tf1

F1-score metrics for classification models in TensorFlow. There are 3 average modes provided:

binary
macro
micro

Usage

from tf1 import f1_binary

# use f1_binary as any other metric from tf.metrics.*

Note, that due to streaming nature of metric computation process, "macro" and "micro" average metrics should know total number of classes. Use them as follows:

from tf1 import f1_macro, f1_micro

def my_task_f1_macro(
    labels, predictions, num_classes, weights=None,
    metrics_collections=None, updates_collections=None,
    name=None):

    return f1_macro(
        labels=labels,
        predictions=predictions,
        num_classes=123,  # Required
        weights=weights,
        metrics_collections=metrics_collections,
        updates_collections=updates_collections,
        name=name
    )

# use my_task_f1_macro as any other metric from tf.metrics.*

Release files for tf1 1.1.1

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for tf1 1.1.1
File	Size	Uploaded
tf1-1.1.1.tar.gz	5.9 kB	Jan 29, 2019	Details

Release files / tf1-1.1.1.tar.gz

Download URL	tf1-1.1.1.tar.gz
Size	5.9 kB
Tags	Source
SHA-256 checksum How to use checksums	`3edd395f346453c549a5ee1c061c2bc14d8d64e5da09d826c3f52e2a67f78f9f`
BLAKE2b-256 checksum How to use checksums	`f5239bd29ec433ba776a80e57e04c374abfed9378fa3b44f77bfde6e4fc12254`
Upload date	Jan 29, 2019
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/1.12.1 pkginfo/1.4.2 requests/2.19.1 setuptools/39.1.0 requests-toolbelt/0.8.0 tqdm/4.23.4 CPython/3.6.5`