refactoring

.python-version

@@ -1 +1 @@
-3.11
+cpython-3.13.5-linux-x86_64-gnu

ml/train.py

@@ -1,5 +1,5 @@
 import os
-import matplotlib.pyplot as plt
+import matplotlib.pyplot as plt  # type: ignore
 import torch
 import torch.nn as nn
 from torchvision.datasets import ImageFolder  # type: ignore
@@ -19,7 +19,7 @@ def get_labels(input_dir, img_size):
             transforms.Resize(img_size),
             transforms.RandomHorizontalFlip(),
             transforms.ToTensor(),
-            transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
+            transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225]),
         ]
     )
     dataset = ImageFolder(root=input_dir, transform=transform)
@@ -47,7 +47,7 @@ def load_model(model_path: str, labels_dict: dict, device: str = "cuda") -> nn.M
         model = torchvision.models.resnet50(weights=ResNet50_Weights.DEFAULT)
         model.fc = nn.Sequential(
             nn.Dropout(0.5),  # Регуляризация
-            torch.nn.Linear(model.fc.in_features, len(labels_dict))
+            torch.nn.Linear(model.fc.in_features, len(labels_dict)),
         )
         return model
     model = torch.load(model_path, map_location=device, weights_only=False)

pyproject.toml

@@ -3,21 +3,31 @@ name = "ai"
 version = "0.1.0"
 description = "Add your description here"
 readme = "README.md"
-requires-python = "~=3.11"
+requires-python = "~=3.13"
 dependencies = [
-    "granian>=2.2.4",
+    "granian==2.5",
     "jinja2>=3.1.6",
-    "starlite>=1.51.16",
-    "numpy==1.23.5",
+    "numpy==2.3.4",
     "pillow>=11.1.0",
     "markdown>=3.9",
+    "aiocache",
+    "torch>=2.9.1",
+    "ruff>=0.14.5",
+    "mypy>=1.18.2",
+    "uvicorn>=0.38.0",
+    "pydantic>=2.12.4",
+    "litestar==2.18.0",
+    "ujson>=5.11.0",
+    "torchvision>=0.24.1",
+    "types-requests>=2.32.4.20250913",
+    "types-markdown>=3.10.0.20251106",
 ]
 
 [project.optional-dependencies]
 default = [
-    "torch>=2.6.0",
+    "torch>=2.9.1",
     "torchvision>=0.21.0",
-    "mypy>=1.15.0",
+    "mypy>=1.18",
     "pyqt5>=5.15.11",
     "requests>=2.32.3",
     "ruff>=0.11.5",
@@ -25,35 +35,15 @@ default = [
     "matplotlib>=3.10.1",
 ]
 
-api = [
-    "torch>=2.6.0",
-    "torchvision>=0.21.0",
-]
-
-[tool.uv]
-conflicts = [
-    [
-      { extra = "default" },
-      { extra = "api" },
-    ],
-]
-
 [tool.uv.sources]
 torch = [
-    { index = "pytorch-cu124", extra = "default" },
-    { index = "pytorch-cpu", extra = "api" },
+    { index = "pytorch-cpu", extra = "default" },
 ]
 
 torchvision = [
-    { index = "pytorch-cu124", extra = "default" },
-    { index = "pytorch-cpu", extra = "api" },
+    { index = "pytorch-cpu", extra = "default" },
 ]
 
-[[tool.uv.index]]
-name = "pytorch-cu124"
-url = "https://download.pytorch.org/whl/cu124"
-explicit = true
-
 [[tool.uv.index]]
 name = "pytorch-cpu"
 url = "https://download.pytorch.org/whl/cpu"

server/main.py

@@ -1,93 +1,35 @@
 from pathlib import Path
+import os
 
 import markdown
-from PIL import Image
-from starlite import (
+from litestar import (
     Controller,
-    StaticFilesConfig,
     get,
     post,
-    Body,
     MediaType,
-    RequestEncodingType,
-    Starlite,
-    UploadFile,
-    Template,
-    TemplateConfig,
-    HTTPException
+    Litestar,
 )
-from starlite.contrib.jinja import JinjaTemplateEngine
-import io
-import os
-import json
+from litestar.enums import RequestEncodingType
+from litestar.datastructures import UploadFile
+from litestar.params import Body
+from litestar.exceptions import HTTPException
+from litestar.contrib.jinja import JinjaTemplateEngine
+from litestar.template.config import TemplateConfig
+from litestar.response import Template
+from litestar.static_files import create_static_files_router
 
-import torch
-from torchvision import transforms  # type: ignore
-import torch.nn.functional as F
+
+from server.services.descriptions import CharactersService, Breed, CharactersRepository
+from server.services.recognizer import RecognizerService, RecognizerRepository
 
 os.environ["CUDA_VISIBLE_DEVICES"] = "-1"
 
 
-def load_model(model_path, device="cpu"):
-    model = torch.load(model_path, map_location=device, weights_only=False)
-    model.eval()
-    return model
+recognizer_service = RecognizerService(RecognizerRepository())
+characters_service = CharactersService(CharactersRepository())
 
 
-DOG_MODEL = load_model("server/models/dogs_model.pth")
-CAT_MODEL = load_model("server/models/cats_model.pth")
-
-with open("server/meta/labels_dogs.json", "r") as f:
-    data_labels = f.read()
-    labels_dogs = json.loads(data_labels)
-
-with open("server/meta/labels_cats.json", "r") as f:
-    data_labels = f.read()
-    labels_cats = json.loads(data_labels)
-
-
-with open("server/meta/images.json", "r") as f:
-    IMAGES = json.loads(f.read())
-
-def predict_image(image, model, device="cuda") -> list[tuple]:
-    img_size = (224, 224)
-    preprocess = transforms.Compose(
-        [
-            transforms.Resize(img_size),
-            transforms.ToTensor(),
-            transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5)),
-        ]
-    )
-    input_tensor = preprocess(image)
-    input_batch = input_tensor.unsqueeze(0).to(device)  # Добавляем dimension для батча
-
-    with torch.no_grad():
-        output = model(input_batch)
-
-    probabilities = torch.nn.functional.softmax(output[0], dim=0)
-    k = 5
-    topk_probs, predicted_idx = torch.topk(probabilities, k)
-
-    data = []
-    for i in range(k):
-        data.append((predicted_idx[i].item(), float(topk_probs[i].item())))
-    return data
-
-breed_dir = Path("server/meta/breed_descriptions")
-DOGS_BEERS = []
-
-# Идем по каждому текстовому файлу с описанием породы
-for breed_file in breed_dir.glob("*.txt"):
-    breed_name = breed_file.stem  # имя файла без расширения - название породы
-    description = breed_file.read_text(encoding="utf-8")  # читаем описание из файла
-    DOGS_BEERS.append({
-        "name": breed_name.replace("_", " "),
-        "alias": breed_file.stem,
-        "description": description.strip()
-    })
-DOGS_BEERS.sort(key=lambda b: b["name"])
-
-class BeerdsController(Controller):
+class BreedsController(Controller):
     path = "/beerds"
 
     @post("/dogs")
@@ -95,96 +37,69 @@ class BeerdsController(Controller):
         self, data: UploadFile = Body(media_type=RequestEncodingType.MULTI_PART)
     ) -> dict:
         body = await data.read()
-
-        img_file = Image.open(io.BytesIO(body))
-        predicted_data = predict_image(img_file, DOG_MODEL, "cpu")
-        results = {}
-        images = []
-        for d in predicted_data:
-            predicted_idx, probabilities = d
-            predicted_label = labels_dogs[str(predicted_idx)]
-            name = predicted_label.replace("_", " ")
-            images.append({
-                "name": name, 
-                "url": [f"/static/assets/dog/{predicted_label}/{i}" for i in IMAGES['dog'][predicted_label]]
-            })
-            results[probabilities] = name
-        return {
-            "results": results,
-            "images": images,
-        }
+        return await recognizer_service.predict_dog_image(body)
 
     @post("/cats")
     async def beerds_cats(
         self, data: UploadFile = Body(media_type=RequestEncodingType.MULTI_PART)
     ) -> dict:
         body = await data.read()
-
-        img_file = Image.open(io.BytesIO(body))
-        predicted_data = predict_image(img_file, CAT_MODEL, "cpu")
-        results = {}
-        images = []
-        for d in predicted_data:
-            predicted_idx, probabilities = d
-            predicted_label = labels_cats[str(predicted_idx)]
-            name = predicted_label.replace("_", " ")
-            images.append({
-                "name": name, 
-                "url": [f"/static/assets/cat/{predicted_label}/{i}" for i in IMAGES['cat'][predicted_label]]
-            })
-            results[probabilities] = predicted_label
-        return {
-            "results": results,
-            "images": images,
-        }
+        return await recognizer_service.predict_cat_image(body)
 
 
-class BaseController(Controller):
+class DescriptionController(Controller):
     path = "/"
 
     @get("/")
     async def dogs(self) -> Template:
-        return Template(name="dogs.html")
-    
+        return Template(template_name="dogs.html")
+
     @get("/cats")
     async def cats(self) -> Template:
-        return Template(name="cats.html")
-    
+        return Template(template_name="cats.html")
+
     @get("/contacts")
     async def contacts(self) -> Template:
-        return Template(name="contacts.html")
-    
+        return Template(template_name="contacts.html")
+
     @get("/donate")
     async def donate(self) -> Template:
-        return Template(name="donate.html")
-    
+        return Template(template_name="donate.html")
+
     @get("/dogs-characteristics")
     async def dogs_characteristics(self) -> Template:
-        return Template(name="dogs-characteristics.html", context={"breeds": DOGS_BEERS})
-    
+        breeds = await characters_service.get_characters()
+        return Template(
+            template_name="dogs-characteristics.html", context={"breeds": breeds}
+        )
+
     @get("/dogs-characteristics/{name:str}")
     async def beer_description(self, name: str) -> Template:
-        data = [b for b in DOGS_BEERS if b.get("alias") == name]
-        if len(data) == 0:
+        breed = await characters_service.get_character(name)
+        if breed is None:
             raise HTTPException(status_code=404, detail="Порода не найдена")
-        return Template(name="beers-description.html", context={
-            "text": markdown.markdown(data[0].get("description")),
-            "title": data[0].get("name"),
-            "images":  [f"/static/assets/dog/{name}/{i}" for i in IMAGES['dog'][name]],
-        })
-        
+        images = await recognizer_service.images_dogs()
+        return Template(
+            template_name="beers-description.html",
+            context={
+                "text": markdown.markdown(breed.description),
+                "title": breed.name,
+                "images": [f"/static/assets/dog/{name}/{i}" for i in images[name]],
+            },
+        )
 
     @get("/sitemap.xml", media_type=MediaType.XML)
     async def sitemaps(self) -> bytes:
+        breeds: list[Breed] = await characters_service.get_characters()
         lastmod = "2025-10-04T19:01:03+00:00"
         beers_url = ""
-        for b in DOGS_BEERS:
-            beers_url += f'''
+        for b in breeds:
+            beers_url += f"""
 <url>
-  <loc>https://xn-----6kcp3cadbabfh8a0a.xn--p1ai/dogs-characteristics/{b.get("alias")}</loc>
+  <loc>https://xn-----6kcp3cadbabfh8a0a.xn--p1ai/dogs-characteristics/{b.alias}</loc>
   <lastmod>{lastmod}</lastmod>
 </url>
-'''
+"""
         return f"""<?xml version="1.0" encoding="UTF-8"?>
 <urlset
       xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"
@@ -214,7 +129,6 @@ class BaseController(Controller):
 
 </urlset>
 """.encode()
-    
 
     @get("/robots.txt", media_type=MediaType.TEXT)
     async def robots(self) -> str:
@@ -226,11 +140,12 @@ Sitemap: https://xn-----6kcp3cadbabfh8a0a.xn--p1ai/sitemap.xml
 """
 
 
-app = Starlite(
+app = Litestar(
     debug=True,
-    route_handlers=[BeerdsController, BaseController],
-    static_files_config=[
-        StaticFilesConfig(directories=[Path("server/static")], path="/static"),
+    route_handlers=[
+        BreedsController,
+        DescriptionController,
+        create_static_files_router(path="/static", directories=["server/static"]),
     ],
     template_config=TemplateConfig(
         directory=Path("server/templates"),

server/services/descriptions/__init__.py

@@ -0,0 +1,13 @@
+from server.services.descriptions.repository import (
+    CharactersRepository,
+    ACharactersRepository,
+)
+from server.services.descriptions.service import CharactersService
+from server.services.descriptions.domain import Breed
+
+__all__ = (
+    "CharactersRepository",
+    "ACharactersRepository",
+    "CharactersService",
+    "Breed",
+)

server/services/descriptions/domain.py

@@ -0,0 +1,8 @@
+from dataclasses import dataclass
+
+
+@dataclass(frozen=True)
+class Breed:
+    name: str
+    alias: str
+    description: str

server/services/descriptions/repository/__init__.py

@@ -0,0 +1,6 @@
+from server.services.descriptions.repository.repository import (
+    CharactersRepository,
+    ACharactersRepository,
+)
+
+__all__ = ("CharactersRepository", "ACharactersRepository")