mlflow test

learning/Jenkinsfile

@@ -21,6 +21,8 @@ pipeline {
         stage('learning') {
             steps {
                 sh 'python ./learning/ml.py'
+                sh "export MLFLOW_TRACKING_URI=http://172.17.0.1:5000"
+                sh "python ./learning/ml-mlflow.py"
             }
         }
         stage('archiveArtifacts') {

learning/ml-mlflow.py

@@ -0,0 +1,87 @@
+import torch
+import mlflow
+from urllib.parse import urlparse
+import torch.nn as nn
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+import torch
+import datetime
+from torch.autograd import Variable
+
+
+INPUT_DIM = 1
+OUTPUT_DIM = 1
+LEARNING_RATE = 0.01
+EPOCHS = 100
+
+
+mlflow.set_tracking_uri("http://172.17.0.1:5000")
+dataset = pd.read_csv('datasets/train_set.csv')
+testset = pd.read_csv('datasets/test_set.csv')
+
+x_values = [datetime.datetime.strptime(
+    item, "%Y-%m-%d").month for item in dataset['date'].values]
+x_train = np.array(x_values, dtype=np.float32)
+x_train = x_train.reshape(-1, 1)
+
+y_values = [min(dataset['result_1'].values[i]/dataset['result_2'].values[i], dataset['result_2'].values[i] /
+                dataset['result_1'].values[i]) for i in range(len(dataset['result_1'].values))]
+y_train = np.array(y_values, dtype=np.float32)
+y_train = y_train.reshape(-1, 1)
+
+
+class LinearRegression(torch.nn.Module):
+    def __init__(self, inputSize, outputSize):
+        super(LinearRegression, self).__init__()
+        self.linear = torch.nn.Linear(inputSize, outputSize)
+
+    def forward(self, x):
+        out = self.linear(x)
+        return out
+
+
+model = LinearRegression(INPUT_DIM, OUTPUT_DIM)
+
+
+criterion = torch.nn.MSELoss()
+optimizer = torch.optim.SGD(model.parameters(), lr=LEARNING_RATE)
+
+for epoch in range(EPOCHS):
+    inputs = Variable(torch.from_numpy(x_train))
+    labels = Variable(torch.from_numpy(y_train))
+
+    optimizer.zero_grad()
+
+    outputs = model(inputs)
+
+    loss = criterion(outputs, labels)
+    print(loss)
+    loss.backward()
+
+    optimizer.step()
+
+    print('epoch {}, loss {}'.format(epoch, loss.item()))
+
+torch.save(model.state_dict(), 'model.pt')
+
+with mlflow.start_run():
+    test_input = x_train[0]
+
+    mlflow.log_param("train size", dataset.size)
+    mlflow.log_param("test size", testset.size)
+    mlflow.log_param("epochs", EPOCHS)
+
+    signature = mlflow.models.signature.infer_signature(
+        x_train.values, model.predict(x_train.values))
+
+    mlflow.set_experiment("s434700")
+    tracking_url_type_store = urlparse(mlflow.get_tracking_uri()).scheme
+    if tracking_url_type_store != "file":
+        mlflow.keras.log_model(model, "model.pt", registered_model_name="s434700", signature=signature,
+                               input_example=test_input)
+    else:
+        mlflow.keras.log_model(model, "model.pt",
+                               signature=signature, input_example=test_input)
+        mlflow.keras.save_model(
+            model, "model.pt", signature=signature, input_example=test_input)

learning/ml.py

@@ -3,7 +3,6 @@ import torch.nn as nn
 import numpy as np
 import pandas as pd
 import matplotlib.pyplot as plt
-import seaborn as sns
 import torch
 import datetime
 from torch.autograd import Variable

requirements.txt

@@ -5,4 +5,5 @@ numpy
 matplotlib
 seaborn
 sacred
-pymongo
+pymongo
+mlflow