language: cs
license: cc-by-nc-sa-4.0
tags:
- Czech
- RoBERTa
- ÚFAL
Model Card for RobeCzech
Model Details
Model Description
RobeCzech is a monolingual RoBERTa language representation model trained on Czech data.
- Developed by: Institute of Formal and Applied Linguistics, Charles University, Prague (UFAL)
- Shared by [Optional]: Hugging Face and LINDAT
- Model type: Fill-Mask
- Language(s) (NLP): cs
- License: cc-by-nc-sa-4.0
- Related Models: More information needed
- Parent Model: RoBERTa
- Resources for more information:
Uses
Direct Use
Fill-Mask tasks.
Downstream Use [Optional]
Morphological tagging and lemmatization, dependency parsing, named entity recognition and semantic parsing.
Out-of-Scope Use
More information needed
Bias, Risks, and Limitations
Significant research has explored bias and fairness issues with language models (see, e.g., Sheng et al. (2021) and Bender et al. (2021)). Predictions generated by the model may include disturbing and harmful stereotypes across protected classes; identity characteristics; and sensitive, social, and occupational groups.
Recommendations
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
Training Details
Training Data
The model creators note in the associated paper:
We trained RobeCzech on a collection of the following publicly available texts: - SYN v4, a large corpus of contemporary written Czech, 4,188M tokens; - Czes, a collection of Czech newspaper and magazine articles, 432M tokens; - documents with at least 400 tokens from the Czech part of the web corpus.W2C , tokenized with MorphoDiTa, 16M tokens; - plain texts extracted from Czech Wikipedia dump 20201020 using WikiEx-tractor, tokenized with MorphoDiTa, 123M tokens
All these corpora contain whole documents, even if the SYN v4 is block-shuffled (blocks with at most 100 words respecting sentence boundaries are permuted in a document) and in total contain 4,917M tokens.
Training Procedure
Preprocessing
The texts are tokenized into subwords with a byte-level BPE (BBPE) to- kenizer [33]. The tokenizer is trained on the entire corpus and we limit its vocabulary size to 52,000 items.
Speeds, Sizes, Times
The model creators note in the associated paper:
The training batch size is 8,192 and each training batch consists of sentences sampled contiguously, even across document boundaries, such that the total length of each sample is at most 512 tokens (FULL-SENTENCES setting). We use Adam optimizer with β1 = 0.9 and β2 = 0.98 to minimize the masked language-modeling objective.
Evaluation
Testing Data, Factors & Metrics
Testing Data
The model creators note in the associated paper:
We evaluate RobeCzech in five NLP tasks, three of them leveraging frozen contextualized word embeddings, two approached with fine-tuning: morphological analysis and lemmatization: frozen contextualized word embeddings, dependency parsing: frozen contextualized word embeddings, named entity recognition: frozen contextualized word embeddings, semantic parsing: fine-tuned, sentiment analysis: fine-tuned.
Factors
More information needed
Metrics
Morphosynt PDT3.5 (POS) | Morphosynt PDT3. (LAS) | Morphosynt UD2.3 (XPOS) | Morphosynt UD2.3 ( LAS) | NER CNEC1.1 (nested) | NER CNEC1.1 (flat) | Semant. PTG (Avg) | Sentim. CDF (F1) |
Results
| Model | Morphosynt PDT3.5 (POS) (LAS) | Morphosynt UD2.3 (XPOS) (LAS) | NER CNEC1.1 (nested) (flat) | Semant. PTG (Avg) (F1) |
|---|---|---|---|---|
| RobeCzech | 98.50 91.42 | 98.31 93.77 | 87.82 87.47 | 92.36 80.13 |
Model Examination
More information needed
Environmental Impact
Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).
- Hardware Type: 8 QUADRO P5000 GPU
- Hours used: 2190 (~3 months)
- Cloud Provider: More information needed
- Compute Region: More information needed
- Carbon Emitted: More information needed
Technical Specifications [optional]
Model Architecture and Objective
The model creators note in the associated paper: We employ the standard text classification architecture consisting of a BERT encoder, followed by a softmax-activated classification layer processing the computed embedding of the given document text obtained from the CLS token embedding from the last layer
Compute Infrastructure
Hardware
8 QUADRO P5000 GPU
Software
More information needed
Citation
APA:
Straka, M., Náplava, J., Straková, J., & Samuel, D. (2021). RobeCzech: Czech RoBERTa, a monolingual contextualized language representation model. arXiv. https://doi.org/10.1007/978-3-030-83527-9_17
Glossary [optional]
More information needed
More Information [optional]
More information needed
Model Card Authors [optional]
Institute of Formal and Applied Linguistics, Charles University, Prague (UFAL), in collaboration with Ezi Ozoani and the Hugging Face Team.
Model Card Contact
More information needed
How to Get Started with the Model
Use the code below to get started with the model.
Click to expand
from transformers import AutoTokenizer, AutoModelForMaskedLM
tokenizer = AutoTokenizer.from_pretrained("ufal/robeczech-base")
model = AutoModelForMaskedLM.from_pretrained("ufal/robeczech-base")