Update example README files

Author: deliahu

examples/image-classifier/README.md

@@ -8,13 +8,16 @@ A `deployment` specifies a set of resources that are deployed as a single unit.
 
 ```yaml
 - kind: deployment
-  name: image-classifier
+  name: image
 
 - kind: api
-  name: alexnet
+  name: classifier
   model: s3://cortex-examples/image-classifier/alexnet.onnx
   request_handler: alexnet_handler.py
+  tracker:
+    model_type: classification
 ```
+
 <!-- CORTEX_VERSION_MINOR x2 -->
 You can run the code that generated the exported models used in this example folder here:
 - [Pytorch Alexnet](https://colab.research.google.com/github/cortexlabs/cortex/blob/master/examples/image-classifier/alexnet.ipynb)
@@ -76,7 +79,7 @@ Behind the scenes, Cortex containerizes the models, makes them servable using ON
 You can track the statuses of the APIs using `cortex get`:
 
 ```bash
-$ cortex get alexnet --watch
+$ cortex get classifier --watch
 
 status   up-to-date   available   requested   last update   avg latency
 live     1            1           1           12s           -
@@ -87,11 +90,11 @@ The output above indicates that one replica of the API was requested and one rep
 ## Serve real-time predictions
 
 ```bash
-$ cortex get alexnet
+$ cortex get classifier
 
-url: http://***.amazonaws.com/image-classifier/alexnet
+url: http://***.amazonaws.com/image/classifier
 
-$ curl http://***.amazonaws.com/image-classifier/alexnet \
+$ curl http://***.amazonaws.com/image/classifier \
     -X POST -H "Content-Type: application/json" \
     -d '{"url": "https://bowwowinsurance.com.au/wp-content/uploads/2018/10/akita-700x700.jpg"}'
 

examples/iris-classifier/README.md

@@ -14,7 +14,10 @@ Define a `deployment` and an `api` resource in `cortex.yaml`. A `deployment` spe
   name: classifier
   model: s3://cortex-examples/iris-classifier/tensorflow
   request_handler: handlers/tensorflow.py
+  tracker:
+    model_type: classification
 ```
+
 <!-- CORTEX_VERSION_MINOR x5 -->
 You can run the code that generated the exported models used in this folder example here:
 - [Tensorflow](https://colab.research.google.com/github/cortexlabs/cortex/blob/master/examples/iris-classifier/models/tensorflow.ipynb)

examples/sentiment-analysis/README.md

@@ -4,7 +4,7 @@ This example shows how to deploy a sentiment analysis classifier trained using [
 
 ## Define a deployment
 
-A `deployment` specifies a set of resources that are deployed as a single unit. An `api` makes a model available as a web service that can serve real-time predictions. This configuration will download the model from the `cortex-examples` S3 bucket and preprocess the payload and postprocess the inference with functions defined in `sentiment.py`.
+A `deployment` specifies a set of resources that are deployed as a single unit. An `api` makes a model available as a web service that can serve real-time predictions. This configuration will download the model from the `cortex-examples` S3 bucket and preprocess the payload and postprocess the inference with functions defined in `handler.py`.
 
 ```yaml
 - kind: deployment
@@ -13,8 +13,11 @@ A `deployment` specifies a set of resources that are deployed as a single unit.
 - kind: api
   name: classifier
   model: s3://cortex-examples/sentiment-analysis/bert
-  request_handler: sentiment.py
+  request_handler: handler.py
+  tracker:
+    model_type: classification
 ```
+
 <!-- CORTEX_VERSION_MINOR -->
 You can run the code that generated the exported BERT model [here](https://colab.research.google.com/github/cortexlabs/cortex/blob/master/examples/sentiment-analysis/bert.ipynb).
 
@@ -74,11 +77,11 @@ The output above indicates that one replica of the API was requested and one rep
 ## Serve real-time predictions
 
 ```bash
-$ cortex get analysis
+$ cortex get classifier
 
-url: http://***.amazonaws.com/sentiment/analysis
+url: http://***.amazonaws.com/sentiment/classifier
 
-$ curl http://***.amazonaws.com/sentiment/analysis \
+$ curl http://***.amazonaws.com/sentiment/classifier \
     -X POST -H "Content-Type: application/json" \
     -d '{"review": "The movie was great!"}'
 

examples/text-generator/README.md

@@ -4,7 +4,7 @@ This example shows how to deploy OpenAI's GPT-2 model as a service on AWS.
 
 ## Define a deployment
 
-A `deployment` specifies a set of resources that are deployed as a single unit. An `api` makes a model available as a web service that can serve real-time predictions. This configuration will download the 124M GPT-2 model from the `cortex-examples` S3 bucket, preprocess the payload and postprocess the inference with functions defined in `encoder.py` and deploy each replica of the API on 1 GPU.
+A `deployment` specifies a set of resources that are deployed as a single unit. An `api` makes a model available as a web service that can serve real-time predictions. This configuration will download the 124M GPT-2 model from the `cortex-examples` S3 bucket, preprocess the payload and postprocess the inference with functions defined in `handler.py` and deploy each replica of the API on 1 GPU.
 
 ```yaml
 - kind: deployment
@@ -13,16 +13,18 @@ A `deployment` specifies a set of resources that are deployed as a single unit.
 - kind: api
   name: generator
   model: s3://cortex-examples/text-generator/gpt-2/124M
-  request_handler: encoder.py
+  request_handler: handler.py
   compute:
+    cpu: 1
     gpu: 1
 ```
+
 <!-- CORTEX_VERSION_MINOR -->
 You can run the code that generated the exported GPT-2 model [here](https://colab.research.google.com/github/cortexlabs/cortex/blob/master/examples/text-generator/gpt-2.ipynb).
 
 ## Add request handling
 
-The model requires encoded data for inference, but the API should accept strings of natural language as input. It should also decode the model’s prediction before responding to the client. This can be implemented in a request handler file using the pre_inference and post_inference functions. See [encoder.py](encoder.py) for the complete code.
+The model requires encoded data for inference, but the API should accept strings of natural language as input. It should also decode the model’s prediction before responding to the client. This can be implemented in a request handler file using the pre_inference and post_inference functions.
 
 ```python
 from encoder import get_encoder
@@ -66,6 +68,10 @@ The output above indicates that one replica of the API was requested and one rep
 ## Serve real-time predictions
 
 ```bash
+$ cortex get generator
+
+url: http://***.amazonaws.com/text/generator
+
 $ curl http://***.amazonaws.com/text/generator \
     -X POST -H "Content-Type: application/json" \
     -d '{"text": "machine learning"}'