Name		Name	Last commit message	Last commit date
parent directory ..
README.md		README.md
api_call_debug.py		api_call_debug.py
api_call_demo.py		api_call_demo.py
bert_inference.py		bert_inference.py
emissions.csv		emissions.csv
logging_to_file.py		logging_to_file.py
logging_to_google_cloud.py		logging_to_google_cloud.py
mnist-comet.py		mnist-comet.py
mnist-sklearn.py		mnist-sklearn.py
mnist.py		mnist.py
mnist_callback.py		mnist_callback.py
mnist_context_manager.py		mnist_context_manager.py
mnist_decorator.py		mnist_decorator.py
mnist_grid_search.py		mnist_grid_search.py
mnist_inference.py		mnist_inference.py
mnist_random_search.py		mnist_random_search.py
notebook.ipynb		notebook.ipynb
prometheus_call.py		prometheus_call.py
pue.py		pue.py
pytorch-multigpu-example.py		pytorch-multigpu-example.py
requirements-examples.txt		requirements-examples.txt
task_inference.py		task_inference.py

README.md

Getting Started

Setup

The following examples use Keras from TensorFlow 2.0. The dependencies can be installed as follows

pip install -r requirements-examples.txt

Examples

mnist.py: Usage using explicit CO2Tracker objects.
mnist_decorator.py: Using the @track_co2 decorator.
mnist_callback.py: Using Keras callbacks to save emissions after each epoch.
mnist-comet.py: Using CO2Tracker with Comet for automatic experiment and emissions tracking.
api_call_demo.py: Simplest demo to send computer emissions to CodeCarbon API.
api_call_debug.py: Script to send computer emissions to CodeCarbon API. Made for debugging: debug log and send data every 20 seconds.