Hugging Face's logo

ModerRAS
/
AniFileBERT

Token Classification
Transformers
ONNX
Safetensors
English
Japanese
Chinese
bert
anime
filename-parsing
Eval Results (legacy)
Model card Files
xet
 Community
AniFileBERT

Commit History

  1. Add robust LLM relabel pipeline and enforce contiguous title

    ModerRAS
    fed9d99

  2. set gpt-5.4-mini default token pricing

    ModerRAS
    8165cc3

  3. feat: add concurrency and token/cost telemetry for llm relabel runs

    ModerRAS
    7c5b41e

  4. feat: add tool-call based llm relabel pipeline and update dataset pointer

    ModerRAS
    e964ae5

  5. chore: update AnimeName submodule checkpoint

    ModerRAS
    1804249

  6. chore: checkpoint current training and manual relabel progress

    ModerRAS
    efb213a

  1. Add tests for combo generator

    ModerRAS
    1bef818

  2. Add balanced focus dataset generator

    ModerRAS
    651ad49

  3. Add path-aware focus dataset support

    ModerRAS
    1e1bc1f

  1. Train virtual-shard anime parser

    ModerRAS
    359ff82

  1. Train augmented anime filename parser

    ModerRAS
    be6a29a

  1. Organize parser modules and tools

    ModerRAS
    8c50d16

  2. Clean stale local scripts

    ModerRAS
    f95ce71

  3. Remove structural parser rule assists

    ModerRAS
    116c87c

  4. Document training lineage

    ModerRAS
    f2ec095

  5. Train thin anime filename parser

    ModerRAS
    f712f4b

  1. Remove stale diagnostics reports

    ModerRAS
    76e084f

  2. Add inference performance benchmark

    ModerRAS
    ce3a60d

  3. Polish Hugging Face repository docs

    ModerRAS
    376db19

  1. Clean special code parsing

    ModerRAS
    e8412e3

  2. Fix GM-Team bilingual title parsing

    ModerRAS
    adf92db

  3. Improve anime filename parser model

    ModerRAS
    e63569d

  1. Add parser diagnostics and inference debugging⁠

    ModerRAS
    b57780c

  2. Add Codex Colab training workflow

    ModerRAS
    e458112

  3. Add agent repository guidelines

    ModerRAS
    beb8c7e

  4. Mirror character vocab in model repo

    ModerRAS
    f408729

  1. Add character-token DMHY training path

    ModerRAS
    0779202

  2. colab: onnx导出改为非阻塞+补全onnxscript

    ModerRAS
    e34dc04

  3. colab: 加上onnx/onnxruntime依赖

    ModerRAS
    c5ba3cb

  4. 修复colab_train.py shell引号嵌套导致的Syntax error

    ModerRAS
    4df510c

  5. 重建词表：从632K dmhy_weak.jsonl统计频率取top8000，覆盖96.2%⁠

    ModerRAS
    410e000

  6. 重新训练模型，补充训练结果

    ModerRAS
    24a2cb6

  7. 提高训练速度并使用GPU进行训练

    ModerRAS
    3a13ca3

  8. 完成整个数据集的整理

    ModerRAS
    f4f4e0e

  1. Document AniFileBERT maintenance workflow

    ModerRAS
    3197202

  2. Add AniFileBERT model and training project⁠

    ModerRAS
    be5f706

  3. initial commit

    ModerRAS
    Verified signature
    a94e250