From 4aa18c3b0e5fde6ba5038f0ab9355564753e742d Mon Sep 17 00:00:00 2001
From: Maciej Sobkowiak <masobkowiak@gmail.com>
Date: Sat, 12 Jun 2021 22:59:33 +0200
Subject: [PATCH] dvc.yaml

---
 dvc.yaml        | 17 +++++++++++++
 dvc_training.py | 68 +++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 85 insertions(+)
 create mode 100644 dvc.yaml
 create mode 100644 dvc_training.py

diff --git a/dvc.yaml b/dvc.yaml
new file mode 100644
index 0000000..48f6552
--- /dev/null
+++ b/dvc.yaml
@@ -0,0 +1,17 @@
+stages:
+    preprocess:
+      cmd: python3 preprocesing.py
+      deps:
+        - preprocesing.py
+        - who_suicide_statistics.csv
+      outs:
+        - train.csv
+    train:
+      cmd: python3 training.py
+      deps:
+      - training.py
+      - train.csv
+      outs:
+      - results.csv
+      - suicide_model.h5
+
diff --git a/dvc_training.py b/dvc_training.py
new file mode 100644
index 0000000..4a3ee14
--- /dev/null
+++ b/dvc_training.py
@@ -0,0 +1,68 @@
+import sys
+import pandas as pd
+import numpy as np
+import matplotlib.pyplot as plt
+import os
+from countries_map import countries
+import tensorflow as tf
+from tensorflow.keras.layers import Input, Dense, Activation, Dropout
+from tensorflow.keras.models import Model
+from tensorflow.keras.callbacks import EarlyStopping
+from keras.models import Sequential
+from sklearn.metrics import mean_squared_error
+from tensorflow import keras
+from tensorflow.keras import layers
+from tensorflow.keras.layers.experimental import preprocessing
+
+EPOCHS = int(sys.argv[1])
+BATCH_SIZE = int(sys.argv[2])
+
+train = pd.read_csv('train.csv')
+validate = pd.read_csv('validate.csv')
+test = pd.read_csv('test.csv')
+
+# podział train set
+X_train = train.loc[:, train.columns != 'suicides_no']
+y_train = train[['suicides_no']]
+X_test = test.loc[:, train.columns != 'suicides_no']
+y_test = test[['suicides_no']]
+
+normalizer = preprocessing.Normalization()
+normalizer.adapt(np.array(X_train))
+
+first = np.array(X_train[:1])
+with np.printoptions(precision=2, suppress=True):
+    print('First example:', first)
+    print()
+    print('Normalized:', normalizer(first).numpy())
+
+model = tf.keras.Sequential([
+    normalizer,
+    layers.Dense(units=1)
+])
+model.predict(X_train[:10])
+
+# Compile model
+model.compile(
+    optimizer=tf.optimizers.Adam(learning_rate=0.1),
+    loss='mean_absolute_error')
+
+# Train model
+history = model.fit(
+    X_train, y_train,
+    batch_size=BATCH_SIZE,
+    epochs=EPOCHS,
+    validation_split=0.2)
+
+model.save_weights('suicide_model.h5')
+
+test_results = {}
+
+test_results['model'] = model.evaluate(
+    X_test, y_test, verbose=0)
+
+test_predictions = model.predict(X_test).flatten()
+
+predictions = model.predict(X_test)
+pd.DataFrame(predictions).to_csv('results.csv')
+model.summary()