deadbeats 0.2.0

No project description provided

DEADBEATS

An easy to use Slack messaging library for research.

Usage

from deadbeats import DEADBEATS
# use environment variables implicitly
# SLACK_ACCESS_TOKEN=xxxx-xxxxxxxxxxxxx-xxxxxxxxxxxxxx-xxxxxxxxxxxxxxxxxxxxxxxx (Get your own Slack API access token)
# SLACK_CHANNEL_ID=deadbeats (set slack channel id whatever you like!)


# or you can set configurations manually.
DEADBEATS.set_access_token("SLACK_ACCESS_TOKEN")
DEADBEATS.set_channel_id("SLACK_CHANNEL_ID")


# `DEADBEATS.wrap` sends a message at the beginning and end of the function.
# `DEADBEATS.wrap` catch every errors and raise it after sending a error message.
@DEADBEATS.wrap
def main():
    # A simple "heartbeating" message.
    DEADBEATS.ping()


    # You can add extra information like below.
    params = {"loss": 0.5, "val_loss": 1.6, "acc": 100.0}
    DEADBEATS.ping(text="message whatever you like", params=params, additional="info", huga="huga")


    # Start threading!
    # All subsequent messages will be sent to the thread.
    DEADBEATS.start_thread()


    # If you want to stop threading, you can use this method.
    # This method reset "thread_ts" of a instance variable, which is a id of thread.
    DEADBEATS.reset_thread()

Research Usage with PyTorch Lightning

class MyModel(pl.LightningModule):
    def validation_epoch_end()

The library is named after the wonderful work of Mori Calliope, DEAD BEATS.