| --- |
| license: mit |
| language: |
| - en |
| - zh |
| library_name: transformers |
| tags: |
| - translation |
| - fine tune |
| - fine_tune |
| widget: |
| - text: >- |
| I {i}should{/i} say that I feel a little relieved to find out that |
| {i}this{/i} is why you’ve been hanging out with Kaori lately, though. She’s |
| really pretty and I got jealous and...I’m sorry. |
| --- |
| |
| # Normal1919/mbart-large-50-one-to-many-lil-fine-tune |
|
|
| * base model: mbart-large-50 |
| * pretrained_ckpt: facebook/mbart-large-50-one-to-many-mmt |
| * This model was trained for [rpy dl translate](https://github.com/O5-7/rpy_dl_translate) |
| |
| ## Model description |
| * source group: English |
| * target group: Chinese |
| * model: transformer |
| * source language(s): eng |
| * target language(s): cjy_Hans cjy_Hant cmn cmn_Hans cmn_Hant gan lzh lzh_Hans nan wuu yue yue_Hans yue_Hant |
| * fine_tune: On the basis of mbart-large-50-one-to-many-mmt checkpoints, train English original text with renpy text features (including but not limited to {i} [text] {/i}) to Chinese with the same reserved flag, as well as training for English name retention for LIL |
| |
| ## How to use |
| ```python |
| >>> from transformers import MBartForConditionalGeneration, MBart50TokenizerFast |
| >>> mode_name = 'Normal1919/mbart-large-50-one-to-many-lil-fine-tune' |
| >>> model = MBartForConditionalGeneration.from_pretrained(mode_name) |
| >>> tokenizer = MBart50TokenizerFast.from_pretrained(mode_name, src_lang="en_XX", tgt_lang="zh_CN") |
| >>> translation = pipeline("Marian-NMT-en-zh-lil-fine-tune", model=model, tokenizer=tokenizer) |
| >>> translation('I {i} should {/i} say that I feel a little relieved to find out that {i}this {/i} is why you’ve been hanging out with Kaori lately, though. She’s really pretty and I got jealous and...I’m sorry', max_length=400) |
| [{'我{i}应该{/i}说发现{i}这{/i}是你最近和Kaori出去的原因,我有点松了一口气。她很漂亮,我嫉妒,而且......我很抱歉。'}] |
| ``` |
| |
| ## Contact |
| |
| 517205163@qq.com or |
| a4564563@gmail.com |
