KnutJaegersberg / topic-classification-IPTC-subject-labels

huggingface.co
Total runs: 151
24-hour runs: 0
7-day runs: 27
30-day runs: 136
Model's Last Updated: March 16 2023
text-classification

Introduction of topic-classification-IPTC-subject-labels

Model Details of topic-classification-IPTC-subject-labels

IPTC topic classifier (multilingual)

A SetFit model fit on 166 downlsampled multilingual IPTC Subject labels (concatenated for the lowest hierarchy level into artificial sentences of keywords) to predict the mid level news categories. The purpose of this classifier is to support exploring corpora as weak labeler, since the representations of these descriptions are only approximations of real documents from those topics. The dataset I used to train the model is based on this file: https://huggingface.co/datasets/KnutJaegersberg/News_topics_IPTC_codes_long

Accuracy on highest level labels in eval: 0.9779412 Accuracy/F1/mcc on mid level labels in eval: 0.6992481/0.6666667/0.6992617

More interestingly, I used the kaggle dataset with headlines from huffington post and manually selected 15 overlapping high level categories to evaluate the performance. https://www.kaggle.com/datasets/rmisra/news-category-dataset

While mcc 0.1968043 on this dataset does not sound as good as before, the mistakes usually could also be seen as a re-interpretation. I.e. news on arrests where categorized as entertainment in the huffington post dataset, the classifier put it into the crime category. My current impression is this system is useful for the aimed for purpose.

The numeric categories can be joined with the labels by using this table: https://huggingface.co/datasets/KnutJaegersberg/IPTC-topic-classifier-labels

Looks like try out api box to the right by huggingface does not yet handle setfit models, can't do anything about that.

Use like any other SetFit model

from setfit import SetFitModel

Download from Hub and run inference

model = SetFitModel.from_pretrained("KnutJaegersberg/IPTC-classifier-ml")

Run inference

preds = model(["Rachel Dolezal Faces Felony Charges For Welfare Fraud", "Elon Musk just got lucky", "The hype on AI is different from the hype on other tech topics"])

Runs of KnutJaegersberg topic-classification-IPTC-subject-labels on huggingface.co

151
Total runs
0
24-hour runs
11
3-day runs
27
7-day runs
136
30-day runs

More Information About topic-classification-IPTC-subject-labels huggingface.co Model

topic-classification-IPTC-subject-labels huggingface.co

topic-classification-IPTC-subject-labels huggingface.co is an AI model on huggingface.co that provides topic-classification-IPTC-subject-labels's model effect (), which can be used instantly with this KnutJaegersberg topic-classification-IPTC-subject-labels model. huggingface.co supports a free trial of the topic-classification-IPTC-subject-labels model, and also provides paid use of the topic-classification-IPTC-subject-labels. Support call topic-classification-IPTC-subject-labels model through api, including Node.js, Python, http.

topic-classification-IPTC-subject-labels huggingface.co Url

https://huggingface.co/KnutJaegersberg/topic-classification-IPTC-subject-labels

KnutJaegersberg topic-classification-IPTC-subject-labels online free

topic-classification-IPTC-subject-labels huggingface.co is an online trial and call api platform, which integrates topic-classification-IPTC-subject-labels's modeling effects, including api services, and provides a free online trial of topic-classification-IPTC-subject-labels, you can try topic-classification-IPTC-subject-labels online for free by clicking the link below.

KnutJaegersberg topic-classification-IPTC-subject-labels online free url in huggingface.co:

https://huggingface.co/KnutJaegersberg/topic-classification-IPTC-subject-labels

topic-classification-IPTC-subject-labels install

topic-classification-IPTC-subject-labels is an open source model from GitHub that offers a free installation service, and any user can find topic-classification-IPTC-subject-labels on GitHub to install. At the same time, huggingface.co provides the effect of topic-classification-IPTC-subject-labels install, users can directly use topic-classification-IPTC-subject-labels installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

topic-classification-IPTC-subject-labels install url in huggingface.co:

https://huggingface.co/KnutJaegersberg/topic-classification-IPTC-subject-labels

Url of topic-classification-IPTC-subject-labels

topic-classification-IPTC-subject-labels huggingface.co Url

Provider of topic-classification-IPTC-subject-labels huggingface.co

KnutJaegersberg
ORGANIZATIONS

Other API from KnutJaegersberg