Upload 3 files

DBS_SingDollar.csv

@@ -0,0 +1,123 @@
+,Date,DBS,SGD
+1,6/20/2017,20.45,1.38925
+2,6/19/2017,20.6,1.38707
+3,6/16/2017,20.45,1.38324
+4,6/15/2017,20.41,1.38378
+5,6/14/2017,20.59,1.37595
+6,6/13/2017,20.78,1.38172
+7,6/12/2017,20.59,1.38424
+8,6/9/2017,20.72,1.38401
+9,6/8/2017,20.47,1.3827
+10,6/7/2017,20.43,1.38187
+11,6/6/2017,20.49,1.37925
+12,6/5/2017,20.49,1.38157
+13,6/2/2017,20.48,1.38104
+14,6/1/2017,20.65,1.38653
+15,5/31/2017,20.47,1.3832
+16,5/30/2017,20.6,1.3853
+17,5/29/2017,20.84,1.38665
+18,5/26/2017,20.8,1.38071
+19,5/25/2017,21.04,1.38634
+20,5/24/2017,21,1.38451
+21,5/23/2017,21,1.38995
+22,5/22/2017,20.86,1.38642
+23,5/19/2017,20.7,1.38523
+24,5/18/2017,20.63,1.39297
+25,5/17/2017,20.8,1.38981
+26,5/16/2017,20.65,1.39474
+27,5/15/2017,20.77,1.39749
+28,5/12/2017,20.68,1.40394
+29,5/11/2017,20.93,1.40746
+30,5/9/2017,20.5,1.41128
+31,5/8/2017,20.3,1.40604
+32,5/5/2017,20.42,1.40496
+33,5/4/2017,20.45,1.40069
+34,5/3/2017,20.53,1.39724
+35,5/2/2017,19.57,1.39379
+36,4/28/2017,19.07,1.39739
+37,4/27/2017,18.93,1.3971
+38,4/26/2017,19.03,1.39598
+39,4/25/2017,18.92,1.39412
+40,4/24/2017,18.73,1.39261
+41,4/21/2017,18.5,1.39655
+42,4/20/2017,18.64,1.39772
+43,4/19/2017,18.52,1.39821
+44,4/18/2017,18.65,1.39584
+45,4/17/2017,18.63,1.39794
+46,4/13/2017,18.82,1.39799
+47,4/12/2017,18.97,1.39653
+48,4/11/2017,18.88,1.40404
+49,4/10/2017,18.85,1.4048
+50,4/7/2017,18.78,1.40474
+51,4/6/2017,18.85,1.40211
+52,4/5/2017,18.93,1.40067
+53,4/4/2017,18.95,1.39817
+54,4/3/2017,19.12,1.3971
+55,3/31/2017,19.12,1.39639
+56,3/30/2017,19.02,1.39769
+57,3/29/2017,19.01,1.39407
+58,3/28/2017,18.75,1.39677
+59,3/27/2017,18.43,1.39397
+60,3/24/2017,18.52,1.39885
+61,3/23/2017,18.38,1.3997
+62,3/22/2017,18.4,1.39854
+63,3/21/2017,18.73,1.39934
+64,3/20/2017,18.74,1.39688
+65,3/17/2017,18.86,1.40201
+66,3/16/2017,18.84,1.40121
+67,3/15/2017,18.85,1.40224
+68,3/14/2017,18.91,1.41592
+69,3/13/2017,18.83,1.41477
+70,3/10/2017,18.66,1.41173
+71,3/9/2017,18.67,1.42132
+72,3/8/2017,19,1.41763
+73,3/7/2017,18.82,1.41135
+74,3/6/2017,18.79,1.41235
+75,3/3/2017,18.74,1.41037
+76,3/2/2017,18.76,1.41213
+77,3/1/2017,18.55,1.40907
+78,2/28/2017,18.46,1.405
+79,2/27/2017,18.54,1.40429
+80,2/24/2017,18.6,1.40447
+81,2/23/2017,18.62,1.40763
+82,2/22/2017,18.48,1.41323
+83,2/21/2017,18.19,1.42045
+84,2/20/2017,18.23,1.41793
+85,2/17/2017,18.33,1.41901
+86,2/16/2017,18.27,1.41659
+87,2/15/2017,17.97,1.41906
+88,2/14/2017,18,1.41995
+89,2/13/2017,18.64,1.42416
+90,2/10/2017,18.7,1.42058
+91,2/9/2017,18.55,1.42066
+92,2/8/2017,18.53,1.41667
+93,2/7/2017,18.59,1.41857
+94,2/6/2017,18.66,1.40917
+95,2/3/2017,18.39,1.40746
+96,2/2/2017,18.51,1.41138
+97,2/1/2017,18.61,1.41244
+98,1/31/2017,18.7,1.40954
+99,1/27/2017,18.89,1.43178
+100,1/26/2017,18.76,1.42589
+101,1/25/2017,18.58,1.41285
+102,1/24/2017,18.39,1.42117
+103,1/23/2017,18.08,1.41598
+104,1/20/2017,18.01,1.42705
+105,1/19/2017,18.02,1.42616
+106,1/18/2017,17.94,1.42876
+107,1/17/2017,18.09,1.41601
+108,1/16/2017,18.06,1.4303
+109,1/13/2017,18.05,1.42592
+110,1/12/2017,17.86,1.42728
+111,1/11/2017,17.78,1.42678
+112,1/10/2017,17.73,1.43654
+113,1/9/2017,17.59,1.43529
+114,1/6/2017,17.42,1.43938
+115,1/5/2017,17.54,1.42847
+116,1/4/2017,17.27,1.43942
+117,1/3/2017,17.07,1.44992
+118,12/30/2016,17.09,1.44655
+119,12/29/2016,17.13,1.44814
+120,12/28/2016,17.22,1.45018
+121,12/27/2016,17.19,1.44954
+122,12/23/2016,17.14,1.44766

app.py

@@ -0,0 +1,69 @@
+# -*- coding: utf-8 -*-
+"""2.7 (Optional DBS).ipynb
+
+Automatically generated by Colab.
+
+Original file is located at
+    https://colab.research.google.com/drive/1Js2DNDDuusJx06SBvdh-lmKbuCBTBw9B
+
+chatgpt prompt:
+Please create a regression model to predict DBS base on SGD exchange rate, create the model and then use gradio for the interface
+"""
+
+# Install dependencies (uncomment if needed)
+# !pip install pandas scikit-learn gradio
+
+import pandas as pd
+from sklearn.model_selection import train_test_split
+from sklearn.linear_model import LinearRegression
+from sklearn.metrics import mean_squared_error, r2_score
+import gradio as gr
+
+# 1. LOAD DATA
+# Make sure the CSV file is in the same folder, or give full path
+df = pd.read_csv("DBS_SingDollar.csv")
+
+# If your header row is exactly: Date,DBS,SGD this will work directly
+# Keep only needed columns and drop missing values
+df = df[["DBS", "SGD"]].dropna()
+
+X = df[["SGD"]]   # feature
+y = df["DBS"]     # target
+
+# 2. TRAIN / TEST SPLIT
+X_train, X_test, y_train, y_test = train_test_split(
+    X, y, test_size=0.2, random_state=42
+)
+
+# 3. TRAIN MODEL (Linear Regression)
+model = LinearRegression()
+model.fit(X_train, y_train)
+
+# 4. EVALUATE (optional, shows in console)
+y_pred = model.predict(X_test)
+rmse = mean_squared_error(y_test, y_pred) ** 0.5
+r2 = r2_score(y_test, y_pred)
+
+print(f"RMSE: {rmse:.4f}")
+print(f"R²:   {r2:.4f}")
+
+# 5. PREDICTION FUNCTION FOR GRADIO
+def predict_dbs_price(sgd_rate: float) -> str:
+    """
+    Input: SGD exchange rate (float)
+    Output: predicted DBS price (string for display)
+    """
+    pred = model.predict([[sgd_rate]])[0]
+    return f"Predicted DBS price: {pred:.2f}"
+
+# 6. GRADIO INTERFACE
+iface = gr.Interface(
+    fn=predict_dbs_price,
+    inputs=gr.Number(label="SGD Exchange Rate"),
+    outputs=gr.Textbox(label="Predicted DBS Price"),
+    title="DBS Price Predictor",
+    description="Enter the SGD exchange rate to predict the DBS share price (trained with historical DBS & SGD data)."
+)
+
+iface.launch()
+

currency_gradio_app.py

@@ -0,0 +1,101 @@
+
+import pandas as pd
+import joblib
+import gradio as gr
+
+# 1. Load & train the model (you can also pre-train and just load a .joblib file)
+df = pd.read_csv("DBS_SingDollar.csv")
+
+# Parse dates and sort just in case
+df["Date"] = pd.to_datetime(df["Date"])
+df = df.sort_values("Date")
+
+# Create simple supervised features:
+# - Use today's SGD and DBS price to predict tomorrow's SGD rate
+df["SGD_today"] = df["SGD"]
+df["DBS_today"] = df["DBS"]
+df["SGD_next"] = df["SGD"].shift(-1)
+
+data = df.dropna(subset=["SGD_next"]).copy()
+
+X = data[["SGD_today", "DBS_today"]]
+y = data["SGD_next"]
+
+from sklearn.ensemble import RandomForestRegressor
+from sklearn.model_selection import train_test_split
+from sklearn.metrics import mean_absolute_error, r2_score
+
+X_train, X_test, y_train, y_test = train_test_split(
+    X, y, test_size=0.2, shuffle=False
+)
+
+model = RandomForestRegressor(
+    n_estimators=300,
+    random_state=42,
+    min_samples_leaf=2,
+)
+
+model.fit(X_train, y_train)
+
+# Evaluate (printed in console)
+y_pred = model.predict(X_test)
+mae = mean_absolute_error(y_test, y_pred)
+r2 = r2_score(y_test, y_pred)
+print(f"Test MAE: {mae:.6f}")
+print(f"Test R^2: {r2:.4f}")
+
+# 2. Define Gradio prediction function
+
+def predict_next_sgd(current_sgd, current_dbs):
+    """Predict tomorrow's SGD rate from today's SGD + DBS price."""
+    import numpy as np
+
+    X_input = np.array([[current_sgd, current_dbs]])
+    next_sgd = model.predict(X_input)[0]
+
+    change = next_sgd - current_sgd
+    pct_change = (change / current_sgd) * 100 if current_sgd != 0 else 0.0
+    direction = "up" if change > 0 else "down or flat"
+
+    return (
+        f"Predicted next-day SGD rate: {next_sgd:.5f}\n"
+        f"Expected move: {direction} ({change:+.5f}, {pct_change:+.3f}%)"
+    )
+
+# 3. Build Gradio UI
+
+with gr.Blocks() as demo:
+    gr.Markdown(
+        """
+        # SGD Currency Movement Predictor
+
+        This demo model is trained on your DBS / SGD dataset.
+
+        Enter today's SGD rate and DBS price to predict the next day's SGD rate.
+        (Note: this is a simple ML model for experimentation/learning, not trading advice.)
+        """
+    )
+
+    with gr.Row():
+        current_sgd = gr.Number(
+            label="Today's SGD rate",
+            value=float(X.iloc[-1]["SGD_today"]),
+            precision=5,
+        )
+        current_dbs = gr.Number(
+            label="Today's DBS price",
+            value=float(X.iloc[-1]["DBS_today"]),
+            precision=2,
+        )
+
+    predict_button = gr.Button("Predict next-day SGD")
+    output = gr.Textbox(label="Prediction", lines=3)
+
+    predict_button.click(
+        fn=predict_next_sgd,
+        inputs=[current_sgd, current_dbs],
+        outputs=output,
+    )
+
+if __name__ == "__main__":
+    demo.launch()