Upload app.py

app.py

@@ -0,0 +1,174 @@
+import os
+os.system('git clone --recursive https://github.com/dmlc/xgboost')
+os.system('cd xgboost')
+os.system('sudo cp make/minimum.mk ./config.mk;')
+os.system('sudo make -j4;')
+os.system('sh build.sh')
+os.system('cd python-package')
+os.system('python setup.py install')
+os.system('pip install graphviz')
+os.system('pip install python-pydot')
+os.system('pip install python-pydot-ng')
+os.system('pip install -U scikit-learn scipy matplotlib')
+os.system('pip install wandb --upgrade')
+os.system('pip install tensorboardX --upgrade')
+os.system('pip install ipython --upgrade')
+os.system('wandb login 5a0e81f39777351977ce52cf57ea09c4f48f3d93 --relogin')
+
+from collections import namedtuple
+import altair as alt
+import math
+import streamlit as st
+import pandas
+import numpy
+import xgboost
+import graphviz
+from sklearn.metrics import mean_squared_error
+from sklearn.model_selection import train_test_split
+import matplotlib.pyplot
+os.system('load_ext tensorboard')
+import os
+import datetime
+from tensorboardX import SummaryWriter
+import wandb
+from wandb.xgboost import wandb_callback
+
+wandb.init(project="australian_rain", entity="epitech1")
+
+"""
+# MLOPS
+"""
+
+
+max_depth_input = st.slider("Max depth", 1, 100, 5)
+colsample_bytree_input = st.slider("Colsample bytree", 0.0, 1.0, 0.5)
+learning_rate_input = st.slider("Learning rate", 0.0, 1.0, 0.2)
+alpha_input = st.slider("Alpha", 1, 100, 10)
+n_estimators_input = st.slider("n estimators", 1, 100, 20)
+city_input = st.selectbox(
+     'Which city do you want to predict rain ?',
+     ("Canberra",
+    "Albury",
+    "Penrith",
+    "Sydney",
+    "MountGinini",
+    "Bendigo",
+    "Brisbane",
+    "Portland"), index=0)
+
+dataset = pandas.read_csv('weatherAUS.csv')
+
+location_dataset = dataset["Location"].unique()
+wind_dataset = dataset["WindGustDir"].unique()
+date_dataset = dataset["Date"].unique()
+
+dataset.drop(dataset.loc[dataset['Location'] != city_input].index, inplace=True)
+
+i_RainTomorrow = dataset.columns.get_loc("RainTomorrow")
+#i_Location = dataset.columns.get_loc("Location")
+i_WindGustDir = dataset.columns.get_loc("WindGustDir")
+i_Date = dataset.columns.get_loc("Date")
+yes = dataset.iat[8, dataset.columns.get_loc("RainTomorrow")]
+no = dataset.iat[0, dataset.columns.get_loc("RainTomorrow")]
+
+for i in range(len(dataset)):
+    if (dataset.iat[i, i_RainTomorrow] == yes):
+        dataset.iat[i, i_RainTomorrow] = True
+    else:
+        dataset.iat[i, i_RainTomorrow] = False
+    #dataset.iat[i, i_Location] = numpy.where(location_dataset == dataset.iat[i, i_Location])[0][0]
+    if (pandas.isna(dataset.iat[i, i_WindGustDir])):
+        dataset.iat[i, i_WindGustDir] = 0
+    else:
+        dataset.iat[i, i_WindGustDir] = numpy.where(wind_dataset == dataset.iat[i, i_WindGustDir])[0][0] + 1
+    dataset.iat[i, i_Date] = numpy.where(date_dataset == dataset.iat[i, i_Date])[0][0]
+    
+    
+dataset = dataset.astype({'RainTomorrow': 'bool'})
+#dataset = dataset.astype({'Location': 'int'})
+dataset = dataset.astype({'WindGustDir': 'int'})
+dataset = dataset.astype({'Date': 'int'})
+
+dataset.drop(columns=["WindDir9am", "WindDir3pm", "WindSpeed9am", "WindSpeed3pm", "Temp9am", "Temp3pm", "RainToday"], inplace=True)
+dataset.drop(dataset.index[dataset.isnull().any(axis=1)], 0, inplace=True)
+
+dataset["Humidity"] = 0.0
+dataset["Pressure"] = 0.0
+dataset["Cloud"] = 0.0
+
+for i in dataset.index:
+    humidity = (dataset["Humidity9am"][i] + dataset["Humidity3pm"][i]) / 2
+    dataset.at[i, "Humidity"] = humidity
+    pressure = (dataset["Pressure9am"][i] + dataset["Pressure3pm"][i]) / 2
+    dataset.at[i, "Pressure"] = pressure
+    cloud = (dataset["Cloud9am"][i] + dataset["Cloud3pm"][i]) / 2
+    dataset.at[i, "Cloud"] = cloud
+
+dataset.drop(columns=["Humidity9am", "Humidity3pm", "Pressure9am", "Pressure3pm", "Cloud9am", "Cloud3pm"], inplace=True)
+
+x, y = dataset.iloc[:,[False, False, True, True, False, True, True, True, True, True, True, True, True]],dataset.iloc[:,4]
+
+data_dmatrix = xgboost.DMatrix(data=x,label=y)
+
+X_train, X_test, y_train, y_test = train_test_split(x, y, test_size=0.2, random_state=123)
+
+class TensorBoardCallback(xgboost.callback.TrainingCallback):
+    def __init__(self, experiment: str = None, data_name: str = None):
+        self.experiment = experiment or "logs"
+        self.data_name = data_name or "test"
+        self.datetime_ = datetime.datetime.now().strftime("%Y%m%d-%H%M%S")
+        self.log_dir = f"runs/{self.experiment}/{self.datetime_}"
+        self.train_writer = SummaryWriter(log_dir=os.path.join(self.log_dir, "train/"))
+        if self.data_name:
+            self.test_writer = SummaryWriter(log_dir=os.path.join(self.log_dir, f"{self.data_name}/"))
+
+    def after_iteration(
+        self, model, epoch: int, evals_log: xgboost.callback.TrainingCallback.EvalsLog
+    ) -> bool:
+        if not evals_log:
+            return False
+
+        for data, metric in evals_log.items():
+            for metric_name, log in metric.items():
+                score = log[-1][0] if isinstance(log[-1], tuple) else log[-1]
+                if data == "train":
+                    self.train_writer.add_scalar(metric_name, score, epoch)
+                else:
+                    self.test_writer.add_scalar(metric_name, score, epoch)
+
+        return False
+        
+xg_reg = xgboost.XGBRegressor(colsample_bytree = colsample_bytree_input, learning_rate = learning_rate_input, max_depth = max_depth_input, alpha = alpha_input, n_estimators = n_estimators_input, eval_metric = ['rmse', 'error', 'logloss', 'map'],
+        callbacks=[TensorBoardCallback(experiment='exp_1', data_name='test')])
+
+xg_reg.fit(X_train,y_train, eval_set=[(X_train, y_train)])
+
+preds = xg_reg.predict(X_test)
+
+rmse = numpy.sqrt(mean_squared_error(y_test, preds))
+st.write("RMSE: %f" % (rmse))
+
+params = {'colsample_bytree': colsample_bytree_input,'learning_rate': learning_rate_input,
+                'max_depth': max_depth_input, 'alpha': alpha_input}
+
+cv_results = xgboost.cv(dtrain=data_dmatrix, params=params, nfold=3,
+                    num_boost_round=50,early_stopping_rounds=10,metrics="rmse", as_pandas=True, seed=123)
+
+st.write((cv_results["test-rmse-mean"]).tail(1))
+
+xg_reg = xgboost.train(params=params, dtrain=data_dmatrix, num_boost_round=10)
+
+os.system('tensorboard --logdir runs')
+
+#xgboost.plot_tree(xg_reg,num_trees=0)
+#matplotlib.pyplot.rcParams['figure.figsize'] = [200, 200]
+#matplotlib.pyplot.show()
+
+#xgboost.plot_importance(xg_reg)
+#matplotlib.pyplot.rcParams['figure.figsize'] = [5, 5]
+#matplotlib.pyplot.show()
+
+#xg_reg = xgboost.train(params=params, dtrain=data_dmatrix, num_boost_round=10, callbacks=[wandb_callback()])
+
+# MLOPS - W&B analytics
+# added the wandb to the callbacks