Add initial Dockerfile, API implementation, and requirements

Dockerfile

@@ -0,0 +1,17 @@
+FROM python:3.13-slim
+
+WORKDIR /code
+
+COPY ./requirements.txt /code/requirements.txt
+
+RUN apt-get update && apt-get install -y procps && \
+    pip install -U pip && \
+    rm /etc/localtime && \
+    ln -s /usr/share/zoneinfo/America/Mexico_City /etc/localtime && \
+    pip install -r ./requirements.txt
+
+COPY ./api.py /code/
+
+EXPOSE 8000
+
+CMD ["uvicorn","api:app", "--host", "0.0.0.0", "--port", "8000"]

README.md

@@ -5,6 +5,7 @@ colorFrom: blue
 colorTo: blue
 sdk: docker
 pinned: false
+app_port: 8000
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

api.py

@@ -0,0 +1,82 @@
+import pickle
+import mlflow
+from fastapi import FastAPI
+from pydantic import BaseModel
+from mlflow import MlflowClient
+from dotenv import load_dotenv
+import os
+import pandas as pd
+import xgboost as xgb
+
+load_dotenv(override=True)  # Carga las variables del archivo .env
+
+mlflow.set_tracking_uri("databricks")
+client = MlflowClient()
+
+EXPERIMENT_NAME = "/Users/[email protected]/nyc-taxi-experiment-prefect"
+
+run_ = mlflow.search_runs(order_by=['metrics.rmse ASC'],
+                          output_format="list",
+                          experiment_names=[EXPERIMENT_NAME]
+                          )[0]
+
+run_id = run_.info.run_id
+
+run_uri = f"runs:/{run_id}/preprocessor"
+
+client.download_artifacts(
+    run_id=run_id,
+    path='preprocessor',
+    dst_path='.'
+)
+
+with open("preprocessor/preprocessor.b", "rb") as f_in:
+    dv = pickle.load(f_in)
+
+model_name = "workspace.default.nyc-taxi-model-prefect"
+alias = "champion"
+
+model_uri = f"models:/{model_name}@{alias}"
+
+champion_model = mlflow.pyfunc.load_model(
+    model_uri=model_uri
+)
+
+def preprocess(input_data):
+
+    input_dict = {
+        'PU_DO': input_data.PULocationID + "_" + input_data.DOLocationID,
+        'trip_distance': input_data.trip_distance,
+    }
+    X = dv.transform([input_dict])
+
+    # Names depend on sklearn version
+    try:
+        cols = dv.get_feature_names_out()
+    except AttributeError:
+        cols = dv.get_feature_names()
+
+    # 
+    X_df = pd.DataFrame(X.toarray(), columns=cols)
+
+    return X_df
+
+def predict(input_data):
+
+    X_val = preprocess(input_data)
+    
+    return champion_model.predict(X_val)
+
+app = FastAPI()
+
+class InputData(BaseModel):
+    PULocationID: str
+    DOLocationID: str
+    trip_distance: float
+
+
+@app.post("/api/v1/predict")
+def predict_endpoint(input_data: InputData):
+    result = predict(input_data)[0]
+    print(f"Prediction: {result}")
+    return {"prediction": float(result)}

requirements.txt

@@ -0,0 +1,8 @@
+fastapi==0.121.0
+mlflow==3.4.0
+mlflow_skinny==3.4.0
+mlflow_tracing==3.4.0
+pandas==2.3.3
+pydantic==2.12.4
+python-dotenv==1.2.1
+xgboost==3.1.1