{"id":18791006,"url":"https://github.com/hansbug/genshin-nlp-encoding","last_synced_at":"2025-12-27T14:30:14.478Z","repository":{"id":182511410,"uuid":"667304154","full_name":"HansBug/genshin-nlp-encoding","owner":"HansBug","description":"NLP Encoder fine-tuning for Genshin Impact","archived":false,"fork":false,"pushed_at":"2023-10-25T14:25:22.000Z","size":47,"stargazers_count":1,"open_issues_count":0,"forks_count":1,"subscribers_count":2,"default_branch":"main","last_synced_at":"2024-12-29T15:10:15.479Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/HansBug.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null}},"created_at":"2023-07-17T07:52:05.000Z","updated_at":"2023-07-28T08:58:53.000Z","dependencies_parsed_at":"2023-07-26T00:16:13.110Z","dependency_job_id":"7b520125-bad2-44e7-bbd9-4f594a954a40","html_url":"https://github.com/HansBug/genshin-nlp-encoding","commit_stats":null,"previous_names":["hansbug/genshin-nlp-encoding"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/HansBug%2Fgenshin-nlp-encoding","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/HansBug%2Fgenshin-nlp-encoding/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/HansBug%2Fgenshin-nlp-encoding/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/HansBug%2Fgenshin-nlp-encoding/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/HansBug","download_url":"https://codeload.github.com/HansBug/genshin-nlp-encoding/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":239718105,"owners_count":19685718,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-11-07T21:14:24.830Z","updated_at":"2025-12-27T14:30:14.409Z","avatar_url":"https://github.com/HansBug.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# genshin-nlp-encoding\n\nNLP Encoder fine-tuning for Genshin Impact\n\n## Install\n\n```shell\ngit clone https://github.com/HansBug/genshin-nlp-encoding.git\ncd genshin-nlp-encoding\npip install -r requirements.txt\n```\n\n## Train\n\n```python\nfrom ditk import logging\n\nfrom gende.train import train\n\nif __name__ == '__main__':\n    logging.try_init_root(logging.INFO)\n    train(\n        'runs/bert_mean_seed_0',\n        model_name='bert_mean',\n        datasource='annotation.xlsx',\n        seed=0,\n        max_epochs=50\n    )\n\n```\n\n## Infer\n\n```python\nimport torch\nfrom huggingface_hub import hf_hub_download\nfrom torch import nn\nfrom treevalue import FastTreeValue\n\nfrom gende import tokenize\nfrom gende.models import load_model_from_ckpt\n\n\nclass ModelWithEncoding(nn.Module):\n    def __init__(self, model):\n        nn.Module.__init__(self)\n        self.model = model\n\n    def forward(self, x):\n        encoded = self.model.encoder(x)\n        predicted = self.model.ft(encoded)\n\n        return FastTreeValue({\n            'encoded': encoded,\n            'predicted': predicted,\n        })\n\n\n_torch_stack = FastTreeValue.func(subside=True)(torch.stack)\n\nif __name__ == '__main__':\n    # use your ckpt here!\n    ckpt_file = hf_hub_download(\n        'HansBug/genshin-nlp-finetuning',\n        'bert_mean_v1.1_seed_0/model.ckpt',\n    )\n    m = load_model_from_ckpt(ckpt_file)\n    m = ModelWithEncoding(m)\n    print(m)\n\n    with torch.no_grad():\n        # batch input\n        input_ = _torch_stack([\n            tokenize(\n                \"\"\"\n        \"Combat Action: When your active character is Electro Hypostasis, heal that character for 3 HP and attach the Electro Crystal Core to them.\n        (You must have Electro Hypostasis in your deck to add this card to your deck.)\"\n                \"\"\", tokenizer_name='bert',\n            ),\n            tokenize(\n                \"\"\"\n        \"The character deals +1 DMG.\n        When your character triggers an Elemental Reaction: Deal +1 DMG. (Twice per Round)\n        (Only Catalyst Characters can equip this. A character can equip a maximum of 1 Weapon)\"\n                \"\"\", tokenizer_name='bert',\n            ),\n        ])\n\n        print(input_)\n        output = m(input_)\n        print(output)\n\n```\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhansbug%2Fgenshin-nlp-encoding","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fhansbug%2Fgenshin-nlp-encoding","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhansbug%2Fgenshin-nlp-encoding/lists"}