import os
import pickle
import bentoml

@bentoml.service(name="sklearn_inference")
class Service:
    def __init__(self) -> None:
        model_path = os.getenv("MODEL_PATH", "/workspace/model")
        with open(f"{model_path}/model.pkl", "rb") as f:
            self.model = pickle.load(f)

    @bentoml.api
    def predict(self, inputs: dict) -> dict:
        features = inputs.get("features")
        prediction = self.model.predict([features])[0]
        return {"prediction": float(prediction)}

import os
import bentoml
from catboost import CatBoostClassifier

@bentoml.service(name="catboost_inference")
class Service:
    def __init__(self) -> None:
        model_path = os.getenv("MODEL_PATH", "/workspace/model")
        self.model = CatBoostClassifier()
        self.model.load_model(f"{model_path}/catboost_model.cbm")

    @bentoml.api
    def predict(self, inputs: dict) -> dict:
        features = inputs.get("features")
        prediction = self.model.predict([features])[0]
        probability = self.model.predict_proba([features])[0]
        return {
            "prediction": int(prediction),
            "probability": float(probability.max()),
        }

import os
import bentoml
import torch

@bentoml.service(name="pytorch_inference")
class Service:
    def __init__(self) -> None:
        model_path = os.getenv("MODEL_PATH", "/workspace/model")
        self.model = torch.load(f"{model_path}/model.pt", weights_only=False)
        self.model.eval()

    @bentoml.api
    def predict(self, inputs: dict) -> dict:
        features = torch.tensor(inputs.get("features"))
        with torch.no_grad():
            output = self.model(features)
            prediction = output.argmax().item()
        return {"prediction": prediction}

import os
import logging
from typing import Any, Dict, Optional

import bentoml
from transformers import pipeline

logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)

MODEL_PATH = os.getenv("MODEL_PATH", "/workspace/model")
TASK = os.getenv("TASK", "text-classification")

@bentoml.service(name="inference")
class Service:
    def __init__(self) -> None:
        logger.info(f"Load transformers pipeline: task={TASK}, model_path={MODEL_PATH}")
        self.pipeline = pipeline(TASK, model=MODEL_PATH, tokenizer=MODEL_PATH)

    @bentoml.api
    def predict(self, inputs: Any, parameters: Optional[Dict[str, Any]] = None) -> Dict[str, Any]:
        """Универсальный endpoint. Поддерживает строку или список строк."""
        params = parameters or {}
        data = inputs if isinstance(inputs, list) else [inputs]
        outputs = self.pipeline(data, **params)
        return {"results": outputs, "count": len(outputs)}

curl -X POST "https://qubu.ai/api/models/salary-predict/inference/proxy/predict" \
  -H "Authorization: Bearer <API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{
    "inputs": {
      "skills": ["Python", "ML"],
      "variable_1": "remote",
      "variable_2": "between1And3",
      "variable_3": "fullDay"
    }
  }'

import requests

url = "https://qubu.ai/api/models/salary-predict/inference/proxy/predict"
headers = {
    "Authorization": "Bearer <API_KEY>",
    "Content-Type": "application/json"
}
data = {
    "inputs": {
        "skills": ["Python", "ML"],
        "variable_1": "remote",
        "variable_2": "between1And3",
        "variable_3": "fullDay"
    }
}

response = requests.post(url, headers=headers, json=data)
print(response.json())

const response = await fetch(
  'https://qubu.ai/api/models/salary-predict/inference/proxy/predict',
  {
    method: 'POST',
    headers: {
      'Authorization': 'Bearer <API_KEY>',
      'Content-Type': 'application/json'
    },
    body: JSON.stringify({
      inputs: {
        skills: ['Python', 'ML'],
        variable_1: 'remote',
        variable_2: 'between1And3',
        variable_3: 'fullDay'
      }
    })
  }
);

const data = await response.json();
console.log(data);

curl -X POST "https://qubu.ai/api/models/model-name/inference/proxy/predict" \
  -H "Content-Type: application/json" \
  -d '{"inputs": {...}}'

curl -X POST "https://qubu.ai/api/models/model-name/inference/proxy/predict" \
  -H "Authorization: Bearer <API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{"inputs": {...}}'

curl -X POST "https://qubu.ai/api/models/model-name/inference/proxy/predict" \
  -H "Authorization: Bearer <API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{"inputs": {...}}'