Merge pull request #161 from datmo/r-example

R script example

examples/R/README.md

@@ -0,0 +1,47 @@
+# CLI + R examples
+
+In order to run the examples, make sure that you have datmo properly installed with the latest 
+stable or development version. You can install it with the following command:
+```
+$ pip install datmo
+```
+*Note: Datmo installation is currently only officially supported through pip, Python's package manager. If you don't already have it installed, [you can find it here.](https://pip.pypa.io/en/stable/installing/)*
+
+## Using the Examples
+### R
+1. Navigate to desired project folder or create a new one 
+
+        $ mkdir MY_DATMO_PROJECT
+        $ cd MY_DATMO_PROJECT
+        
+2. Initialize the datmo project
+
+        $ datmo init
+
+3. Copy/save example files within project folder (if directory, copy the contents of the directory)
+
+        $ cp /path/to/SCRIPT_NAME.R .
+        
+   If the filename for the example is a directory then you can run the following
+
+        $ cp /path/to/DIRECTORY/* .
+        
+4. Open RStudio from the terminal (allows $PATH to find Datmo in MacOS)
+
+        $ open -a RStudio
+
+5. Change your current working directory within RStudio to the root directory of your project. 
+
+This can be done either through the GUI, or with the following R Console command.
+
+        `> setwd("~/path/of/projectfolder")`
+
+NOTE: You may find that some of the libraries requested in the script are not present, you can use 
+the RStudio GUI to install those packages if you run into any errors running the lines of code
+in the example.   
+
+### Examples
+
+| feature  | filename(s) | Instructions |
+| ------------- |:-------------:| -----|
+| Create Snapshot | `snapshot_create_iris_caret.R`| (1) Run `snapshot_create_iris_caret.R` <br> (2) See snapshot created with `$ datmo snapshot ls` |

examples/R/snapshot-create-iris-caret.R

@@ -0,0 +1,28 @@
+library(datasets)
+library(caret)
+library(rpart.plot)
+
+## Import data
+df <- iris
+head(df) # View first few rows of dataset
+
+## Fit Model
+modFit <- train(Species ~., method = "rpart", data=df) #Fit model
+print(modFit$finalModel)   #Summarize model
+
+# Visualize Results
+rpart.plot(modFit$finalModel) #create decision tree visualization
+
+## Define the properties we want to log in the snapshot
+# Define configuration to save
+config<- paste(sep="",
+               " --config method:", modFit$method,
+               " --config modelType:", modFit$modelType)
+
+#define metrics to save from the model
+stats<- paste(sep="",
+              " --stats Accuracy:", modFit$results$Accuracy[1],
+              " --stats Kappa:", modFit$results$Kappa[1])
+
+# Create snapshot with configuration and stats
+system2("datmo", args=paste("snapshot create", "-m 'Whoah, my first snapshot!'", config, stats))

examples/README.md

@@ -8,15 +8,27 @@ $ pip install datmo
 ```
 
 ## Using the Examples
-### CLI flow
-See [CLI flow examples](/examples/cli) for instructions
+### CLI (Standalone)
+The datmo CLI can work entirely as a standalone tool. It can also be used in conjunction with our language-specific SDKs to enhance the experience of the user and enable more granular control over model management during runtime.
+
+See [CLI flow examples](/examples/cli) for instructions.
+
+### Python
+We offer a fully supported Python SDK, that works in conjunction with the CLI.
 
-### CLI + Python flow
 See [CLI + Python flow examples](/examples/python) for instructions
 
-### CLI + Jupyter Notebook flow
+### Jupyter Notebook (IPython)
+For Python kernels, the Python SDK is compatible with Jupyter notebooks. 
+For running a containerized notebook environment, the user can do this with the CLI.
+
 See [CLI + Jupyter Notebook flow examples](/examples/jupyter_notebook) for instructions
 
+### R (beta)
+Users can call the CLI natively using the `system2()` command in R, allowing for granular control over snapshot creation.
+In the future, these system calls will be replaced with a more intuitive SDK.
+
+See [R flow examples](/examples/R) for instructions
 
 ## Examples
 Listed below are actions you might want to take with Datmo. For each
@@ -31,6 +43,8 @@ each example.
     * snapshot_create_iris_sklearn
 * [CLI + Jupyter Notebook flow](/examples/jupyter_notebook)
     * snapshot_create_iris_sklearn
+* [CLI + R flow](/examples/R)
+    * snapshot_create_iris_caret
 
 #### Running a containerized task (with option to create Snapshot)
 * [CLI + Python flow](/examples/python)