A Day in the Life of a Data Scientist Conquer Machine Learning Lifecycle on Kubernetes Brian Redmond • Cloud Architect @ Microsoft (18 years) • Azure Global Black Belt Team • Live in Pittsburgh, PA Repeatable/consistent • CI/CD • This has worked well for App Dev. Now time for AI/ML • But, must ensure data scientist are not hindered by structure Why Containers, Kubernetes & Helm? • Container • Contains everything

Camilla Falconi Crispim James Lewis Scott Shaw Rachel Laycock （CTO） Martin Fowler （Chief Scientist） Erik Dörnenburg Marisa Hoenig Selvakumar Natesan Shangqi Liu Sofia Tania Vanya Seth Bharani

max_value represent the upper bound  axis represent the dimension in which the constraint to be applied. e.g. in Shape (2,3,4) axis 0 denotes first dimension, 1 denotes second dimension and 2 denotes kernel_constraint=my_constrain)) where, rate represent the rate at which the weight constrain is applied. Regularizers In machine learning, regularizers are used in the optimization phase. It applies kernel_regularizer represents the regularizer function to be applied to the kernel weights matrix.  bias_regularizer represents the regularizer function to be applied to the bias vector.  activity_regularizer

value 'my frontend' # If the same annotation is set multiple times, only the last value will be applied oc annotate pods foo description='my frontend' # Update a pod identified by type and name in (OC) 23 2.5.1.5. oc apply edit-last-applied 编辑资源/对象的最新 last-applied-configuration 注解 用法示例 用法示例 2.5.1.6. oc apply set-last-applied 设置 live 对象上的 last-applied-configuration 注释，以匹配文件的内容。 用法示例 用法示例 view-last-applied 查看资源/对象最新的最后应用配置注解 用法示例 用法示例 2.5.1.8. oc attach 附加到正在运行的容器 用法示例 用法示例 # Edit the last-applied-configuration annotations by type/name in YAML oc apply edit-last-applied deployment/nginx

value 'my frontend' # If the same annotation is set multiple times, only the last value will be applied oc annotate pods foo description='my frontend' # Update a pod identified by type and name in (OC) 27 2.7.1.5. oc apply edit-last-applied 编辑资源/对象的最新 last-applied-configuration 注解 用法示例 用法示例 2.7.1.6. oc apply set-last-applied 设置 live 对象上的 last-applied-configuration 注释，以匹配文件的内容。 用法示例 用法示例 2.7.1.7. oc apply view-last-applied 查看资源/对象最新的最后应用配置注解 # Apply the configuration in pod.json to a pod oc apply -f ./pod.json # Apply resources from a directory containing kustomization.yaml -

value 'my frontend'. # If the same annotation is set multiple times, only the last value will be applied oc annotate pods foo description='my frontend' # Update a pod identified by type and name in (OC) 23 2.5.1.5. oc apply edit-last-applied 编辑资源/对象的最新 last-applied-configuration 注解 用法示例 用法示例 2.5.1.6. oc apply set-last-applied 设置 live 对象上的 last-applied-configuration 注释，以匹配文件的内容。 用法示例 用法示例 view-last-applied 查看资源/对象最新的最后应用配置注解 用法示例 用法示例 2.5.1.8. oc attach 附加到正在运行的容器 用法示例 用法示例 # Edit the last-applied-configuration annotations by type/name in YAML. oc apply edit-last-applied deployment/nginx

statistics (mean, median, min, max, counts...) are easily calculable. These or custom aggregations can be applied on the entire data set, a sliding window of the data, or grouped by categories. The latter is also pandas and Matplotlib explicit enables all the power of matplotlib to the plot. This strategy is applied in the previous example: fig, axs = plt.subplots(figsize=(12, 4)) # Create an empty matplotlib Figure 2019-05-07 06:00:00 NaN 61.9 NaN NaN␣ ˓→ NaN The calculation is again element-wise, so the / is applied for the values in each row. Also other mathematical operators (+, -, \*, /) or logical operators

statistics (mean, median, min, max, counts...) are easily calculable. These or custom aggregations can be applied on the entire data set, a sliding window of the data, or grouped by categories. The latter is also pandas and Matplotlib explicit enables all the power of matplotlib to the plot. This strategy is applied in the previous example: fig, axs = plt.subplots(figsize=(12, 4)) # Create an empty matplotlib Figure 2019-05-07 06:00:00 NaN 61.9 NaN NaN␣ ˓→ NaN The calculation is again element-wise, so the / is applied for the values in each row. Also other mathematical operators (+, -, \*, /) or logical operators

statistics (mean, median, min, max, counts...) are easily calculable. These or custom aggregations can be applied on the entire data set, a sliding window of the data, or grouped by categories. The latter is also pandas and Matplotlib explicit enables all the power of matplotlib to the plot. This strategy is applied in the previous example: fig, axs = plt.subplots(figsize=(12, 4)) # Create an empty matplotlib Figure 2019-05-07 06:00:00 NaN 61.9 NaN ˓→NaN NaN The calculation is again element-wise, so the / is applied for the values in each row. Also other mathematical operators (+, -, \*, /) or logical operators

statistics (mean, median, min, max, counts...) are easily calculable. These or custom aggregations can be applied on the entire data set, a sliding window of the data, or grouped by categories. The latter is also pandas and Matplotlib explicit enables all the power of matplotlib to the plot. This strategy is applied in the previous example: fig, axs = plt.subplots(figsize=(12, 4)) # Create an empty matplotlib Figure 2019-05-07 06:00:00 NaN 61.9 NaN NaN␣ ˓→ NaN The calculation is again element-wise, so the / is applied for the values in each row. Also other mathematical operators (+, -, \*, /) or logical operators