Finished training and deploying

combined_app.py

@@ -27,7 +27,7 @@
 st.pyplot(fig)
 
 
-st.subheader('Closing Price vs Time Chart (MOVING AVERAGE 100 (R) AND 200 (G))')
+st.subheader('MOVING AVERAGE 100 (R) AND 200 (G)')
 
 ma100 = df.Close.rolling(100).mean()
 ma200 = df.Close.rolling(200).mean()
@@ -42,97 +42,44 @@
 testing = pd.DataFrame(df['Close'][int(len(df) * 0.7):int(len(df))])
 
 scaler = MinMaxScaler(feature_range=(0,1))
-
 training_array = scaler.fit_transform(training)
 
-# Splitting training data into x_train and y_train
-
-# x_train = []
-# y_train = []
+model = load_model('/Users/aaditroychowdhury/Documents/CS 222/Main branch/main-project-shmoney/keras_model.h5')
 
-# for i in range(100, training_array.shape[0]):
-
-#     x_train.append(training_array[i - 100 : i])
-#     y_train.append(training_array[i, 0])
-
-# x_train, y_train = np.array(x_train), np.array(y_train)
 
-filename = 'random_forest_model.joblib'
-model = joblib.load(filename)
+past_100_days = training.tail(100)
+final_df = pd.concat([past_100_days, testing], ignore_index=True)
+input_data = scaler.fit_transform(final_df)
 
-# Ensure 'Tomorrow' column is created by shifting 'Close' by -1
-df["Tomorrow"] = df["Close"].shift(-1)
 
-# Check if 'Tomorrow' column was successfully created
-if "Tomorrow" in df.columns:
-    # Drop rows where 'Tomorrow' has NaN values (last row will have NaN after shift)
-    df = df.dropna(subset=["Tomorrow"])
+x_test = []
+y_test = []
 
-    # Now safely compare 'Tomorrow' with 'Close'
-    df["Target"] = (df["Tomorrow"] > df["Close"]).astype(int)
-else:
-    st.error("The 'Tomorrow' column was not created.")
+for i in range(100, input_data.shape[0]):
 
-
-
-if "Target" in df.columns:
-    predictions = model.predict(df)
-
-    st.subheader("Predictions vs Original")
-    fig2 = plt.figure(figsize=(12, 6))
-    plt.plot(df["Target"], 'b', label='Original Price')
-    plt.plot(predictions, 'r', label='Predicted Price')
-    plt.legend()
-    st.pyplot(fig2)
-else:
-    st.error("The 'Target' column was not created. Unable to make predictions.")
+    x_test.append(input_data[i - 100 : i])
+    y_test.append(input_data[i, 0])
+
 
-
-if "Target" in df.columns:
-    # Exclude 'Tomorrow' and 'Target' columns for prediction
-    features = df.drop(["Tomorrow", "Target"], axis=1)
-    predictions = model.predict(features)
-    # ... rest of the plotting code ...
-    st.subheader("Predictions vs Original")
-    fig2 = plt.figure(figsize= (12, 6))
-    plt.plot(df["Target"], 'b', label='Original Price')
-    plt.plot(predictions, 'r', label='Predicted Price')
-
-    st.pyplot(fig2)
-
-st.write("DataFrame columns:", df.columns)
-
-# past_100_days = training.tail(100)
-# final_df = pd.concat([past_100_days, testing], ignore_index=True)
-# input_data = scaler.fit_transform(final_df)
+x_test, y_test = np.array(x_test), np.array(y_test)
 
 
-# x_test = []
-# y_test = []
-
-# for i in range(100, input_data.shape[0]):
-
-#     x_test.append(input_data[i - 100 : i])
-#     y_test.append(input_data[i, 0])
-
-# x_test, y_test = np.array(y_test), np.array(y_test)
-# x_test = x_test.reshape(1, -1)
-# y_test = y_test.reshape(1, -1)
-# y_predicted = model.predict(x_test)
-# scaler = scaler.scale_
+# Now you can make predictions
+y_predicted = model.predict(x_test)
+scaler = scaler.scale_
 
-# scale_factor = 1 / scaler[0]
-# y_predicted = y_predicted * scale_factor
-# y_test = y_test * scale_factor
+scale_factor = 1 / scaler[0]
+y_predicted = y_predicted * scale_factor
+y_test = y_test * scale_factor
 
 # final figure
 
-# st.subheader("Predictions vs Original")
-# fig2 = plt.figure(figsize= (12, 6))
-# plt.plot(y_test, 'b', label='Original Price')
-# plt.plot(y_predicted, 'r', label='Predicted Price')
+st.subheader("Predictions vs Original")
+fig2 = plt.figure(figsize= (12, 6))
+plt.plot(y_test, 'b', label='Original Price')
+plt.plot(y_predicted, 'r', label='Predicted Price')
 
-# plt.xlabel('Time')
-# plt.ylabel('Price')
-# plt.legend()
-# st.pyplot(fig2)
+plt.xlabel('Time')
+plt.ylabel('Price')
+plt.legend()
+st.pyplot(fig2)