README for CIFAR and tweaks to MNIST README

cifar10/README.md

@@ -0,0 +1,97 @@
+# CIFAR10
+
+This is a [Guild AI](http://guild.ai) example that defines a CIFAR10
+model based on TensorFlow's excellent [Convolutional Neural
+Networks](https://www.tensorflow.org/tutorials/deep_cnn/) Guide.
+
+## Project files
+
+- **[Guild](Guild)** - Guild project file ([more info](https://guild.ai/project-reference/))
+- **[samples.py](samples.py)** - Implementation of the *samples* resource
+- **[single.py](expert.py)** - Training script for single GPU
+- **[support.py](support.py)** - Shared code across training scripts
+- **[upstream_single.py](upstream_single.py)** - Wrapper to train
+  using the original TensorFlow (upstream) script for single GPU
+
+## Requirements
+
+To work with this project, [ensure that you have Guild AI installed along
+with its requirements](https://guild.ai/getting-started/setup/).
+
+If you haven't already, clone Guild AI examples:
+
+```
+$ git clone https://github.com/guildai/guild-examples.git
+```
+
+## Training
+
+Prepare and train the CIFAR10 model:
+
+``` bash
+$ cd guild-examples/cifar10
+$ guild prepare
+$ guild train
+```
+
+The `prepare` command downloads the CIFAR10 data, which will be used
+for subsequent operations.
+
+## Viewing
+
+At any point you can view project run results by in [Guild
+View](https://guild.ai/user-guide/guild-view) by running the
+`view` command in a separate terminal:
+
+``` bash
+$ guild view
+```
+
+Guild View runs on port `6333` by default — to view it,
+open [http://localhost:6333](http://localhost:6333) in your browser.
+
+## Evaluating
+
+To calculate the most recently trained model's final accuracy using
+test data, use the `evaluate` command:
+
+``` bash
+$ guild evaluate --latest-run
+```
+
+## Serving
+
+To run a model as an HTTP service in [Guild
+Serve](https://guild.ai/user-guide/guild-serve/) run:
+
+``` bash
+$ guild serve RUN
+```
+
+where `RUN` is a value returned by `list-runs` (i.e. a path to the run
+directory) or `--latest-run` to serve the model exported in the last
+run.
+
+## Generating sample inputs
+
+This example provides a `samples` resource that can be used to
+generate a number of MNIST images and their corresponding JSON
+encodings. These file can be used to run ad hoc inference on the model
+in Guild View (see the **Serve** tab) or Guild Serve
+(see [Serving](#user-content-serving) above).
+
+Images are generated from the MNIST training/test data.
+
+To generate samples, run:
+
+``` bash
+$ guild prepare samples
+```
+
+This will create a local `samples` subdirectory containing the samples
+images.
+
+## More about Guild AI
+
+For more information about the Guild AI project, see
+[https://guild.ai](https://guild.ai).

mnist/README.md

@@ -1,6 +1,6 @@
 # MNIST
 
-This is a [Guild AI](http://guild.ai) example that provides two MNIST
+This is a [Guild AI](http://guild.ai) example that defines two MNIST
 models:
 
 - **intro** - logistic regression
@@ -19,7 +19,7 @@ themselves — are from TensorFlow's excellent
 
 ## Requirements
 
-To work with these models, [ensure that you have Guild AI installed along
+To work with this project, [ensure that you have Guild AI installed along
 with its requirements](https://guild.ai/getting-started/setup/).
 
 If you haven't already, clone Guild AI examples:
@@ -39,7 +39,7 @@ $ guild train
 ```
 
 The `prepare` command downloads the MNIST data, which will be used for
-all subsequent trainin.
+subsequent operations.
 
 This project is configured to train the `intro` model by trained. You
 can train `expert` by specifying it explicitly:
@@ -51,7 +51,7 @@ $ guild train expert
 
 ## Viewing
 
-At any point you can project run results by in [Guild
+At any point you can view project run results by in [Guild
 View](https://guild.ai/user-guide/guild-view) by running the
 `view` command in a separate terminal:
 
@@ -77,9 +77,13 @@ To run a model as an HTTP service in [Guild
 Serve](https://guild.ai/user-guide/guild-serve/) run:
 
 ``` bash
-$ guild serve
+$ guild serve RUN
 ```
 
+where `RUN` is a value returned by `list-runs` (i.e. a path to the run
+directory) or `--latest-run` to serve the model exported in the last
+run.
+
 ## Generating sample inputs
 
 This example provides a `samples` resource that can be used to