
Picture by Creator
The Kaggle CLI (Command Line Interface) permits you to work together with Kaggle’s datasets, competitions, notebooks, and fashions straight out of your terminal. That is helpful for automating downloads, submissions, and dataset administration while not having an online browser. Most of my GitHub Motion workflows use Kaggle CLI for downloading or pushing datasets, as it’s the quickest and best approach.
1. Set up & Setup
Be sure to have Python 3.10+ put in. Then, run the next command in your terminal to put in the official Kaggle API:
To acquire your Kaggle credentials, obtain the kaggle.json file out of your Kaggle account settings by clicking “Create New Token.”
Subsequent, set the setting variables in your native system:
- KAGGLE_USERNAME=<username>
- KAGGLE_API_KEY=<key>
2. Competitions
Kaggle Competitions are hosted challenges the place you’ll be able to remedy machine studying issues, obtain information, submit predictions, and see your outcomes on the leaderboard.
The CLI helps you automate every little thing: searching competitions, downloading recordsdata, submitting options, and extra.
Record Competitions
kaggle competitions record -s
Reveals a listing of Kaggle competitions, optionally filtered by a search time period. Helpful for locating new challenges to hitch.
Record Competitors Recordsdata
kaggle competitions recordsdata
Shows all recordsdata out there for a selected competitors, so you understand what information is offered.
Obtain Competitors Recordsdata
kaggle competitions obtain [-f ] [-p ]
Downloads all or particular recordsdata from a contest to your native machine. Use -f to specify a file, -p to set the obtain folder.
Undergo a Competitors
kaggle competitions submit -f -m ""
Add your resolution file to a contest with an elective message describing your submission.
Record Your Submissions
kaggle competitions submissions
Reveals all of your earlier submissions for a contest, together with scores and timestamps.
View Leaderboard
kaggle competitions leaderboard [-s]
Shows the present leaderboard for a contest. Use -s to point out solely the highest entries.
3. Datasets
Kaggle Datasets are collections of knowledge shared by the neighborhood. The dataset CLI instructions aid you discover, obtain, and add datasets, in addition to handle dataset variations.
Record Datasets
Finds datasets on Kaggle, optionally filtered by a search time period. Nice for locating information in your tasks.
Record Recordsdata in a Dataset
Reveals all recordsdata included in a selected dataset, so you’ll be able to see what’s out there earlier than downloading.
Obtain Dataset Recordsdata
kaggle datasets obtain / [-f ] [--unzip]
Downloads all or particular recordsdata from a dataset. Use –unzip to mechanically extract zipped recordsdata.
Initialize Dataset Metadata
Creates a metadata file in a folder, making ready it for dataset creation or versioning.
Create a New Dataset
kaggle datasets create -p
Uploads a brand new dataset from a folder containing your information and metadata.
Create a New Dataset Model
kaggle datasets model -p -m ""
Uploads a brand new model of an current dataset, with a message describing the modifications.
4. Notebooks
Kaggle Notebooks are executable code snippets or notebooks. The CLI permits you to record, obtain, add, and verify the standing of those notebooks, which is helpful for sharing or automating evaluation.
Record Kernels
Finds public Kaggle notebooks (kernels) matching your search time period.
Get Kernel Code
Downloads the code for a selected kernel to your native machine.
Initialize Kernel Metadata
Creates a metadata file in a folder, making ready it for kernel creation or updates.
Replace Kernel
Uploads new code and runs the kernel, updating it on Kaggle.
Get Kernel Output
kaggle kernels output / -p
Downloads the output recordsdata generated by a kernel run.
Examine Kernel Standing
Reveals the present standing (e.g., working, full, failed) of a kernel.
5. Fashions
Kaggle Fashions are versioned machine studying fashions you’ll be able to share, reuse, or deploy. The CLI helps handle these fashions, from itemizing and downloading to creating and updating them.
Record Fashions
Finds public fashions on Kaggle matching your search time period.
Get a Mannequin
Downloads a mannequin and its metadata to your native machine.
Initialize Mannequin Metadata
Creates a metadata file in a folder, making ready it for mannequin creation.
Create a New Mannequin
Uploads a brand new mannequin to Kaggle out of your native folder.
Replace a Mannequin
Uploads a brand new model of an current mannequin.
Delete a Mannequin
Removes a mannequin from Kaggle.
6. Config
Kaggle CLI configuration instructions management default behaviors, resembling obtain places and your default competitors. Regulate these settings to make your workflow smoother.
View Config
Shows your present Kaggle CLI configuration settings (e.g., default competitors, obtain path).
Set Config
Units a configuration worth, resembling default competitors or obtain path.
Unset Config
Removes a configuration worth, reverting to default habits.
7. Suggestions
- Use -h or –help after any command for detailed choices and utilization
- Use -v for CSV output, -q for quiet mode
- You will need to settle for competitors guidelines on the Kaggle web site earlier than downloading or submitting to competitions
Abid Ali Awan (@1abidaliawan) is an authorized information scientist skilled who loves constructing machine studying fashions. Presently, he’s specializing in content material creation and writing technical blogs on machine studying and information science applied sciences. Abid holds a Grasp’s diploma in expertise administration and a bachelor’s diploma in telecommunication engineering. His imaginative and prescient is to construct an AI product utilizing a graph neural community for college students combating psychological sickness.