some changes

Jenkinsfile

@@ -1,5 +1,10 @@
 pipeline {
-    agent { dockerfile true }
+    agent { 
+        dockerfile {
+        filename 'Dockerfile'
+        args '-v /tmp:/tmp'
+        }
+    }
 
    parameters{
       buildSelector(

mlruns/0/f5c7f50e058248b0b2ccdb2a43dbac38/meta.yaml

@@ -0,0 +1,15 @@
+artifact_uri: file:///home/tomasz/IUM/ium_434695/mlruns/0/f5c7f50e058248b0b2ccdb2a43dbac38/artifacts
+end_time: 1621803527317
+entry_point_name: ''
+experiment_id: '0'
+lifecycle_stage: active
+name: ''
+run_id: f5c7f50e058248b0b2ccdb2a43dbac38
+run_uuid: f5c7f50e058248b0b2ccdb2a43dbac38
+source_name: ''
+source_type: 4
+source_version: ''
+start_time: 1621803499546
+status: 3
+tags: []
+user_id: tomasz

mlruns/0/f5c7f50e058248b0b2ccdb2a43dbac38/metrics/rmse

@@ -0,0 +1 @@
+1621803525760 0.0427710843373494 0

mlruns/0/f5c7f50e058248b0b2ccdb2a43dbac38/params/batch_size

@@ -0,0 +1 @@
+16

mlruns/0/f5c7f50e058248b0b2ccdb2a43dbac38/params/epochs

@@ -0,0 +1 @@
+15

mlruns/0/f5c7f50e058248b0b2ccdb2a43dbac38/tags/mlflow.source.git.commit

@@ -0,0 +1 @@
+eee8d50045b94596c5132c29b393ab28d589ca2c

mlruns/0/f5c7f50e058248b0b2ccdb2a43dbac38/tags/mlflow.source.name

@@ -0,0 +1 @@
+/home/tomasz/IUM/ium_434695/vgsales-mlflow.py

mlruns/0/f5c7f50e058248b0b2ccdb2a43dbac38/tags/mlflow.source.type

@@ -0,0 +1 @@
+LOCAL

mlruns/0/f5c7f50e058248b0b2ccdb2a43dbac38/tags/mlflow.user

@@ -0,0 +1 @@
+tomasz

my_model/MLmodel

@@ -18,4 +18,4 @@ signature:
     {"name": "JP_Sales", "type": "double"}, {"name": "Other_Sales", "type": "double"},
     {"name": "Global_Sales", "type": "double"}]'
   outputs: '[{"type": "long"}]'
-utc_time_created: '2021-05-23 17:35:24.878064'
+utc_time_created: '2021-05-23 20:58:45.764716'

predict434788.py

@@ -0,0 +1,7 @@
+import mlflow
+import pandas as pd
+
+model = mlflow.keras.load_model("my_model")
+data = pd.read_json('my_model/input_example.json', orient='split')
+
+print(model.predict(data))

vgsales-mlflow.py

@@ -35,7 +35,7 @@ def my_main(epochs, batch_size):
     early_stop = EarlyStopping(monitor="val_loss", mode="min", verbose=1, patience=10)
 
 
-    model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])
+    model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
 
 
     model.fit(X_train, y_train, epochs=epochs, batch_size=batch_size, validation_data=(X_test, y_test))