALYSIS: The VGG19 model’s performance achieved an accuracy score of 87.09% after ten epochs using the training dataset. When we applied the model to the validation dataset, the model achieved an accuracy score of 76.00%.

CONCLUSION: In this iteration, the TensorFlow VGG19 CNN model appeared suitable for modeling this dataset.

Dataset ML Model: Multi-Class classification with numerical features

Dataset Used: Reptiles and Amphibians Image Dataset

Dataset Reference: https://www.kaggle.com/datasets/vencerlanz09/reptiles-and-amphibians-image-dataset

One source of potential performance benchmarks: https://www.kaggle.com/code/vencerlanz09/reptiles-and-amphibians-classification-using-cnn/notebook

The HTML formatted report can be found here on GitHub.

Binary-Class Tabular Model for Pumpkin Seeds Identification Using Python and AutoKeras

Template Credit: Adapted from a template made available by Dr. Jason Brownlee of Machine Learning Mastery.

SUMMARY: The project aims to construct a predictive model using various machine learning algorithms and document the end-to-end steps using a template. The Pumpkin Seeds Identification dataset is a binary-class modeling situation where we attempt to predict one of two possible outcomes.

INTRODUCTION: Pumpkin seeds are frequently consumed as confection worldwide because of their adequate amount of protein, fat, carbohydrate, and mineral contents. The research team carried out a study on the two most important types of pumpkin seed, “Ürgüp Sivrisi” and “Çerçevelik,” generally grown in Ürgüp and Karacaören regions in Turkey. Furthermore, the morphological measurements of 2500 pumpkin seeds of both varieties were captured using the gray and binary forms of threshold techniques.

ANALYSIS: After 100 trials, the best AutoKeras model processed the training dataset with an accuracy score of 87.88%.

CONCLUSION: In this iteration, AutoKeras appeared to be a suitable algorithm for modeling this dataset.

Dataset Used: Pumpkin Seeds Dataset

Dataset ML Model: Binary classification with numerical features

Dataset Reference: https://www.muratkoklu.com/datasets/

One source of potential performance benchmarks: https://doi.org/10.1007/s10722-021-01226-0

The HTML formatted report can be found here on GitHub.

Binary-Class Tabular Model for Pumpkin Seeds Identification Using Python and TensorFlow

Template Credit: Adapted from a template made available by Dr. Jason Brownlee of Machine Learning Mastery.

SUMMARY: The project aims to construct a predictive model using various machine learning algorithms and document the end-to-end steps using a template. The Pumpkin Seeds Identification dataset is a binary-class modeling situation where we attempt to predict one of two possible outcomes.

INTRODUCTION: Pumpkin seeds are frequently consumed as confection worldwide because of their adequate amount of protein, fat, carbohydrate, and mineral contents. The research team carried out a study on the two most important types of pumpkin seed, “Ürgüp Sivrisi” and “Çerçevelik,” generally grown in Ürgüp and Karacaören regions in Turkey. Furthermore, the morphological measurements of 2500 pumpkin seeds of both varieties were captured using the gray and binary forms of threshold techniques.

ANALYSIS: The average performance of the cross-validated TensorFlow models achieved an accuracy benchmark of 87.40%.

CONCLUSION: In this iteration, TensorFlow appeared to be a suitable algorithm for modeling this dataset.

Dataset Used: Pumpkin Seeds Dataset

Dataset ML Model: Binary classification with numerical features

Dataset Reference: https://www.muratkoklu.com/datasets/

One source of potential performance benchmarks: https://doi.org/10.1007/s10722-021-01226-0

The HTML formatted report can be found here on GitHub.

Binary-Class Tabular Model for Pumpkin Seeds Identification Using Python and XGBoost

Template Credit: Adapted from a template made available by Dr. Jason Brownlee of Machine Learning Mastery.

SUMMARY: The project aims to construct a predictive model using various machine learning algorithms and document the end-to-end steps using a template. The Pumpkin Seeds Identification dataset is a binary-class modeling situation where we attempt to predict one of two possible outcomes.

INTRODUCTION: Pumpkin seeds are frequently consumed as confection worldwide because of their adequate amount of protein, fat, carbohydrate, and mineral contents. The research team carried out a study on the two most important types of pumpkin seed, “Ürgüp Sivrisi” and “Çerçevelik,” generally grown in Ürgüp and Karacaören regions in Turkey. Furthermore, the morphological measurements of 2500 pumpkin seeds of both varieties were captured using the gray and binary forms of threshold techniques.

ANALYSIS: The performance of the preliminary XGBoost model achieved an accuracy benchmark of 88.40%. After a series of tuning trials, the final model processed the training dataset with an accuracy score of 88.68%.

CONCLUSION: In this iteration, the XGBoost model appeared to be a suitable algorithm for modeling this dataset.

Dataset Used: Pumpkin Seeds Dataset

Dataset ML Model: Binary classification with numerical features

Dataset Reference: https://www.muratkoklu.com/datasets/

One source of potential performance benchmarks: https://doi.org/10.1007/s10722-021-01226-0

The HTML formatted report can be found here on GitHub.

Steven Pressfield on Put Your Ass Where Your Heart Wants to Be, Part 5

In the book, Put Your Ass Where Your Heart Wants to Be, Steven Pressfield shares his inspiration and techniques to help us make the life-altering transformation.

These are some of my favorite takeaways from reading the book.

Up to this point, Steve has been talking about making the commitment. But “put your ass” is not a short, one-time endeavor. This concept applies lifelong.

When we commit to “put our ass,” the duration of the journey can be long, repetitive, and never-ending. Consider the passage in the book from James Rhodes, the concert pianist.

Steve lists some of the accomplished artists in the book and compares them to the statement of “going the distance.” We will find… “When you and I put our ass where our heart wants to be, we do it for keeps. We’re in it to the end of the line.”

Just like those accomplished artists, we have a body of work, too. For most of us, that body of work exists inside us, just waiting for us to bring it out to the world. These bodies of work exist as alternative futures. If we choose not to commit over a long haul, that body of work may exist only in our imagination.

So, can we put our ass where our hearts want to be if we have a family, a job, and a mortgage? “Yes,” says Steve.

The Muse does not count hours. She counts commitment. We can commit to a smaller time slice, but our commitment needs to be consistent. We can add up our work and commitment drip by drip.

At some point, the practice of our vocation moves from being a challenge that we must step up and accept to becoming simply… our life.

Like a mother raising her children or a farmer tending his crops, the commitment becomes who we are and what we do.

這還要多久？

（從我一個尊敬的作家，賽斯·高汀）

那要看看下面：

  • 我們開始後規格會改變嗎？
  • 我們是否要依賴其他人的供應或投入？
  • 這預算會改變嗎？
  • 這項工作以前有人做過嗎？
  • 這是這個團隊以前做過的工作嗎？
  • 快速完成它，比做得好，還是按著預算走是那個最重要？
  • 你想參與這項工作嗎（見關於規範的部分）？
  • 從事該項目的人員的動機是什麼？
  • 會涉及多少不同的人？
  • 所有的人員、預算和資產都已經準備到位了嗎？
  • 是誰在選擇工具？

尋路會比跟踪路徑來的花費更長的時間。討論這會導致規範的變化，因變的要素總是會增加時間。