metadata

license: apache-2.0
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
model-index:
  - name: BERTUncased_Fake_news_classification
    results: []
widget:
  - text: >-
      Cop Shares Racist Meme About Michelle Obama; Now That Cop Is Having A VERY
      Bad Day (IMAGES)After the election of Donald Trump many folks seem to see
      it as a permission slip to be as racist and vile as possible. However,
      here s the thing, you re still going to get called out as racist and vile.
      And one Alabama police officer just found this out the hard way.According
      to the Washington Post: Talladega Police Officer Joel Husk was terminated
      Wednesday for violating the department s social media and code of conduct
      policies, City Manager Patrick Bryant said. What did he do? So glad you
      asked: Husk had posted several memes on his Facebook page, including one
      showing Obama and Melania Trump.  Fluent in Slovenian, English, French,
      Serbian, and German,  it said over Trump s photo. Over Obama s, it read: 
      Fluent in Ghetto. Not only that, he posted several extraordinarily racist
      memes:via Washington Postvia Washington PostAccording to the City Manager,
      the statements were  deemed to be biased or racially insensitive or
      derogatory  and because of that, they  have to take action to correct it.
      If you re going to be a police officer and serve all the public, you can t
      assume black people standing up for their rights are equivalent to the
      KKK. That s about the most horrific equivalence imaginable.Also, according
      to WaPo: Husk, 37, who had been with the department for about two and a
      half years, had also shared a meme showing President Obama with the
      words:  Was Dallas a terrorist attack? Yes! Carried out by Obama s own
      homegrown terrorist group! Which is a blatant lie and anyone who were to
      feel that way belongs nowhere near law enforcement. The city took the
      proper action letting this racist cop go, and hopefully it will be an
      example to police departments all over the country that this sort of
      behavior simply cannot be tolerated.Trump s election must not be allowed
      to serve as a permission slip to bigots everywhere that it s fine to be as
      awful as possible, because here in the land of the free and the home of
      the brave, everyone is protected. Everyone, regardless of color, class,
      gender, sexual orientation, or creed.Featured Photo by Chip
      Somodevilla/Getty Images
    example_title: Fake
  - text: >-
      Several homes have been evacuated as firefighters tackle a fire at an
      industrial building in Mansfield. Nottinghamshire Fire and Rescue Service
      said it was called to a building in Forest Road shortly before 15:45 GMT
      on Saturday.The service said it received a large number of calls from the
      public to report smoke coming from the area. An evacuation centre had been
      opened at a leisure centre in the town for residents waiting to be allowed
      home. More than 100 firefighters from 16 crews have been involved in the
      operation, with Nottingham Road, Forest Road and Sandhurst Avenue closed
      and several properties evacuated. Bus services have also been diverting
      from the area. Ch Insp Amy Styles-Jones, from Nottinghamshire Police,
      said: ‘Thankfully there are no reports of any injuries but neighbouring
      streets have been evacuated as a precaution.’‘The Oak Tree Leisure Centre,
      in Jubilee Way South, will open from 19:00 as a place of safety for
      members of the public affected by the fire.‘This incident is likely to be
      ongoing for some time so please avoid the area if possible.’
    example_title: 'True'

BERTUncased_Fake_news_classification

This model is a fine-tuned version of bert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.0027
Accuracy: 0.9997
F1: 0.9998
Percision: 0.9995

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 3e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Percision
0.0081	1.0	3865	0.0000	1.0	1.0	1.0
0.0015	2.0	7730	0.0027	0.9997	0.9998	0.9995

Framework versions

Transformers 4.27.1
Pytorch 1.13.1+cu116
Datasets 2.10.1
Tokenizers 0.13.2