Update app.py

app.py

@@ -1,68 +1,68 @@
-from datetime import datetime
-import numpy as np
-import pandas as pd
-from sklearn.ensemble import RandomForestRegressor
-import gradio as gr
-import os
-import plotly.graph_objects as go
-from huggingface_hub import from_pretrained_keras
-def predictPPM(df, split):
-    ts= pd.read_csv('datappm.csv')
-    df2 =ts.copy()
-    ttSplit=split/100
-    ts['Date']=pd.to_datetime(ts['Date'])
-    ts.rename(columns={'#PPM':'PPM'},inplace=True)
-    ts=ts.set_index(['Date'])
-    ts['months'] = [x.month for x in ts.index]
-    ts['years'] = [x.year for x in ts.index]
-    ts.reset_index(drop=True, inplace=True)
-
-    # Split Data
-    X=ts.drop("PPM",axis=1)
-    Y= ts["PPM"]
-    X_train=X[:int (len(Y)*ttSplit)] 
-    X_test=X[int(len(Y)*ttSplit):]
-    Y_train=Y[:int (len(Y)*ttSplit)] 
-    Y_test=Y[int(len(Y)*ttSplit):]
-
-    # fit the model
-    rf = RandomForestRegressor()
-    rf.fit(X_train, Y_train)
-    
-    df1=df2.set_index(['Date'])
-    df1.rename(columns={'#PPM':'PPM'},inplace=True)
-    train=df1.PPM[:int (len(ts.PPM)*ttSplit)]
-    test=df1.PPM[int(len(ts.PPM)*ttSplit):]
-    preds=rf.predict(X_test).astype(int) 
-    predictions=pd.DataFrame(preds,columns=['PPM'])
-    predictions.index=test.index
-    predictions.reset_index(inplace=True)
-    predictions['Date']=pd.to_datetime(predictions['Date'])
-    print(predictions)
-    
-    #combine all into one table
-    ts_df=df.copy()
-    ts_df.rename(columns={'#PPM':'PPM'},inplace=True)
-    train= ts_df[:int (len(ts_df)*ttSplit)]
-    test= ts_df[int(len(ts_df)*ttSplit):] 
-
-    df2['Date']=pd.to_datetime(df2['Date'])
-    df2.rename(columns={'#PPM':'PPM'},inplace=True)
-    df3= predictions
-    df2['origin']='ground truth'
-    df3['origin']='prediction'
-    df4=pd.concat([df2, df3])
-    print(df4)
-    return df4
-
-demo = gr.Interface(
-    fn =predictPPM,
-    inputs = [
-        gr.Timeseries(label="Input for the timeseries", max_rows=1, interactive=False),
-        gr.Slider(1, 100, value=75, step=1, label="Train test split percentage"),
-    ],
-    outputs= [gr.LinePlot(x='Date', y='PPM', color='origin')#gr.Timeseries(x='Month')
-
-    ]
-)
+from datetime import datetime
+import numpy as np
+import pandas as pd
+from sklearn.ensemble import RandomForestRegressor
+import gradio as gr
+import os
+import plotly.graph_objects as go
+from huggingface_hub import from_pretrained_keras
+def predictPPM(df, split):
+    ts= pd.read_csv('datappm.csv')
+    df2 =ts.copy()
+    ttSplit=split/100
+    ts['Date']=pd.to_datetime(ts['Date'])
+    ts.rename(columns={'#PPM':'PPM'},inplace=True)
+    ts=ts.set_index(['Date'])
+    ts['months'] = [x.month for x in ts.index]
+    ts['years'] = [x.year for x in ts.index]
+    ts.reset_index(drop=True, inplace=True)
+
+    # Split Data
+    X=ts.drop("PPM",axis=1)
+    Y= ts["PPM"]
+    X_train=X[:int (len(Y)*ttSplit)] 
+    X_test=X[int(len(Y)*ttSplit):]
+    Y_train=Y[:int (len(Y)*ttSplit)] 
+    Y_test=Y[int(len(Y)*ttSplit):]
+
+    # fit the model
+    rf = RandomForestRegressor()
+    rf.fit(X_train, Y_train)
+    
+    df1=df2.set_index(['Date'])
+    df1.rename(columns={'#PPM':'PPM'},inplace=True)
+    train=df1.PPM[:int (len(ts.PPM)*ttSplit)]
+    test=df1.PPM[int(len(ts.PPM)*ttSplit):]
+    preds=rf.predict(X_test).astype(int) 
+    predictions=pd.DataFrame(preds,columns=['PPM'])
+    predictions.index=test.index
+    predictions.reset_index(inplace=True)
+    predictions['Date']=pd.to_datetime(predictions['Date'])
+    print(predictions)
+    
+    #combine all into one table
+    ts_df=df.copy()
+    ts_df.rename(columns={'#PPM':'PPM'},inplace=True)
+    train= ts_df[:int (len(ts_df)*ttSplit)]
+    test= ts_df[int(len(ts_df)*ttSplit):] 
+
+    df2['Date']=pd.to_datetime(df2['Date'])
+    df2.rename(columns={'#PPM':'PPM'},inplace=True)
+    df3=predictions
+    df2['origin']='ground truth'
+    df3['origin']='prediction'
+    df4=pd.concat([df2, df3])
+    print(df4)
+    return df4
+
+demo=gr.Interface(
+    fn = predictPPM,
+    inputs=[
+        gr.Timeseries(label="Input for the timeseries", max_rows=1, interactive=False),
+        gr.Slider(1, 100, value=75, step=1, label="Train test split percentage"),
+    ],
+    outputs=[gr.LinePlot(x='Date', y='PPM', color='origin')#gr.Timeseries(x='Month')
+
+    ]
+)
 demo.launch()