AWS wellarchitected documentation change
Summary
Updated figure numbering, formatting consistency changes (dashes to colons in bullet points), minor wording adjustments, and typo fixes in pipeline descriptions
Security assessment
Changes are primarily editorial improvements without addressing security vulnerabilities or introducing security-related content. Formatting changes (like ' - ' to ':') and figure renumbering don't impact security. The mention of 'debugging/profiling' relates to model performance rather than security vulnerabilities. No evidence of security patches or vulnerability disclosures.
Diff
diff --git a/wellarchitected/latest/machine-learning-lens/model-training-and-tuning.md b/wellarchitected/latest/machine-learning-lens/model-training-and-tuning.md index 20db798a5..73abce8f7 100644 --- a//wellarchitected/latest/machine-learning-lens/model-training-and-tuning.md +++ b//wellarchitected/latest/machine-learning-lens/model-training-and-tuning.md @@ -9 +9 @@ In this phase, you select a machine learning algorithm that is appropriate for y - + @@ -11 +11 @@ In this phase, you select a machine learning algorithm that is appropriate for y -_Figure 12: ML model training and tuning: main components_ +_Figure 11: ML model training and tuning main components_ @@ -13 +13 @@ _Figure 12: ML model training and tuning: main components_ -Model training, tuning, and evaluation require prepared data and engineered features. The following are the main activities in this stage, as listed in Figure 12: +Model training, tuning, and evaluation require prepared data and engineered features. The following are the main activities in this stage, as listed in Figure 11: @@ -15 +15 @@ Model training, tuning, and evaluation require prepared data and engineered feat - * **Features** – Features are selected as part of the data processing after data quality is assured + * **Features:** Features are selected as part of the data processing after data quality is assured @@ -17 +17 @@ Model training, tuning, and evaluation require prepared data and engineered feat - * **Building code** – Model development includes building the algorithm and its supporting code. The code-building process should support version control and continuous build, test, and integration through a pipeline. + * **Building code:** Model development includes building the algorithm and its supporting code. The code-building process should support version control and continuous build, test, and integration through a pipeline. @@ -19 +19 @@ Model training, tuning, and evaluation require prepared data and engineered feat - * **Algorithm selection** – Selecting the right algorithm involves running many experiments with parameter tuning across available options. Factors to consider when evaluating each option can include success metrics, model explainability, and compute requirements (training/prediction time and memory requirements). + * **Algorithm selection:** Selecting the right algorithm involves running many experiments with parameter tuning across available options. Factors to consider when evaluating each option can include success metrics, model explainability, and compute requirements (training/prediction time and memory requirements). @@ -21 +21 @@ Model training, tuning, and evaluation require prepared data and engineered feat - * **Model training** (data parallel, model parallel)– The process of training a ML model involves providing the algorithm with training data to learn from. Distributed training enables splitting large models and training datasets across compute instances to reduce training time significantly. Model parallelism and data parallelism are techniques to achieve distributed training. + * **Model training (data parallel, model parallel):** The process of training a ML model involves providing the algorithm with training data to learn from. Distributed training enables splitting large models and training datasets across compute instances to reduce training time significantly. Model parallelism and data parallelism are techniques to achieve distributed training. @@ -27 +27 @@ Model training, tuning, and evaluation require prepared data and engineered feat - * **Debugging/profiling** – A machine learning training job can have problems including: system bottlenecks, overfitting, saturated activation functions, and vanishing gradients. These problems can compromise model performance. A debugger provides visibility into the ML training process through monitoring, recording, and analyzing data. It captures the state of a training job at periodic intervals. + * **Debugging or profiling:** A machine learning training job can have problems including: system bottlenecks, overfitting, saturated activation functions, and vanishing gradients. These problems can compromise model performance. A debugger provides visibility into the ML training process through monitoring, recording, and analyzing data. It captures the state of a training job at periodic intervals. @@ -29 +29 @@ Model training, tuning, and evaluation require prepared data and engineered feat - * **Validation metrics** – Typically, a training algorithm computes several metrics such as loss and prediction accuracy. These metrics determine if the model is learning and generalizing well for making predictions on unseen data. Metrics reported by the algorithm depend on the business problem and the ML technique used. For example, a _confusion matrix_ is one of the metrics used for classification models, and Root Mean Squared Error (RMSE) is one of the metrics for regression models. + * **Validation metrics:** Typically, a training algorithm computes several metrics such as loss and prediction accuracy. These metrics determine if the model is learning and generalizing well for making predictions on unseen data. Metrics reported by the algorithm depend on the business problem and the ML technique used. For example, a _confusion matrix_ is one of the metrics used for classification models, and Root Mean Squared Error (RMSE) is one of the metrics for regression models. @@ -31 +31 @@ Model training, tuning, and evaluation require prepared data and engineered feat - * **Hyperparameter tuning** – Settings that can be tuned to control the behavior of the ML algorithm are referred to as _hyperparameters_. The number and type of hyperparameters in ML algorithms are specific to each model. Examples of commonly used hyperparameters include: learning rate, number of epochs, hidden layers, hidden units and activation functions. Hyperparameter tuning, or optimization, is the process of choosing the optimal hyperparameters for an algorithm. + * **Hyperparameter tuning:** Settings that can be tuned to control the behavior of the ML algorithm are referred to as _hyperparameters_. The number and type of hyperparameters in ML algorithms are specific to each model. Examples of commonly used hyperparameters include: learning rate, number of epochs, hidden layers, hidden units and activation functions. Hyperparameter tuning, or optimization, is the process of choosing the optimal hyperparameters for an algorithm. @@ -33 +33 @@ Model training, tuning, and evaluation require prepared data and engineered feat - * **Training code container** – Create container images with your training code and its entire dependency stack. This will enable the training and deployment of models quickly and reliably at any scale. + * **Training code container:** Create container images with your training code and its entire dependency stack. This will enable the training and deployment of models quickly and reliably at scale. @@ -35 +35 @@ Model training, tuning, and evaluation require prepared data and engineered feat - * **Model artifacts** – Model artifacts are the outputs that results from training a model. They typically consist of trained parameters, a model definition that describes how to compute inferences, and other metadata. + * **Model artifacts:** Model artifacts are the outputs that results from training a model. They typically consist of trained parameters, a model definition that describes how to compute inferences, and other metadata. @@ -37 +37 @@ Model training, tuning, and evaluation require prepared data and engineered feat - * **Visualization** – Enables exploring and understanding data during metrics validation, debugging, profiling, and hyperparameter tuning. + * **Visualization:** Enables exploring and understanding data during metrics validation, debugging, profiling, and hyperparameter tuning. @@ -42 +42 @@ Model training, tuning, and evaluation require prepared data and engineered feat - + @@ -44 +44 @@ Model training, tuning, and evaluation require prepared data and engineered feat -_Figure 13: ML lifecycle with pre-production pipelines_ +_Figure 12: ML lifecycle with pre-production pipelines_ @@ -46 +46 @@ _Figure 13: ML lifecycle with pre-production pipelines_ -Figure 13 shows the pre-production pipelines. The _data prepare_ pipeline automates data preparation tasks. The feature pipeline automates the storing, fetching, and copying of the features into and from online/offline store. The CI/CD/CT pipeline automates the build, train, and release to staging and production environments. +Figure 12 shows the pre-production pipelines. The _data prepare_ pipeline automates data preparation tasks. The feature pipeline automates the storing, fetching, and copying of the features into and from online/offline store. The CI/CD/CT pipeline automates the build, train, and release to staging and production environments. @@ -54 +54 @@ To use the Amazon Web Services Documentation, Javascript must be enabled. Please -Model development lifecycle phase +Model development