Modelpublic

vocabtrimmer/xlm-roberta-base-tweet-sentiment-de-trimmed-de-10000

sourceHugging Faceupdated 4y agoView on Hugging Face

0likes27downloads

Model Card

Vocabulary Trimmed cardiffnlp/xlm-roberta-base-tweet-sentiment-de: `vocabtrimmer/xlm-roberta-base-tweet-sentiment-de-trimmed-de-10000`

This model is a trimmed version of cardiffnlp/xlm-roberta-base-tweet-sentiment-de by `vocabtrimmer`, a tool for trimming vocabulary of language models to compress the model size. Following table shows a summary of the trimming process.

	cardiffnlp/xlm-roberta-base-tweet-sentiment-de	vocabtrimmer/xlm-roberta-base-tweet-sentiment-de-trimmed-de-10000
parametersizefull	278,045,955	93,725,955
parametersizeembedding	192,001,536	7,681,536
vocab_size	250,002	10,002
compressionratefull	100.0	33.71
compressionrateembedding	100.0	4.0

Following table shows the parameter used to trim vocabulary.

language	dataset	dataset_column	dataset_name	dataset_split	target_vocab_size	min_frequency
de	vocabtrimmer/mc4_validation	text	de	validation	10000	2