Spaces:
Sleeping
Sleeping
Clement Vachet
commited on
Commit
·
72500f8
1
Parent(s):
8f37ccc
Add Docker deployment
Browse files
README.md
CHANGED
|
@@ -21,21 +21,25 @@ short_description: IRIS Classification Lambda
|
|
| 21 |

|
| 22 |

|
| 23 |
|
| 24 |
-
|
| 25 |
based on measurements of physical characteristics (sepals and petals).
|
| 26 |
|
| 27 |
-
|
|
|
|
|
|
|
| 28 |
- Front-end: user interface via Gradio library
|
| 29 |
- Back-end: use of AWS Lambda function to run deployed ML model
|
| 30 |
|
| 31 |
You can try out our deployed [Hugging Face Space](https://huggingface.co/spaces/cvachet/iris_classification_lambda
|
| 32 |
)!
|
| 33 |
|
| 34 |
-
|
|
|
|
| 35 |
- [Local development](#1-local-development)
|
| 36 |
- [AWS deployment](#2-deployment-to-aws)
|
| 37 |
- [Hugging Face deployment](#3-deployment-to-hugging-face)
|
| 38 |
-
|
|
|
|
| 39 |
|
| 40 |
## 1. Local development
|
| 41 |
|
|
@@ -73,7 +77,7 @@ python
|
|
| 73 |
|
| 74 |
Use of Gradio library for web interface
|
| 75 |
|
| 76 |
-
|
| 77 |
> export AWS_API=http://localhost:8080
|
| 78 |
|
| 79 |
Command line for execution:
|
|
@@ -114,7 +118,7 @@ Example: ```iris-classification-lambda```
|
|
| 114 |
|
| 115 |
<details>
|
| 116 |
|
| 117 |
-
|
| 118 |
- Create function from container image
|
| 119 |
|
| 120 |
Example name: ```iris-classification```
|
|
@@ -139,7 +143,7 @@ Advanced notes:
|
|
| 139 |
|
| 140 |
<details>
|
| 141 |
|
| 142 |
-
|
| 143 |
- Create a new ```Rest API``` (e.g. ```iris-classification-api```)
|
| 144 |
- Add a new resource to the API (e.g. ```/classify```)
|
| 145 |
- Add a ```POST``` method to the resource
|
|
@@ -172,3 +176,11 @@ Hugging Face space URL:
|
|
| 172 |
https://huggingface.co/spaces/cvachet/iris_classification_lambda
|
| 173 |
|
| 174 |
Note: This space uses the ML model deployed on AWS Lambda
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 21 |

|
| 22 |

|
| 23 |
|
| 24 |
+
**Aims:** Categorization of different species of iris flowers (Setosa, Versicolor, and Virginica)
|
| 25 |
based on measurements of physical characteristics (sepals and petals).
|
| 26 |
|
| 27 |
+
**Method:** Use of Decision Tree Classifier
|
| 28 |
+
|
| 29 |
+
**Architecture:**
|
| 30 |
- Front-end: user interface via Gradio library
|
| 31 |
- Back-end: use of AWS Lambda function to run deployed ML model
|
| 32 |
|
| 33 |
You can try out our deployed [Hugging Face Space](https://huggingface.co/spaces/cvachet/iris_classification_lambda
|
| 34 |
)!
|
| 35 |
|
| 36 |
+
----
|
| 37 |
+
**Table of contents:**
|
| 38 |
- [Local development](#1-local-development)
|
| 39 |
- [AWS deployment](#2-deployment-to-aws)
|
| 40 |
- [Hugging Face deployment](#3-deployment-to-hugging-face)
|
| 41 |
+
- [Docker Hub deployment](#4-deployment-to-docker-hub)
|
| 42 |
+
----
|
| 43 |
|
| 44 |
## 1. Local development
|
| 45 |
|
|
|
|
| 77 |
|
| 78 |
Use of Gradio library for web interface
|
| 79 |
|
| 80 |
+
**Note:** The environment variable ```AWS_API``` should point to the local container
|
| 81 |
> export AWS_API=http://localhost:8080
|
| 82 |
|
| 83 |
Command line for execution:
|
|
|
|
| 118 |
|
| 119 |
<details>
|
| 120 |
|
| 121 |
+
**Steps**:
|
| 122 |
- Create function from container image
|
| 123 |
|
| 124 |
Example name: ```iris-classification```
|
|
|
|
| 143 |
|
| 144 |
<details>
|
| 145 |
|
| 146 |
+
**Steps**:
|
| 147 |
- Create a new ```Rest API``` (e.g. ```iris-classification-api```)
|
| 148 |
- Add a new resource to the API (e.g. ```/classify```)
|
| 149 |
- Add a ```POST``` method to the resource
|
|
|
|
| 176 |
https://huggingface.co/spaces/cvachet/iris_classification_lambda
|
| 177 |
|
| 178 |
Note: This space uses the ML model deployed on AWS Lambda
|
| 179 |
+
|
| 180 |
+
|
| 181 |
+
## 4. Deployment to Docker Hub
|
| 182 |
+
|
| 183 |
+
This web application is available on Docker Hub as a docker image
|
| 184 |
+
|
| 185 |
+
URL:
|
| 186 |
+
https://hub.docker.com/r/cvachet/iris-classification-lambda
|