From ff7ce5da7cb4d2c322e48e830b82a7066acd1411 Mon Sep 17 00:00:00 2001 From: KUAI Date: Thu, 23 Jul 2026 23:03:43 +0800 Subject: [PATCH 1/5] Create easyocr_models.py --- tools/easyocr_models.py | 318 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 318 insertions(+) create mode 100644 tools/easyocr_models.py diff --git a/tools/easyocr_models.py b/tools/easyocr_models.py new file mode 100644 index 00000000..f75cf6bd --- /dev/null +++ b/tools/easyocr_models.py @@ -0,0 +1,318 @@ +""" +EasyOCR 模型下载助手（支持 GitHub/EdgeOne/ModelScope 镜像） +GitHub 源：从官方地址自动下载 zip 并解压提取 .pth 文件。 +EdgeOne 源：从 EdgeOne 镜像下载模型文件，由 kuai 提供。 +ModelScope 源：从 ModelScope 镜像下载模型文件，由 Ceceliachenen 提供。 +""" + +import hashlib +import importlib.util +import os +import shutil +import subprocess +import sys +import urllib.parse +import urllib.request +import zipfile +from typing import Any + + +# -------------------- 常量 -------------------- +MODEL_FILES = ['craft_mlt_25k.pth', 'zh_sim_g2.pth'] + +EXPECTED_MD5 = { + 'craft_mlt_25k.pth': '2f8227d2def4037cdb3b34389dcf9ec1', + 'zh_sim_g2.pth': 'b601ce7143293387d3ec4f41a66edc07', +} + +MIRROR_OPTIONS: dict[str, dict[str, Any]] = { + '0': { + 'name': 'GitHub', + 'type': 'github', + 'urls': { + 'craft_mlt_25k.pth': 'https://github.com/JaidedAI/EasyOCR/releases/download/pre-v1.1.6/craft_mlt_25k.zip', + 'zh_sim_g2.pth': 'https://github.com/JaidedAI/EasyOCR/releases/download/v1.3/zh_sim_g2.zip', + }, + }, + '1': { + 'name': 'EdgeOne', + 'base_url': 'http://easyocr.v.ekuai.tech/', + 'type': 'direct', + }, + '2': { + 'name': 'ModelScope', + 'type': 'modelscope', + }, +} + +MODELSCOPE_MODEL = 'Ceceliachenen/easyocr' + + +# -------------------- 工具函数 -------------------- +def get_input(prompt: str, default: str = '0', choices: list[str] | None = None) -> str: + while True: + user_input = input(prompt).strip() + if not user_input: + user_input = default + if choices is None or user_input in choices: + return user_input + print(f'无效输入，请从 {choices} 中选择。') + + +def print_step(msg: str) -> None: + print(f'\n{"-" * 60}\n>>> {msg}\n{"-" * 60}') + + +def ensure_dir(path: str) -> None: + os.makedirs(path, exist_ok=True) + print(f'目录已就绪: {path}' if os.path.isdir(path) else f'已创建目录: {path}') + + +def _download_hook(count: int, blk: int, total: int) -> None: + if total > 0: + pct = min(100, count * blk * 100 / total) + sys.stdout.write(f'\r下载进度: {pct:.1f}%') + sys.stdout.flush() + if pct >= 100: + sys.stdout.write('\n') + + +def download_file(url: str, dest: str) -> None: + """通用文件下载（带进度）""" + print(f'下载: {url} -> {dest}') + parsed = urllib.parse.urlparse(url) + if parsed.scheme not in ('http', 'https'): + raise ValueError(f'不支持的 URL 协议: {parsed.scheme}') + try: + urllib.request.urlretrieve(url, dest, _download_hook) # noqa: S310 + print('下载完成') + except Exception as e: + print(f'下载失败: {e}') + raise + + +def md5(path: str) -> str: + h = hashlib.md5() + with open(path, 'rb') as f: + for chunk in iter(lambda: f.read(4096), b''): + h.update(chunk) + return h.hexdigest() + + +def download_modelscope(filename: str, local_dir: str = '.') -> None: + """使用 ModelScope Python API 下载单个文件""" + try: + from modelscope.hub.file_download import model_file_download + except ImportError as err: + raise RuntimeError('modelscope 安装不完整，缺少 hub.file_download 模块') from err + print(f'ModelScope API 下载: {filename} -> {local_dir}') + model_file_download( + model_id=MODELSCOPE_MODEL, + file_path=filename, + local_dir=local_dir, + ) + print('下载完成') + + +def download_edgeone_split(base_url: str, filename: str, dest: str) -> None: + """EdgeOne 分片下载并合并""" + parts = 4 + tmp_parts: list[str] = [] + print(f'EdgeOne 分片下载: {filename} (共 {parts} 部分)') + for i in range(1, parts + 1): + suffix = f'.part_{i:03d}' + part_url = base_url + filename + suffix + part_local = os.path.join(os.getcwd(), filename + suffix) + print(f'\n[部分 {i}/{parts}]') + download_file(part_url, part_local) + tmp_parts.append(part_local) + + print('\n合并分片...') + with open(dest, 'wb') as out: + for p in tmp_parts: + with open(p, 'rb') as f: + data = f.read() + out.write(data) + print(f'已合并: {os.path.basename(p)} ({len(data) / 1024 / 1024:.2f} MB)') + print('合并完成，删除临时分片...') + for p in tmp_parts: + os.remove(p) + + +def download_github_zip(zip_url: str, pth_name: str, dest_path: str) -> None: + """下载 zip 文件并解压提取指定的 .pth 模型文件""" + tmp_zip = os.path.join(os.getcwd(), pth_name + '.zip') + download_file(zip_url, tmp_zip) + + print(f'解压 {os.path.basename(tmp_zip)} ...') + with zipfile.ZipFile(tmp_zip, 'r') as zf: + pth_files = [f for f in zf.namelist() if f.endswith('.pth')] + if not pth_files: + os.remove(tmp_zip) + raise RuntimeError(f'压缩包中未找到任何 .pth 文件: {tmp_zip}') + + target = _find_pth_target(pth_files, pth_name) + print(f'提取文件: {target}') + + with zf.open(target) as src, open(dest_path, 'wb') as dst: + shutil.copyfileobj(src, dst) + + os.remove(tmp_zip) + print(f'解压完成，模型文件保存为: {dest_path}') + + +def _find_pth_target(pth_files: list[str], pth_name: str) -> str: + """在 zip 文件列表中查找匹配的 .pth 文件""" + for f in pth_files: + if os.path.basename(f) == pth_name: + return f + return pth_files[0] + + +# -------------------- 主流程 -------------------- +def check_easyocr() -> bool: + """检测 EasyOCR 环境，返回是否已安装""" + print_step('1. 检测 EasyOCR 环境') + if importlib.util.find_spec('easyocr') is not None: + print('EasyOCR 已安装。') + return True + + print('未检测到 EasyOCR') + opt = get_input('[0] 跳过继续下载 (默认) [1] 退出: ', '0', ['0', '1']) + if opt == '1': + sys.exit(0) + print('跳过安装，将继续下载模型。') + return False + + +def prepare_model_dir() -> str: + """准备模型目录并返回路径""" + print_step('2. 准备模型目录') + model_dir = os.path.join(os.path.expanduser('~'), '.EasyOCR', 'model') + print(f'目标: {model_dir}') + ensure_dir(model_dir) + return model_dir + + +def select_mirror() -> tuple[str, dict[str, Any]]: + """选择下载镜像源，返回 (choice, mirror)""" + print_step('3. 选择下载镜像源') + for k, v in MIRROR_OPTIONS.items(): + print(f' [{k}] {v["name"]}') + choice = get_input('请输入数字 (默认0): ', '0', ['0', '1', '2']) + mirror = MIRROR_OPTIONS[choice] + + if choice == '2': + _ensure_modelscope() + + return choice, mirror + + +def _ensure_modelscope() -> None: + """确保 modelscope 已安装""" + if importlib.util.find_spec('modelscope') is not None: + print('modelscope 已安装。') + return + + print('未安装 modelscope') + ins = get_input('自动安装? [0] 是 (默认) [1] 否: ', '0', ['0', '1']) + if ins == '0': + print('安装 modelscope ...') + subprocess.check_call([sys.executable, '-m', 'pip', 'install', 'modelscope']) + print('安装成功') + else: + print('请手动安装 modelscope 后重试。') + sys.exit(0) + + +def _download_one(fname: str, choice: str, mirror: dict[str, Any], tmp: str) -> None: + """下载单个模型文件""" + try: + if mirror['type'] == 'github': + download_github_zip(mirror['urls'][fname], fname, tmp) + elif choice == '1' and fname == 'craft_mlt_25k.pth': + download_edgeone_split(mirror['base_url'], fname, tmp) + elif mirror['type'] == 'direct': + download_file(mirror['base_url'] + fname, tmp) + elif mirror['type'] == 'modelscope': + download_modelscope(fname, os.getcwd()) + else: + print(f'未知下载方式: {mirror["type"]}') + sys.exit(1) + except Exception as e: + print(f'下载或解压失败: {e}') + sys.exit(1) + + +def _verify_md5(fname: str, tmp: str) -> None: + """校验文件 MD5""" + print(f'校验 {fname} ...') + real_md5 = md5(tmp) + if real_md5 != EXPECTED_MD5[fname]: + print(f'MD5 不匹配! 期望: {EXPECTED_MD5[fname]} 实际: {real_md5}') + os.remove(tmp) + sys.exit(1) + print(f'校验通过 ({real_md5})') + + +def _move_to_model_dir(tmp: str, model_dir: str, fname: str) -> None: + """将文件移动到模型目录""" + dst = os.path.join(model_dir, fname) + if os.path.exists(dst): + os.remove(dst) + shutil.move(tmp, dst) + print(f'已移动到 {dst}') + + +def download_models( + choice: str, + mirror: dict[str, Any], + model_dir: str, +) -> None: + """下载所有模型文件""" + print_step('4. 下载模型文件') + for fname in MODEL_FILES: + print(f'\n--- 处理: {fname} ---') + tmp = os.path.join(os.getcwd(), fname) + _download_one(fname, choice, mirror, tmp) + _verify_md5(fname, tmp) + _move_to_model_dir(tmp, model_dir, fname) + + +def verify_loading(model_dir: str) -> None: + """验证 EasyOCR 模型加载""" + print_step('5. 验证 EasyOCR 加载') + try: + import easyocr + + easyocr.Reader( + ['ch_sim', 'en'], + model_storage_directory=model_dir, + download_enabled=False, + verbose=False, + ) + print('验证成功，模型可正常加载。') + except Exception as e: + print(f'验证失败: {e}') + + +def main() -> None: + print_step('EasyOCR 模型下载助手') + + easyocr_ok = check_easyocr() + model_dir = prepare_model_dir() + choice, mirror = select_mirror() + download_models(choice, mirror, model_dir) + + print_step('所有模型就绪') + + if easyocr_ok: + verify_loading(model_dir) + else: + print_step('5. 跳过验证 (EasyOCR 未安装)') + + print('\n脚本完成。') + + +if __name__ == '__main__': + main() From 63f08158238dd2acc427de9da75dafe0278186ad Mon Sep 17 00:00:00 2001 From: KUAI Date: Fri, 24 Jul 2026 13:15:31 +0800 Subject: [PATCH 2/5] Update easyocr_models.py --- tools/easyocr_models.py | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/tools/easyocr_models.py b/tools/easyocr_models.py index f75cf6bd..a5075c17 100644 --- a/tools/easyocr_models.py +++ b/tools/easyocr_models.py @@ -36,7 +36,7 @@ }, '1': { 'name': 'EdgeOne', - 'base_url': 'http://easyocr.v.ekuai.tech/', + 'base_url': 'https://easyocr.v.ekuai.tech/', 'type': 'direct', }, '2': { @@ -273,6 +273,19 @@ def download_models( print_step('4. 下载模型文件') for fname in MODEL_FILES: print(f'\n--- 处理: {fname} ---') + dst = os.path.join(model_dir, fname) + + # 检查目标目录是否已存在模型文件 + if os.path.exists(dst): + print(f'目标目录已存在: {dst}') + print(f'校验已有文件 {fname} ...') + real_md5 = md5(dst) + if real_md5 == EXPECTED_MD5[fname]: + print(f'已有文件校验通过 ({real_md5})，跳过下载。') + continue + else: + print(f'已有文件 MD5 不匹配 (期望: {EXPECTED_MD5[fname]} 实际: {real_md5})，将重新下载。') + tmp = os.path.join(os.getcwd(), fname) _download_one(fname, choice, mirror, tmp) _verify_md5(fname, tmp) From 0665bc4bb8f6410759c353e88dc89ec160f2c73d Mon Sep 17 00:00:00 2001 From: KUAI Date: Fri, 24 Jul 2026 13:16:56 +0800 Subject: [PATCH 3/5] Update easyocr_models.py --- tools/easyocr_models.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tools/easyocr_models.py b/tools/easyocr_models.py index a5075c17..3e1148af 100644 --- a/tools/easyocr_models.py +++ b/tools/easyocr_models.py @@ -284,7 +284,9 @@ def download_models( print(f'已有文件校验通过 ({real_md5})，跳过下载。') continue else: - print(f'已有文件 MD5 不匹配 (期望: {EXPECTED_MD5[fname]} 实际: {real_md5})，将重新下载。') + print( + f'已有文件 MD5 不匹配 (期望: {EXPECTED_MD5[fname]} 实际: {real_md5})，将重新下载。' + ) tmp = os.path.join(os.getcwd(), fname) _download_one(fname, choice, mirror, tmp) From 3f16402869e1caf7ea587a2e5fd3398010784f38 Mon Sep 17 00:00:00 2001 From: KUAI Date: Fri, 24 Jul 2026 13:27:50 +0800 Subject: [PATCH 4/5] use OCREngine to replace EasyOCREngine --- autowsgr/combat/actions.py | 4 +--- autowsgr/combat/handlers.py | 2 +- autowsgr/ui/map/panels/sortie.py | 6 +++--- 3 files changed, 5 insertions(+), 7 deletions(-) diff --git a/autowsgr/combat/actions.py b/autowsgr/combat/actions.py index d59fcbb2..d357084c 100644 --- a/autowsgr/combat/actions.py +++ b/autowsgr/combat/actions.py @@ -304,7 +304,7 @@ def click_image(device: AndroidController, template_key: TemplateKey, timeout: f return False -def get_ship_drop(device: AndroidController) -> str | None: +def get_ship_drop(device: AndroidController, ocr: OCREngine) -> str | None: """获取掉落舰船名称。 Parameters @@ -317,10 +317,8 @@ def get_ship_drop(device: AndroidController) -> str | None: str | None 掉落的舰船名称，或 ``None`` 如果未获取到。 """ - from autowsgr.vision import EasyOCREngine screen = device.screenshot() - ocr = EasyOCREngine.create(gpu=False) result = recognize_ship_drop(screen, ocr) if result.ship_name: _log.info('[Combat] 掉落识别: {} ({})', result.ship_name, result.ship_type or '未知') diff --git a/autowsgr/combat/handlers.py b/autowsgr/combat/handlers.py index e8682508..0e3516bc 100644 --- a/autowsgr/combat/handlers.py +++ b/autowsgr/combat/handlers.py @@ -416,7 +416,7 @@ def _handle_result(self) -> ConditionFlag: def _handle_get_ship(self) -> ConditionFlag: """处理获取舰船。""" - ship_name = get_ship_drop(self._device) + ship_name = get_ship_drop(self._device, self._ocr) if ship_name: _log.info('[Combat] 获得舰船: {}', ship_name) diff --git a/autowsgr/ui/map/panels/sortie.py b/autowsgr/ui/map/panels/sortie.py index 0d6089d7..260dadd6 100644 --- a/autowsgr/ui/map/panels/sortie.py +++ b/autowsgr/ui/map/panels/sortie.py @@ -30,7 +30,7 @@ if TYPE_CHECKING: import numpy as np - from autowsgr.vision import EasyOCREngine + from autowsgr.vision import OCREngine _log = get_logger('ui') @@ -103,7 +103,7 @@ def _parse_numerator(text: str, max_val: int) -> int: return None -def recognize_loot_count(screen: np.ndarray, ocr: EasyOCREngine) -> int | None: +def recognize_loot_count(screen: np.ndarray, ocr: OCREngine) -> int | None: """识别出征面板战利品 (胖次) 已获取数量。 OCR ``X/50`` 区域并提取 ``/`` 前的数字, 上限固定为 50。 @@ -123,7 +123,7 @@ def recognize_loot_count(screen: np.ndarray, ocr: EasyOCREngine) -> int | None: return count -def recognize_ship_count(screen: np.ndarray, ocr: EasyOCREngine) -> int | None: +def recognize_ship_count(screen: np.ndarray, ocr: OCREngine) -> int | None: """识别出征面板舰船已获取数量。 OCR ``X/500`` 区域并提取 ``/`` 前的数字, 上限固定为 500。 From 4c2cc596e42426e71395b54daa3ed6a33c11374d Mon Sep 17 00:00:00 2001 From: KUAI Date: Fri, 24 Jul 2026 15:37:55 +0800 Subject: [PATCH 5/5] embed model download --- autowsgr/infra/config.py | 3 + autowsgr/scheduler/launcher.py | 2 +- autowsgr/types.py | 13 + autowsgr/vision/easyocr_models_checker.py | 75 ++++ autowsgr/vision/easyocr_models_cli.py | 403 ++++++++++++++++++++++ autowsgr/vision/ocr.py | 17 +- examples/usersettings.full.yaml | 5 + tools/easyocr_models.py | 333 ------------------ usersettings.yaml | 5 + 9 files changed, 517 insertions(+), 339 deletions(-) create mode 100644 autowsgr/vision/easyocr_models_checker.py create mode 100644 autowsgr/vision/easyocr_models_cli.py delete mode 100644 tools/easyocr_models.py diff --git a/autowsgr/infra/config.py b/autowsgr/infra/config.py index 49e86ac1..1b6decae 100644 --- a/autowsgr/infra/config.py +++ b/autowsgr/infra/config.py @@ -19,6 +19,7 @@ GameAPP, MapEntrance, OcrBackend, + OcrMirror, OSType, RepairMode, ShipType, @@ -80,6 +81,8 @@ class OCRConfig(BaseModel): """OCR 后端""" gpu: bool = False """是否使用 GPU 加速""" + mirror: OcrMirror = OcrMirror.tencent + """EasyOCR 模型下载镜像源""" class LogConfig(BaseModel): diff --git a/autowsgr/scheduler/launcher.py b/autowsgr/scheduler/launcher.py index 6b1bbd13..ea5efd6e 100644 --- a/autowsgr/scheduler/launcher.py +++ b/autowsgr/scheduler/launcher.py @@ -137,7 +137,7 @@ def create_ocr(self) -> OCREngine: cfg = self.config _log.info('[Launcher] 创建 OCR 引擎 (backend={})', cfg.ocr.backend.value) # 目前仅支持 EasyOCR，后续可按 cfg.ocr.backend 分发 - self._ocr = EasyOCREngine.create(gpu=cfg.ocr.gpu) + self._ocr = EasyOCREngine.create(gpu=cfg.ocr.gpu, mirror=cfg.ocr.mirror) return self._ocr # ── 构造 GameContext ── diff --git a/autowsgr/types.py b/autowsgr/types.py index 878e5efc..79a4a84d 100644 --- a/autowsgr/types.py +++ b/autowsgr/types.py @@ -184,6 +184,19 @@ class OcrBackend(StrEnum): paddleocr = 'paddleocr' +class OcrMirror(StrEnum): + """EasyOCR 模型下载镜像源。""" + + origin = 'origin' + """EasyOCR 原生下载""" + github = 'github' + """GitHub releases""" + tencent = 'tencent' + """EdgeOne CDN (腾讯云)""" + modelscope = 'modelscope' + """ModelScope 镜像""" + + # ── 游戏概念 ── diff --git a/autowsgr/vision/easyocr_models_checker.py b/autowsgr/vision/easyocr_models_checker.py new file mode 100644 index 00000000..aa7a3e14 --- /dev/null +++ b/autowsgr/vision/easyocr_models_checker.py @@ -0,0 +1,75 @@ +"""EasyOCR 模型管理 — 检查、校验、下载。 + +在运行时初始化 EasyOCR 前调用 :func:`ensure_models`，自动完成： +1. 检查模型目录中是否存在所需模型文件 +2. 校验 MD5 +3. 若缺失或损坏，从用户配置的镜像源下载 +""" + +from __future__ import annotations + +import os +import shutil + +from autowsgr.infra.logger import get_logger + +# 从 CLI 模块导入共享常量和下载函数（单一维护点） +from .easyocr_models_cli import ( + EXPECTED_MD5, + MODEL_FILES, + check_model_file, + download_model_file, + ensure_model_dir, + ensure_modelscope, +) + + +_log = get_logger('vision.model_download') + + +def ensure_models(mirror: str = 'tencent') -> None: + """确保所有 EasyOCR 模型就绪。 + + Parameters + ---------- + mirror: + 镜像源名称: ``'origin'`` / ``'github'`` / ``'tencent'`` / ``'modelscope'``。 + 为 ``'origin'`` 时不执行任何检查或下载，由 EasyOCR 自行处理。 + """ + model_dir = ensure_model_dir() + + if mirror == 'origin': + return + + if mirror == 'modelscope': + ensure_modelscope() + + for fname in MODEL_FILES: + path = os.path.join(model_dir, fname) + if check_model_file(path, fname): + _log.info('[Model] {} 已就绪', fname) + continue + + if os.path.isfile(path): + _log.warning('[Model] {} MD5 不匹配，将重新下载', fname) + + _log.info('[Model] 从镜像源 {} 下载EasyOCR模型 {}', mirror, path) + _download_and_verify(fname, mirror, model_dir, path) + + +def _download_and_verify(fname: str, mirror: str, model_dir: str, dest: str) -> None: + """下载单个模型文件到临时位置，校验后移入模型目录。""" + tmp = os.path.join(model_dir, fname + '.tmp') + # 清理可能残留的临时文件 + if os.path.exists(tmp): + os.remove(tmp) + try: + download_model_file(fname, mirror, tmp, expected_md5=EXPECTED_MD5[fname]) + if os.path.exists(dest): + os.remove(dest) + shutil.move(tmp, dest) + _log.info('[Model] {} 下载完成并校验通过', fname) + except Exception: + if os.path.exists(tmp): + os.remove(tmp) + raise diff --git a/autowsgr/vision/easyocr_models_cli.py b/autowsgr/vision/easyocr_models_cli.py new file mode 100644 index 00000000..4fb4f506 --- /dev/null +++ b/autowsgr/vision/easyocr_models_cli.py @@ -0,0 +1,403 @@ +""" +EasyOCR 模型下载助手（支持 GitHub/EdgeOne/ModelScope 镜像） +GitHub 源：从官方地址自动下载 zip 并解压提取 .pth 文件。 +EdgeOne 源：从 EdgeOne 镜像下载模型文件，由 kuai 提供。 +ModelScope 源：从 ModelScope 镜像下载模型文件，由 Ceceliachenen 提供。 +""" + +import hashlib +import importlib.util +import logging +import os +import shutil +import subprocess +import sys +import urllib.parse +import urllib.request +import zipfile +from typing import Any, ClassVar + + +class _ColorFormatter(logging.Formatter): + """只输出消息，按级别着色""" + + _COLORS: ClassVar[dict[int, str]] = { + logging.DEBUG: '\033[36m', # 青色 + logging.INFO: '\033[32m', # 绿色 + logging.WARNING: '\033[33m', # 黄色 + logging.ERROR: '\033[31m', # 红色 + } + _RESET: ClassVar[str] = '\033[0m' + + def format(self, record: logging.LogRecord) -> str: + color = self._COLORS.get(record.levelno, '') + return f'{color}{record.getMessage()}{self._RESET}' + + +logger = logging.getLogger(__name__) +logger.setLevel(logging.INFO) +_handler = logging.StreamHandler() +_handler.setFormatter(_ColorFormatter()) +logger.addHandler(_handler) + +# -------------------- 常量 -------------------- +MODEL_FILES = ['craft_mlt_25k.pth', 'zh_sim_g2.pth'] + +EXPECTED_MD5 = { + 'craft_mlt_25k.pth': '2f8227d2def4037cdb3b34389dcf9ec1', + 'zh_sim_g2.pth': 'b601ce7143293387d3ec4f41a66edc07', +} + +# 按镜像名称索引的下载配置，与 YAML 中 ocr.mirror 枚举值对齐 +MIRROR_CONFIG: dict[str, dict[str, Any]] = { + 'github': { + 'type': 'github', + 'urls': { + 'craft_mlt_25k.pth': 'https://github.com/JaidedAI/EasyOCR/releases/download/pre-v1.1.6/craft_mlt_25k.zip', + 'zh_sim_g2.pth': 'https://github.com/JaidedAI/EasyOCR/releases/download/v1.3/zh_sim_g2.zip', + }, + }, + 'tencent': { + 'type': 'edgeone', + 'base_url': 'https://easyocr.v.ekuai.tech/', + 'split': { + 'craft_mlt_25k.pth': 4, + }, + }, + 'modelscope': { + 'type': 'modelscope', + }, +} + +# CLI 交互用的镜像选项（引用 MIRROR_CONFIG 避免重复声明 URL） +MIRROR_OPTIONS: dict[str, dict[str, Any]] = { + '0': { + 'name': 'GitHub', + 'key': 'github', + }, + '1': { + 'name': 'EdgeOne (腾讯云)', + 'key': 'tencent', + }, + '2': { + 'name': 'ModelScope', + 'key': 'modelscope', + }, +} + +MODELSCOPE_MODEL = 'Ceceliachenen/easyocr' + + +def ensure_model_dir() -> str: + """返回 EasyOCR 模型目录路径，不存在则创建。""" + model_dir = os.path.join(os.path.expanduser('~'), '.EasyOCR', 'model') + os.makedirs(model_dir, exist_ok=True) + return model_dir + + +def download_model_file( + fname: str, + mirror: str, + dest: str, + *, + expected_md5: str | None = None, +) -> None: + """根据镜像类型下载单个模型文件到指定路径。 + + Parameters + ---------- + fname: + 模型文件名，如 ``'craft_mlt_25k.pth'``。 + mirror: + 镜像名称: ``'github'`` / ``'tencent'`` / ``'modelscope'``。 + dest: + 目标文件路径。 + expected_md5: + 若提供，下载后校验 MD5；不匹配时删除文件并抛出 ``RuntimeError``。 + """ + cfg = MIRROR_CONFIG[mirror] + mtype = cfg['type'] + + if mtype == 'github': + download_github_zip(cfg['urls'][fname], fname, dest) + elif mtype == 'edgeone': + split_count = cfg.get('split', {}).get(fname) + if split_count is not None: + download_edgeone_split(cfg['base_url'], fname, dest, parts=split_count) + else: + download_file(cfg['base_url'] + fname, dest) + elif mtype == 'modelscope': + dest_dir = os.path.dirname(dest) or '.' + download_modelscope(fname, dest_dir) + # modelscope API 直接写入 local_dir，若目标路径不同则移动 + src = os.path.join(dest_dir, fname) + if src != dest and os.path.exists(src): + if os.path.exists(dest): + os.remove(dest) + shutil.move(src, dest) + else: + raise ValueError(f'未知下载方式: {mtype}') + + if expected_md5 is not None: + real = md5(dest) + if real != expected_md5: + os.remove(dest) + raise RuntimeError(f'{fname} MD5 校验失败: 期望 {expected_md5}, 实际 {real}') + logger.info('校验 %s MD5: %s 正确', dest, real) + + +# -------------------- 工具函数 -------------------- +def get_input(prompt: str, default: str = '0', choices: list[str] | None = None) -> str: + while True: + user_input = input(prompt).strip() + if not user_input: + user_input = default + if choices is None or user_input in choices: + return user_input + logger.warning('无效输入，请从 %s 中选择。', choices) + + +def print_step(msg: str) -> None: + logger.info('-' * 60) + logger.info('>>> %s', msg) + logger.info('-' * 60) + + +def download_file(url: str, dest: str) -> None: + """通用文件下载（带进度）""" + logger.info('下载: %s -> %s', url, dest) + parsed = urllib.parse.urlparse(url) + if parsed.scheme not in ('http', 'https'): + raise ValueError(f'不支持的 URL 协议: {parsed.scheme}') + try: + with urllib.request.urlopen(url) as response: # noqa: S310 + total = int(response.headers.get('Content-Length', 0)) + downloaded = 0 + block_size = 4096 + + with open(dest, 'wb') as f: + while True: + chunk = response.read(block_size) + if not chunk: + break + f.write(chunk) + downloaded += len(chunk) + + if total > 0: + pct = min(100, downloaded * 100 / total) + sys.stdout.write(f'\r下载进度: {pct:.1f}%') + sys.stdout.flush() + + if total > 0: + sys.stdout.write('\n') + logger.info('下载完成') + except Exception as e: + logger.error('下载失败: %s', e) + raise + + +def md5(path: str) -> str: + h = hashlib.md5() + with open(path, 'rb') as f: + for chunk in iter(lambda: f.read(4096), b''): + h.update(chunk) + return h.hexdigest() + + +def check_model_file(path: str, fname: str) -> bool: + """检查单个模型文件是否存在且 MD5 正确。""" + if not os.path.isfile(path): + return False + return md5(path) == EXPECTED_MD5[fname] + + +def download_modelscope(filename: str, local_dir: str = '.') -> None: + """使用 ModelScope Python API 下载单个文件""" + try: + from modelscope.hub.file_download import model_file_download + except ImportError as err: + raise RuntimeError('modelscope 安装不完整，缺少 hub.file_download 模块') from err + logger.info('ModelScope API 下载: %s -> %s', filename, local_dir) + model_file_download( + model_id=MODELSCOPE_MODEL, + file_path=filename, + local_dir=local_dir, + ) + logger.info('下载完成') + + +def download_edgeone_split(base_url: str, filename: str, dest: str, *, parts: int = 4) -> None: + """EdgeOne 分片下载并合并""" + tmp_parts: list[str] = [] + logger.info('EdgeOne 分片下载: %s (共 %d 部分)', filename, parts) + try: + for i in range(1, parts + 1): + suffix = f'.part_{i:03d}' + part_url = base_url + filename + suffix + part_local = dest + suffix + logger.info('[部分 %d/%d]', i, parts) + download_file(part_url, part_local) + tmp_parts.append(part_local) + + logger.info('合并分片...') + with open(dest, 'wb') as out: + for p in tmp_parts: + with open(p, 'rb') as f: + data = f.read() + out.write(data) + logger.info( + '已合并: %s (%.2f MB)', os.path.basename(p), len(data) / 1024 / 1024 + ) + finally: + logger.info('清理临时分片...') + for p in tmp_parts: + if os.path.exists(p): + os.remove(p) + + +def download_github_zip(zip_url: str, pth_name: str, dest_path: str) -> None: + """下载 zip 文件并解压提取指定的 .pth 模型文件""" + tmp_zip = dest_path + '.zip' + try: + download_file(zip_url, tmp_zip) + + logger.info('解压 %s ...', os.path.basename(tmp_zip)) + with zipfile.ZipFile(tmp_zip, 'r') as zf: + pth_files = [f for f in zf.namelist() if f.endswith('.pth')] + if not pth_files: + raise RuntimeError(f'压缩包中未找到任何 .pth 文件: {tmp_zip}') + + target = next((f for f in pth_files if os.path.basename(f) == pth_name), pth_files[0]) + logger.info('提取文件: %s', target) + + with zf.open(target) as src, open(dest_path, 'wb') as dst: + shutil.copyfileobj(src, dst) + finally: + if os.path.exists(tmp_zip): + os.remove(tmp_zip) + logger.info('解压完成，模型文件保存为: %s', dest_path) + + +# -------------------- 主流程 -------------------- +def check_easyocr() -> bool: + """检测 EasyOCR 环境，返回是否已安装""" + print_step('1. 检测 EasyOCR 环境') + if importlib.util.find_spec('easyocr') is not None: + logger.info('EasyOCR 已安装。') + return True + + logger.warning('未检测到 EasyOCR') + opt = get_input('[0] 跳过继续下载 (默认) [1] 退出: ', '0', ['0', '1']) + if opt == '1': + sys.exit(0) + logger.info('跳过安装，将继续下载模型。') + return False + + +def prepare_model_dir() -> str: + """准备模型目录并返回路径""" + print_step('2. 准备模型目录') + model_dir = ensure_model_dir() + logger.info('目标: %s', model_dir) + return model_dir + + +def select_mirror() -> str: + """选择下载镜像源，返回镜像 key（如 'github' / 'tencent' / 'modelscope'）""" + print_step('3. 选择下载镜像源') + for k, v in MIRROR_OPTIONS.items(): + logger.info(' [%s] %s', k, v['name']) + choice = get_input('请输入数字 (默认0): ', '0', ['0', '1', '2']) + mirror_key = MIRROR_OPTIONS[choice]['key'] + + if mirror_key == 'modelscope': + _ensure_modelscope_interactive() + + return mirror_key + + +def ensure_modelscope() -> None: + """确保 modelscope 已安装，未安装时静默自动安装。""" + if importlib.util.find_spec('modelscope') is not None: + return + logger.info('安装 modelscope ...') + subprocess.check_call([sys.executable, '-m', 'pip', 'install', 'modelscope']) + logger.info('modelscope 安装成功') + + +def _ensure_modelscope_interactive() -> None: + """CLI 交互模式：确认后安装 modelscope。""" + if importlib.util.find_spec('modelscope') is not None: + logger.info('modelscope 已安装。') + return + + logger.warning('未安装 modelscope') + ins = get_input('自动安装? [0] 是 (默认) [1] 否: ', '0', ['0', '1']) + if ins == '0': + ensure_modelscope() + else: + logger.warning('请手动安装 modelscope 后重试。') + sys.exit(0) + + +def download_models( + mirror: str, + model_dir: str, +) -> None: + """下载所有模型文件""" + print_step('4. 下载模型文件') + for fname in MODEL_FILES: + logger.info('--- 处理: %s ---', fname) + dst = os.path.join(model_dir, fname) + + if check_model_file(dst, fname): + logger.info('已有文件校验通过，跳过下载: %s', dst) + continue + + if os.path.exists(dst): + logger.warning('已有文件 MD5 不匹配，将重新下载: %s', dst) + + tmp = os.path.join(model_dir, fname + '.tmp') + try: + download_model_file(fname, mirror, tmp, expected_md5=EXPECTED_MD5[fname]) + except Exception as e: + logger.error('下载或解压失败: %s', e) + sys.exit(1) + if os.path.exists(dst): + os.remove(dst) + shutil.move(tmp, dst) + logger.info('已移动到 %s', dst) + + +def verify_loading() -> None: + """验证 EasyOCR 模型加载""" + print_step('5. 验证 EasyOCR 加载') + try: + import easyocr + + easyocr.Reader(['ch_sim', 'en']) + logger.info('验证成功，模型可正常加载。') + except Exception as e: + logger.error('验证失败: %s', e) + + +def main() -> None: + print_step('EasyOCR 模型下载助手') + + easyocr_ok = check_easyocr() + model_dir = prepare_model_dir() + mirror = select_mirror() + download_models(mirror, model_dir) + + print_step('所有模型就绪') + + if easyocr_ok: + verify_loading() + else: + print_step('5. 跳过验证 (EasyOCR 未安装)') + + logger.info('脚本完成。') + + +if __name__ == '__main__': + main() diff --git a/autowsgr/vision/ocr.py b/autowsgr/vision/ocr.py index 535b0497..beb69c7c 100644 --- a/autowsgr/vision/ocr.py +++ b/autowsgr/vision/ocr.py @@ -6,7 +6,7 @@ from autowsgr.vision import OCREngine - engine = OCREngine.create("easyocr", gpu=False) + engine = OCREngine.create("easyocr", gpu=False, mirror="tencent") results = engine.recognize(cropped_image) number = engine.recognize_number(resource_area) """ @@ -379,7 +379,9 @@ def recognize_ship_names( """已创建的引擎单例缓存，key 为 ``":"``。""" @classmethod - def create(cls, engine: str = 'easyocr', gpu: bool = False) -> OCREngine: + def create( + cls, engine: str = 'easyocr', gpu: bool = False, mirror: str = 'tencent' + ) -> OCREngine: """创建或获取 OCR 引擎实例（单例）。 首次调用时创建引擎实例并缓存，后续相同参数的调用直接返回缓存实例。 @@ -390,6 +392,8 @@ def create(cls, engine: str = 'easyocr', gpu: bool = False) -> OCREngine: 引擎名称: ``"easyocr"`` 或 ``"paddleocr"``。 gpu: 是否使用 GPU 加速。 + mirror: + 模型下载镜像源: ``"origin"`` / ``"github"`` / ``"tencent"`` / ``"modelscope"``。 Returns ------- @@ -401,8 +405,8 @@ def create(cls, engine: str = 'easyocr', gpu: bool = False) -> OCREngine: return cls._instances[cache_key] if engine == 'easyocr': - _log.info('[OCR] 初始化 EasyOCR（gpu={}）', gpu) - instance = EasyOCREngine(gpu=gpu) + _log.info('[OCR] 初始化 EasyOCR（gpu={}, mirror={}）', gpu, mirror) + instance = EasyOCREngine(gpu=gpu, mirror=mirror) cls._instances[cache_key] = instance return instance raise ValueError(f'不支持的 OCR 引擎: {engine}，可选: easyocr, paddleocr') @@ -414,7 +418,10 @@ def create(cls, engine: str = 'easyocr', gpu: bool = False) -> OCREngine: class EasyOCREngine(OCREngine): """基于 EasyOCR 的识别引擎。""" - def __init__(self, gpu: bool = False) -> None: + def __init__(self, gpu: bool = False, mirror: str = 'tencent') -> None: + from autowsgr.vision.easyocr_models_checker import ensure_models + + ensure_models(mirror) self._reader = easyocr.Reader(['ch_sim', 'en'], gpu=gpu) def recognize( diff --git a/examples/usersettings.full.yaml b/examples/usersettings.full.yaml index 16ed0071..dfc6c32d 100644 --- a/examples/usersettings.full.yaml +++ b/examples/usersettings.full.yaml @@ -31,6 +31,11 @@ account: ocr: backend: easyocr # easyocr / paddleocr gpu: false # 是否使用 GPU 加速 + mirror: tencent # Easyocr 模型镜像, 用于下载模型, 可选值如下 + # origin: 使用Easyocr原生下载 + # github: 使用Github地址下载, 由官方提供 + # tencent: 使用kuai提供的镜像下载, 由腾讯云托管 + # modelscope: 使用Ceceliachenen提供的镜像下载, 由ModelScope托管 # ═══════════════════════════════════════════ # 日志 diff --git a/tools/easyocr_models.py b/tools/easyocr_models.py deleted file mode 100644 index 3e1148af..00000000 --- a/tools/easyocr_models.py +++ /dev/null @@ -1,333 +0,0 @@ -""" -EasyOCR 模型下载助手（支持 GitHub/EdgeOne/ModelScope 镜像） -GitHub 源：从官方地址自动下载 zip 并解压提取 .pth 文件。 -EdgeOne 源：从 EdgeOne 镜像下载模型文件，由 kuai 提供。 -ModelScope 源：从 ModelScope 镜像下载模型文件，由 Ceceliachenen 提供。 -""" - -import hashlib -import importlib.util -import os -import shutil -import subprocess -import sys -import urllib.parse -import urllib.request -import zipfile -from typing import Any - - -# -------------------- 常量 -------------------- -MODEL_FILES = ['craft_mlt_25k.pth', 'zh_sim_g2.pth'] - -EXPECTED_MD5 = { - 'craft_mlt_25k.pth': '2f8227d2def4037cdb3b34389dcf9ec1', - 'zh_sim_g2.pth': 'b601ce7143293387d3ec4f41a66edc07', -} - -MIRROR_OPTIONS: dict[str, dict[str, Any]] = { - '0': { - 'name': 'GitHub', - 'type': 'github', - 'urls': { - 'craft_mlt_25k.pth': 'https://github.com/JaidedAI/EasyOCR/releases/download/pre-v1.1.6/craft_mlt_25k.zip', - 'zh_sim_g2.pth': 'https://github.com/JaidedAI/EasyOCR/releases/download/v1.3/zh_sim_g2.zip', - }, - }, - '1': { - 'name': 'EdgeOne', - 'base_url': 'https://easyocr.v.ekuai.tech/', - 'type': 'direct', - }, - '2': { - 'name': 'ModelScope', - 'type': 'modelscope', - }, -} - -MODELSCOPE_MODEL = 'Ceceliachenen/easyocr' - - -# -------------------- 工具函数 -------------------- -def get_input(prompt: str, default: str = '0', choices: list[str] | None = None) -> str: - while True: - user_input = input(prompt).strip() - if not user_input: - user_input = default - if choices is None or user_input in choices: - return user_input - print(f'无效输入，请从 {choices} 中选择。') - - -def print_step(msg: str) -> None: - print(f'\n{"-" * 60}\n>>> {msg}\n{"-" * 60}') - - -def ensure_dir(path: str) -> None: - os.makedirs(path, exist_ok=True) - print(f'目录已就绪: {path}' if os.path.isdir(path) else f'已创建目录: {path}') - - -def _download_hook(count: int, blk: int, total: int) -> None: - if total > 0: - pct = min(100, count * blk * 100 / total) - sys.stdout.write(f'\r下载进度: {pct:.1f}%') - sys.stdout.flush() - if pct >= 100: - sys.stdout.write('\n') - - -def download_file(url: str, dest: str) -> None: - """通用文件下载（带进度）""" - print(f'下载: {url} -> {dest}') - parsed = urllib.parse.urlparse(url) - if parsed.scheme not in ('http', 'https'): - raise ValueError(f'不支持的 URL 协议: {parsed.scheme}') - try: - urllib.request.urlretrieve(url, dest, _download_hook) # noqa: S310 - print('下载完成') - except Exception as e: - print(f'下载失败: {e}') - raise - - -def md5(path: str) -> str: - h = hashlib.md5() - with open(path, 'rb') as f: - for chunk in iter(lambda: f.read(4096), b''): - h.update(chunk) - return h.hexdigest() - - -def download_modelscope(filename: str, local_dir: str = '.') -> None: - """使用 ModelScope Python API 下载单个文件""" - try: - from modelscope.hub.file_download import model_file_download - except ImportError as err: - raise RuntimeError('modelscope 安装不完整，缺少 hub.file_download 模块') from err - print(f'ModelScope API 下载: {filename} -> {local_dir}') - model_file_download( - model_id=MODELSCOPE_MODEL, - file_path=filename, - local_dir=local_dir, - ) - print('下载完成') - - -def download_edgeone_split(base_url: str, filename: str, dest: str) -> None: - """EdgeOne 分片下载并合并""" - parts = 4 - tmp_parts: list[str] = [] - print(f'EdgeOne 分片下载: {filename} (共 {parts} 部分)') - for i in range(1, parts + 1): - suffix = f'.part_{i:03d}' - part_url = base_url + filename + suffix - part_local = os.path.join(os.getcwd(), filename + suffix) - print(f'\n[部分 {i}/{parts}]') - download_file(part_url, part_local) - tmp_parts.append(part_local) - - print('\n合并分片...') - with open(dest, 'wb') as out: - for p in tmp_parts: - with open(p, 'rb') as f: - data = f.read() - out.write(data) - print(f'已合并: {os.path.basename(p)} ({len(data) / 1024 / 1024:.2f} MB)') - print('合并完成，删除临时分片...') - for p in tmp_parts: - os.remove(p) - - -def download_github_zip(zip_url: str, pth_name: str, dest_path: str) -> None: - """下载 zip 文件并解压提取指定的 .pth 模型文件""" - tmp_zip = os.path.join(os.getcwd(), pth_name + '.zip') - download_file(zip_url, tmp_zip) - - print(f'解压 {os.path.basename(tmp_zip)} ...') - with zipfile.ZipFile(tmp_zip, 'r') as zf: - pth_files = [f for f in zf.namelist() if f.endswith('.pth')] - if not pth_files: - os.remove(tmp_zip) - raise RuntimeError(f'压缩包中未找到任何 .pth 文件: {tmp_zip}') - - target = _find_pth_target(pth_files, pth_name) - print(f'提取文件: {target}') - - with zf.open(target) as src, open(dest_path, 'wb') as dst: - shutil.copyfileobj(src, dst) - - os.remove(tmp_zip) - print(f'解压完成，模型文件保存为: {dest_path}') - - -def _find_pth_target(pth_files: list[str], pth_name: str) -> str: - """在 zip 文件列表中查找匹配的 .pth 文件""" - for f in pth_files: - if os.path.basename(f) == pth_name: - return f - return pth_files[0] - - -# -------------------- 主流程 -------------------- -def check_easyocr() -> bool: - """检测 EasyOCR 环境，返回是否已安装""" - print_step('1. 检测 EasyOCR 环境') - if importlib.util.find_spec('easyocr') is not None: - print('EasyOCR 已安装。') - return True - - print('未检测到 EasyOCR') - opt = get_input('[0] 跳过继续下载 (默认) [1] 退出: ', '0', ['0', '1']) - if opt == '1': - sys.exit(0) - print('跳过安装，将继续下载模型。') - return False - - -def prepare_model_dir() -> str: - """准备模型目录并返回路径""" - print_step('2. 准备模型目录') - model_dir = os.path.join(os.path.expanduser('~'), '.EasyOCR', 'model') - print(f'目标: {model_dir}') - ensure_dir(model_dir) - return model_dir - - -def select_mirror() -> tuple[str, dict[str, Any]]: - """选择下载镜像源，返回 (choice, mirror)""" - print_step('3. 选择下载镜像源') - for k, v in MIRROR_OPTIONS.items(): - print(f' [{k}] {v["name"]}') - choice = get_input('请输入数字 (默认0): ', '0', ['0', '1', '2']) - mirror = MIRROR_OPTIONS[choice] - - if choice == '2': - _ensure_modelscope() - - return choice, mirror - - -def _ensure_modelscope() -> None: - """确保 modelscope 已安装""" - if importlib.util.find_spec('modelscope') is not None: - print('modelscope 已安装。') - return - - print('未安装 modelscope') - ins = get_input('自动安装? [0] 是 (默认) [1] 否: ', '0', ['0', '1']) - if ins == '0': - print('安装 modelscope ...') - subprocess.check_call([sys.executable, '-m', 'pip', 'install', 'modelscope']) - print('安装成功') - else: - print('请手动安装 modelscope 后重试。') - sys.exit(0) - - -def _download_one(fname: str, choice: str, mirror: dict[str, Any], tmp: str) -> None: - """下载单个模型文件""" - try: - if mirror['type'] == 'github': - download_github_zip(mirror['urls'][fname], fname, tmp) - elif choice == '1' and fname == 'craft_mlt_25k.pth': - download_edgeone_split(mirror['base_url'], fname, tmp) - elif mirror['type'] == 'direct': - download_file(mirror['base_url'] + fname, tmp) - elif mirror['type'] == 'modelscope': - download_modelscope(fname, os.getcwd()) - else: - print(f'未知下载方式: {mirror["type"]}') - sys.exit(1) - except Exception as e: - print(f'下载或解压失败: {e}') - sys.exit(1) - - -def _verify_md5(fname: str, tmp: str) -> None: - """校验文件 MD5""" - print(f'校验 {fname} ...') - real_md5 = md5(tmp) - if real_md5 != EXPECTED_MD5[fname]: - print(f'MD5 不匹配! 期望: {EXPECTED_MD5[fname]} 实际: {real_md5}') - os.remove(tmp) - sys.exit(1) - print(f'校验通过 ({real_md5})') - - -def _move_to_model_dir(tmp: str, model_dir: str, fname: str) -> None: - """将文件移动到模型目录""" - dst = os.path.join(model_dir, fname) - if os.path.exists(dst): - os.remove(dst) - shutil.move(tmp, dst) - print(f'已移动到 {dst}') - - -def download_models( - choice: str, - mirror: dict[str, Any], - model_dir: str, -) -> None: - """下载所有模型文件""" - print_step('4. 下载模型文件') - for fname in MODEL_FILES: - print(f'\n--- 处理: {fname} ---') - dst = os.path.join(model_dir, fname) - - # 检查目标目录是否已存在模型文件 - if os.path.exists(dst): - print(f'目标目录已存在: {dst}') - print(f'校验已有文件 {fname} ...') - real_md5 = md5(dst) - if real_md5 == EXPECTED_MD5[fname]: - print(f'已有文件校验通过 ({real_md5})，跳过下载。') - continue - else: - print( - f'已有文件 MD5 不匹配 (期望: {EXPECTED_MD5[fname]} 实际: {real_md5})，将重新下载。' - ) - - tmp = os.path.join(os.getcwd(), fname) - _download_one(fname, choice, mirror, tmp) - _verify_md5(fname, tmp) - _move_to_model_dir(tmp, model_dir, fname) - - -def verify_loading(model_dir: str) -> None: - """验证 EasyOCR 模型加载""" - print_step('5. 验证 EasyOCR 加载') - try: - import easyocr - - easyocr.Reader( - ['ch_sim', 'en'], - model_storage_directory=model_dir, - download_enabled=False, - verbose=False, - ) - print('验证成功，模型可正常加载。') - except Exception as e: - print(f'验证失败: {e}') - - -def main() -> None: - print_step('EasyOCR 模型下载助手') - - easyocr_ok = check_easyocr() - model_dir = prepare_model_dir() - choice, mirror = select_mirror() - download_models(choice, mirror, model_dir) - - print_step('所有模型就绪') - - if easyocr_ok: - verify_loading(model_dir) - else: - print_step('5. 跳过验证 (EasyOCR 未安装)') - - print('\n脚本完成。') - - -if __name__ == '__main__': - main() diff --git a/usersettings.yaml b/usersettings.yaml index 5abcc8aa..3e818e2b 100644 --- a/usersettings.yaml +++ b/usersettings.yaml @@ -31,6 +31,11 @@ account: ocr: backend: easyocr # easyocr / paddleocr gpu: false # 是否使用 GPU 加速 + mirror: tencent # Easyocr 模型镜像, 用于下载模型, 可选值如下 + # origin: 使用Easyocr原生下载 + # github: 使用Github地址下载, 由官方提供 + # tencent: 使用kuai提供的镜像下载, 由腾讯云托管 + # modelscope: 使用Ceceliachenen提供的镜像下载, 由ModelScope托管 # ═══════════════════════════════════════════ # 日志