I’m planning to learn and test myself with competitions on Kaggle.
Kaggle is a place with real-world problems where Data Scientists and alike can go against each other to solve problems with Machine Learning.
From my understanding:
- There is a validation set where your model is tested against and a public leader board to see how you’re going.
- At the end of the competition, there is an unseen test set where everyones models is tested against and where the final rankings are determined.
- This is quite reflective of real world where preparing a representative validation set is vital, thus will perform well on the test set.
- A common mistake for newbs is over-fitting to the validation set. I’m ready to make that mistake 🤣.
As for the Kaggle API, you can download the kernel which is the necessary datasets and source files to do the competitions.
Alternatively, I can use the notebooks on their website. I plan to try doing competitions both ways.
This is how I set up my Kaggle API
1. Install Library
pip install python
2. Create API token (.json file)
- Go to Kaggle
- Go to Settings
- Create New Token
3. Save to your local .kaggle folder (Windows)
Location: C:\Users\<Windows-username>\.kaggle\kaggle.json
Kaggle Github Reference
3.1 Pasted into the wrong folder?
If you did something wrong then ran kaggle in the terminal, you’ll an error (telling you where to put it):
4 Start Kaggling
Type: module
String form: <module 'kaggle' from 'c:\\Users\\tonyp\\miniconda3\\envs\\fastai\\Lib\\site-packages\\kaggle\\__init__.py'>
File: c:\users\tonyp\miniconda3\envs\fastai\lib\site-packages\kaggle\__init__.py
Source:
#!/usr/bin/python
#
# Copyright 2024 Kaggle Inc
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# coding=utf-8
from __future__ import absolute_import
from kaggle.api.kaggle_api_extended import KaggleApi
from kaggle.api_client import ApiClient
api = KaggleApi(ApiClient())
api.authenticate()