mlflow

train.py

@@ -7,9 +7,19 @@ from sklearn.metrics import accuracy_score
 from sklearn.preprocessing import StandardScaler
 import pandas as pd
 import numpy as np
+import os
+import mlflow
+import mlflow.sklearn
+import logging
+
+logging.basicConfig(level=logging.WARN)
+logger = logging.getLogger(__name__)
+mlflow.set_tracking_uri("http://localhost:5000")
+mlflow.set_experiment("s487194")
 
 # Wczytanie danych
-data = pd.read_csv("Sales.csv")
+print(os.getcwd())
+data = pd.read_csv("./Sales.csv")
 
 # Przygotowanie danych
 data["Profit_Category"] = pd.cut(data["Profit"], bins=[-np.inf, 500, 1000, np.inf], labels=[0, 1, 2])
@@ -47,7 +57,6 @@ class ANN_Model(nn.Module):
 
 torch.manual_seed(20)
 model=ANN_Model()
-model.parameters
 
 def calculate_accuracy(model, X, y):
     with torch.no_grad():
@@ -65,6 +74,14 @@ epochs = 100
 final_losses = []
 accuracy_list = []
 
+with mlflow.start_run() as run:
+    # Logowanie parametrów modelu
+    mlflow.log_param("hidden_layer_1", 20)
+    mlflow.log_param("hidden_layer_2", 20)
+    mlflow.log_param("output_layer", 3)
+    mlflow.log_param("learning_rate", 0.01)
+    mlflow.log_param("epochs", epochs)
+
     for i in range(epochs):
         i = i + 1
         y_pred = model(X_train)
@@ -73,9 +90,18 @@ for i in range(epochs):
 
         train_accuracy = calculate_accuracy(model, X_train, y_train)
         test_accuracy = calculate_accuracy(model, X_test, y_test)
+
+        # Logowanie metryk po każdej epoce
+        mlflow.log_metric("train_loss", loss.item(), step=i)
+        mlflow.log_metric("train_accuracy", train_accuracy, step=i)
+        mlflow.log_metric("test_accuracy", test_accuracy, step=i)
+
         print(f"Epoch: {i}, Loss: {loss.item()}, Train Accuracy: {train_accuracy}%, Test Accuracy: {test_accuracy}%")
         optimizer.zero_grad()
         loss.backward()
         optimizer.step()
 
 torch.save(model,"classificationn_model.pt")
+
+
+