This checkpoint should be loaded into BartForConditionalGeneration.frompretrained. See the BART docs for more information.
Open-weight model · Summarization
rut5_base_headline_gen_telegram
by Ilya Gusev IlyaGusev/rut5_base_headline_gen_telegram
Based on rut5-base model
Parameters—
Context—
Weights977.4 MB
Licenseapache-2.0
AccessOpen weights
Monthly Downloads5.5k
Model Card
By Ilya Gusev, published under apache-2.0, revision 7ee21f469a71.
Based on rut5-base model
Read Ilya Gusev's full model card
RuT5TelegramHeadlines
Model description
Based on rut5-base model
Intended uses & limitations
How to use
from transformers import AutoTokenizer, T5ForConditionalGeneration
model_name = "IlyaGusev/rut5_base_headline_gen_telegram"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = T5ForConditionalGeneration.from_pretrained(model_name)
article_text = "..."
input_ids = tokenizer(
[article_text],
max_length=600,
add_special_tokens=True,
padding="max_length",
truncation=True,
return_tensors="pt"
)["input_ids"]
output_ids = model.generate(
input_ids=input_ids
)[0]
headline = tokenizer.decode(output_ids, skip_special_tokens=True)
print(headline)
Training data
- Dataset: ru_all_split.tar.gz
Training procedure
- Training script: train.py
Configuration
- Architecture
- T5ForConditionalGeneration
- Vocabulary size
- 30,000
- Stored precision
- float32
- Model type
- t5
Identity and Version
- Repository
- IlyaGusev/rut5_base_headline_gen_telegram
- Publisher
- Ilya Gusev
- Task
- Summarization
- Modality
- Text
- Library
- transformers
- Parameters
- Not stated by the source
- Languages
- ru
- Revision
- 7ee21f469a71b19a107eef60766bba8ad29bdbb7
- First published
- 2022-03-02
- Last updated
- 2021-12-18
Files and Weights
8 files, 979.5 MB in total. The weights are 1 file totalling 977.4 MB in bin.
Weights1 file · 977.4 MB
Configuration2 files · 831 B
Tokenizer3 files · 2.1 MB
Documentation1 file · 9.8 KB
Repository1 file · 1.3 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| pytorch_model.bin | Weights | 977.4 MB | a42ea9bd0123 |
| config.json | Configuration | 766 B | — |
| special_tokens_map.json | Configuration | 65 B | — |
| README.md | Documentation | 9.8 KB | — |
| .gitattributes | Repository | 1.3 KB | — |
| spiece.model | Tokenizer | 827.6 KB | 76927654c70c |
| tokenizer.json | Tokenizer | 1.3 MB | — |
| tokenizer_config.json | Tokenizer | 327 B | — |
License and Download
- License
- apache-2.0
- Access
- Open weights, no gate
- Download size
- 977.4 MB
Download from Ilya Gusev
Released by Ilya Gusev through its official repository on Hugging Face. Read the license.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 977.4 MB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About rut5_base_headline_gen_telegram
Can I use rut5_base_headline_gen_telegram commercially?
Yes. rut5_base_headline_gen_telegram is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.
Similar Models
Original TF 1 code here Authors: Jingqing Zhang, Yao Zhao, Mohammad Saleh and Peter J. Liu on Dec 18, 2019 The following is copied from the authors' README. We train a pegasus model with sampled gap sentence ratios on both C4 and HugeNews, and stochastically sample important sentences. The updated the results are reported in this table. The "Mixed & Stochastic" model has the following changes: - trained on both C4 and HugeNews (dataset mixture is weighted by their number of examples). - trained for 1.5M instead of 500k (we observe slower convergence on pretraining perplexity). - the model uniformly sample a gap sentence ratio between 15% and 45%. - importance sentences are sampled using a…
This model was trained using Amazon SageMaker and the new Hugging Face Deep Learning container.
This checkpoint should be loaded into BartForConditionalGeneration.frompretrained. See the BART docs for more information.
This repository contains the mT5 checkpoint finetuned on the 45 languages of XL-Sum dataset. For finetuning details and scripts, see the paper and the official repository. Scores on the XL-Sum test sets are as follows: Language | ROUGE-1 / ROUGE-2 / ROUGE-L Amharic | 20.0485 / 7.4111 / 18.0753 Arabic | 34.9107 / 14.7937 / 29.1623 Azerbaijani | 21.4227 / 9.5214 / 19.3331 Bengali | 29.5653 / 12.1095 / 25.1315 Burmese | 15.9626 / 5.1477 / 14.1819 Chinese (Simplified) | 39.4071 / 17.7913 / 33.406 Chinese (Traditional) | 37.1866 / 17.1432 / 31.6184 English | 37.601 / 15.1536 / 29.8817 French | 35.3398 / 16.1739 / 28.2041 Gujarati | 21.9619 / 7.7417 / 19.86 Hausa | 39.4375 / 17.6786 / 31.6667…