https://github.com/robmarkcole/classificationbox_python

Python script for teaching Classificationbox image classes
https://github.com/robmarkcole/classificationbox_python

machinebox

Last synced: 12 days ago
JSON representation

Python script for teaching Classificationbox image classes

• Host: GitHub
• URL: https://github.com/robmarkcole/classificationbox_python
• Owner: robmarkcole
• License: mit
• Created: 2018-05-28T11:36:54.000Z (almost 7 years ago)
• Default Branch: master
• Last Pushed: 2019-04-01T16:15:59.000Z (about 6 years ago)
• Last Synced: 2025-05-13T03:59:21.268Z (12 days ago)
• Topics: machinebox
• Language: Jupyter Notebook
• Homepage: https://blog.machinebox.io/how-anyone-can-build-a-machine-learning-image-classifier-from-photos-on-your-hard-drive-very-5c20c6f2764f
• Size: 1.09 MB
• Stars: 2
• Watchers: 3
• Forks: 3
• Open Issues: 4
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
Script to teach classificationbox image classes as described in

https://blog.machinebox.io/how-anyone-can-build-a-machine-learning-image-classifier-from-photos-on-your-hard-drive-very-5c20c6f2764f

This script is therefore a python implementation of

https://github.com/machinebox/toys/blob/master/imgclass/main.go

First create a model using CURL:

```CURL

$ curl -XPOST --header "Content-Type: application/json" --header "Accept: application/json"  http://localhost:8080/classificationbox/models --data '{

    "id": "sentiment1",

    "name": "sentimentModel",

    "options": {

        "ngrams": 1,

        "skipgrams": 1

    },

    "classes": [

        "class1",

        "class2",

        "class3"

    ]

}

```

You can check what models you have available:

```CURL

curl -XPOST --header "Content-Type: application/json" --header "Accept: application/json"  http://localhost:8080/classificationbox/models

```

Then run the script `teach_classificationbox.py` from within the root folder containing the folders of images.