senda ![](sendalogo.png)
![License](https://img.shields.io/badge/license-MIT-blue.svg)
senda
is a small python package for fine-tuning transformers for
sentiment analysis (and text classification tasks in general).
senda
builds on the excellent transformers.Trainer
API (all credit goes to the Huggingface
team).
Installation guide
senda
can be installed from PyPI with
pip install senda
If you want the development version then install directly from GitHub.
How to use
You can use senda
to fine-tune any transformer for any text classification task in any language.
Here we will go through how to use senda
for fine-tuning a transformer for detecting the polarity ('positive', 'neutral' or 'negative')
of Danish Tweets. For training we use more than 5,000 Danish Tweets kindly annotated
and hosted by the Alexandra Institute (thanks!).
First, load Danish Tweets annotated with polarity.
from senda import get_danish_tweets
df_train, df_eval, df_test = get_danish_tweets()
Note, that the datasets must be DataFrames containing the columns 'text' and 'label', e.g.
df_train
text label
Cepos: Vi bør diskutere, hvordan vi afvikler j... neutral
Avis: FC København og Brøndby IF i duel om Ste... neutral
@PeterThorup @IntactDenmark Nej - endnu ikke -... positiv
That was pretty close. Theresa May fortsætter ... neutral
Så er der ny Facebook-side til min nye forretn... positiv
... ...
@MtnTeit @aeldresagen Helt enig. Vi må bare ik... negativ
@PrmMortensen @Marchen_Neel @larsloekke @oeste... negativ
Hvordan sikrer vi ØKONOMISK RENTABLE REGIONALE... neutral
@JanEJoergensen @24syv @DanskDf1995 @Spolitik ... negativ
@Fonoudi6eren Ikke enig! Synes vi var godt med... positiv
Next, instantiate and set up the model.
from senda import Model, compute_metrics
from transformers import EarlyStoppingCallback
m = Model(train_dataset = df_train,
eval_dataset = df_eval,
transformer = "Maltehb/danish-bert-botxo",
labels = ['negativ', 'neutral', 'positiv'],
tokenize_args = {'padding':True, 'truncation':True, 'max_length':512},
training_args = {"output_dir":'./results',
"num_train_epochs": 4,
"per_device_train_batch_size":8,
"evaluation_strategy":"steps",
"eval_steps":100,
"logging_steps":100,
"learning_rate":2e-05,
"weight_decay": 0.01,
"per_device_eval_batch_size":32,
"warmup_steps":100,
"seed":42,
"load_best_model_at_end":True,
},
trainer_args = {'compute_metrics': compute_metrics,
'callbacks':[EarlyStoppingCallback(early_stopping_patience=4)],
}
)
Now, all there is left is to initialize the model (including the transformers.Trainer
) and train it:
m.init()
m.train()
The model can then be evaluated on the test set:
m.evaluate(df_test)
{'eval_loss': 0.5771588683128357, 'eval_accuracy': 0.7664399092970522, 'eval_f1': 0.7290485787279956, 'eval_runtime': 4.2016, 'eval_samples_per_second': 104.959}
Predict new observations:
text = "Sikke en dejlig dag det er i dag"
m.predict(text)
PredictionOutput(predictions=array([[-1.2986785 , -0.31318122, 1.2002046 ]], dtype=float32), label_ids=array([0]), metrics={'test_loss': 2.7630457878112793, 'test_accuracy': 0.0, 'test_f1': 0.0, 'test_runtime': 0.07, 'test_samples_per_second': 14.281})
m.predict(text, return_labels=True)
['positiv']
senda
model available on Huggingface
As you see, the model above achieves an accuracy of 0.77 and a macro-averaged F1-score of 0.73 on a small test data set, that Alexandra Institute provides.
The model is published on Huggingface.
Here is how to download and use the model with PyTorch:
from transformers import AutoTokenizer, AutoModelForSequenceClassification, pipeline
tokenizer = AutoTokenizer.from_pretrained("pin/senda")
model = AutoModelForSequenceClassification.from_pretrained("pin/senda")
senda_pipeline = pipeline('sentiment-analysis', model=model, tokenizer=tokenizer)
senda_pipeline("Sikke en dejlig dag det er i dag")
[{'label': 'positiv', 'score': 0.7678486704826355}]
The model can most certainly be improved, and we encourage all NLP-enthusiasts to train a better model - you can use the senda
package to do this.
Background
senda
is developed as a part of Ekstra Bladet’s activities on Platform Intelligence in News (PIN). PIN is an industrial research project that is carried out in collaboration between the Technical University of Denmark, University of Copenhagen and Copenhagen Business School with funding from Innovation Fund Denmark. The project runs from 2020-2023 and develops recommender systems and natural language processing systems geared for news publishing, some of which are open sourced like senda
.
Shout-outs
Contact
We hope, that you will find senda
useful.
Please direct any questions and feedbacks to
us!
If you want to contribute (which we encourage you to), open a
PR.
If you encounter a bug or want to suggest an enhancement, please
open an issue.