Add a canary model (streaming-capable hello-world)

Author: nickstenning

README.md

@@ -7,6 +7,7 @@ Once you've got a working model and want to publish it so others can see it in a
 ## Examples in this repo
 
 - [blur](blur)
+- [canary](canary)
 - [hello-world](hello-world)
 - [notebook](notebook)
 - [resnet](resnet)

canary/README.md

@@ -0,0 +1,21 @@
+# Canary
+
+This simple model takes a string as input and returns a streaming string output.
+
+## Usage
+
+First, make sure you've got the [latest version of Cog](https://github.com/replicate/cog#install) installed.
+
+Build the container image:
+
+```sh
+cog build
+```
+
+Now you can run predictions on the model:
+
+```sh
+cog predict -i text=Athena
+
+cog predict -i text=Zeus
+```

canary/cog.yaml

@@ -0,0 +1,3 @@
+build:
+  python_version: "3.8"
+predict: "predict.py:Predictor"

canary/predict.py

@@ -0,0 +1,8 @@
+from cog import BasePredictor, ConcatenateIterator, Input
+
+
+class Predictor(BasePredictor):
+    def predict(self, text: str = Input(description="Text to prefix with 'hello there, '")) -> ConcatenateIterator[str]:
+        yield "hello "
+        yield "there, "
+        yield text