jobs

Dockerfile

@@ -1,10 +1,10 @@
 FROM ubuntu:latest
 
-RUN apt-get update && apt-get install -y python3-pip && pip3 install setuptools && pip3 install numpy && pip3 install pandas && pip3 install wget && pip3 install scikit-learn && rm -rf /var/lib/apt/lists/*
+RUN apt-get update && apt-get install -y python3-pip && pip3 install setuptools && pip3 install numpy && pip3 install pandas && pip3 install wget && pip3 install scikit-learn && pip3 install matplotlib && rm -rf /var/lib/apt/lists/*
 RUN pip3 install torch torchvision torchaudio 
 WORKDIR /app
 
 COPY ./create.py ./
 COPY ./stats.py ./
 COPY ./stroke-pytorch.py ./
-
+COPY ./stroke-pytorch-eval.py ./

pytorch-training-eval/Dockerfile

@@ -1,10 +1,11 @@
 FROM ubuntu:latest
 
-RUN apt-get update && apt-get install -y python3-pip && pip3 install setuptools && pip3 install numpy && pip3 install pandas && pip3 install wget && pip3 install scikit-learn && rm -rf /var/lib/apt/lists/*
+RUN apt-get update && apt-get install -y python3-pip && pip3 install setuptools && pip3 install numpy && pip3 install pandas && pip3 install wget && pip3 install scikit-learn && pip3 install matplotlib && rm -rf /var/lib/apt/lists/*
 RUN pip3 install torch torchvision torchaudio 
 WORKDIR /app
 
 COPY ./create.py ./
 COPY ./stats.py ./
 COPY ./stroke-pytorch.py ./
+COPY ./stroke-pytorch-eval.py ./
 

pytorch-training-eval/Jenkinsfile-eval

@@ -0,0 +1,32 @@
+pipeline {
+   agent {
+      dockerfile true
+   }
+   stages {
+      stage('checkout') {
+            steps {
+                git 'https://git.wmi.amu.edu.pl/s434766/ium_434766.git'
+                copyArtifacts fingerprintArtifacts: true, projectName: 's434766-create-dataset'
+                copyArtifacts fingerprintArtifacts: true, projectName: 's434766-training'
+            }
+        }
+      stage('Docker'){
+         steps{
+            sh 'python3 "./stroke-pytorch-eval.py" >> eval.txt'
+         }
+      }
+      stage('archiveArtifacts') {
+         steps {
+            archiveArtifacts 'eval.txt'
+
+         }
+                     post {
+                        success {
+                           emailext body: 'Evaluation of stroke predictions is finished',
+                           subject: 's434766 evaluation finished',
+                           to: '26ab8f35.uam.onmicrosoft.com@emea.teams.ms'
+                }
+            }
+      }
+   }
+}

pytorch-training-eval/Jenkinsfile-train

@@ -11,18 +11,20 @@ pipeline {
         }
       stage('Docker'){
          steps{
-            sh 'python3 "./stroke-pytorch.py" > model.txt'
-         }
-      }
-      stage('checkout: Check out from version control') {
-         steps {
-            git 'https://git.wmi.amu.edu.pl/s434766/ium_434766.git'
+            sh 'python3 "./stroke-pytorch.py" ${BATCH_SIZE} ${EPOCHS} > pred.txt'
          }
       }
       stage('archiveArtifacts') {
          steps {
-            archiveArtifacts 'model.txt'
+            archiveArtifacts 'pred.txt'
             archiveArtifacts 'stroke.pkl'
+         }
+                     post {
+                        success {
+                           emailext body: 'Training of stroke predictions is finished',
+                           subject: 's434766 training finished',
+                           to: '26ab8f35.uam.onmicrosoft.com@emea.teams.ms'
+                }
             }
       }
    }

stroke-pytorch-eval.py

@@ -0,0 +1,50 @@
+import torch
+import torch.nn as nn
+import numpy as np
+from os import path
+import torch.nn.functional as F
+from torch import nn
+from torch.autograd import Variable
+import torchvision.transforms as transforms
+import pandas as pd
+from sklearn.metrics import accuracy_score
+from sklearn.metrics import mean_squared_error
+from sklearn.metrics import classification_report
+import matplotlib.pyplot as plt
+
+class LogisticRegressionModel(nn.Module):
+    def __init__(self, input_dim, output_dim):
+        super(LogisticRegressionModel, self).__init__()
+        self.linear = nn.Linear(input_dim, output_dim)
+        self.sigmoid = nn.Sigmoid()
+    def forward(self, x):
+        out = self.linear(x)
+        return self.sigmoid(out)
+
+np.set_printoptions(suppress=False)
+
+data_test  = pd.read_csv("data_test.csv")
+
+FEATURES   =  ['age','hypertension','heart_disease','ever_married', 'avg_glucose_level', 'bmi']
+
+
+x_test = data_test[FEATURES].astype(np.float32)
+y_test = data_test['stroke'].astype(np.float32)
+
+fTest= torch.from_numpy(x_test.values)
+tTest = torch.from_numpy(y_test.values)
+
+model = LogisticRegressionModel(6,1)
+model.load_state_dict(torch.load('stroke.pth'))
+y_pred = model(fTest)
+
+rmse = mean_squared_error(tTest, y_pred.detach().numpy())
+acc = accuracy_score(tTest, np.argmax(y_pred.detach().numpy(), axis=1))
+print('-' * 60)
+print(classification_report(tTest, y_pred.detach().numpy().round()))
+print(f"    RMSE: {rmse}")
+print(f"    Accuracy: {acc}")
+print('-' * 60)
+
+
+

stroke-pytorch.py

@@ -1,4 +1,5 @@
 import torch
+import sys
 import torch.nn.functional as F
 from torch import nn
 from torch.autograd import Variable
@@ -9,6 +10,8 @@ from sklearn.metrics import accuracy_score
 import numpy as np
 import pandas as pd
 
+np.set_printoptions(suppress=False)
+
 class LogisticRegressionModel(nn.Module):
     def __init__(self, input_dim, output_dim):
         super(LogisticRegressionModel, self).__init__()
@@ -35,9 +38,8 @@ tTrain = torch.from_numpy(y_train.values.reshape(2945,1))
 fTest= torch.from_numpy(x_test.values)
 tTest = torch.from_numpy(y_test.values)
 
-batch_size = 95
-n_iters = 1000
-num_epochs = int(n_iters / (len(x_train) / batch_size))
+batch_size = int(sys.argv[1]) if len(sys.argv) > 1 else 16
+num_epochs = int(sys.argv[2]) if len(sys.argv) > 2 else 5
 learning_rate = 0.001
 input_dim = 6
 output_dim = 1
@@ -48,7 +50,7 @@ criterion = torch.nn.BCELoss(reduction='mean')
 optimizer = torch.optim.SGD(model.parameters(), lr = learning_rate)
 
 for epoch in range(num_epochs):
-    print ("Epoch #",epoch)
+    # print ("Epoch #",epoch)
     model.train()
     optimizer.zero_grad()
     # Forward pass
@@ -64,6 +66,4 @@ for epoch in range(num_epochs):
 y_pred = model(fTest)
 print("predicted Y value: ", y_pred.data)
 
-print ("The accuracy is", accuracy_score(tTest, np.argmax(y_pred.detach().numpy(), axis=1)))
-
-torch.save(model, 'stroke.pkl')
+torch.save(model.state_dict(), 'stroke.pth')