Downloads · 30 days
12
3% of all-time downloads
rbroc/twitter-eupol-model
twitter-eupol-model is a feature extraction model from rbroc. Use it when you need embeddings to search or compare text. It is set up for transformers. The card lists the license as mit.
DistilBERT model fine-tuned on 117567 English language tweets from a range of political agencies (EUCommission, UN, OECD, IMF, ECB, Council of the European Union, UK government, Scottish government). Fine-tuned with l…
Downloads · 30 days
12
3% of all-time downloads
All-time downloads
356
Public
Repo size
531 MB
Likes
0
Public
Click a slice to open those files.
.bin265 MB · 100%
From the Hugging Face model README
DistilBERT model fine-tuned on 117567 English language tweets from a range of political agencies (EU_Commission, UN, OECD, IMF, ECB, Council of the European Union, UK government, Scottish government). Fine-tuned with learning rate = 2e-5, 16-sample batches, chunk_size = 50 tokens, 100 epochs (early stopping patience = 3 epochs), 3 warmup epochs. More details can be found at: https://github.com/rbroc/eucomm-twitter. No evaluation was performed, as fine-tuning was merely functional to providing checkpoints for a contextualized topic model.