Upload 3 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tft_model.keras filter=lfs diff=lfs merge=lfs -text

app.py

@@ -0,0 +1,26 @@
+import gradio as gr
+import pandas as pd
+import numpy as np
+import tensorflow as tf
+from sklearn.preprocessing import StandardScaler
+
+# Load your trained TFT model
+model = tf.keras.models.load_model("tft_model.keras", compile=False)
+
+# Load scalers if saved separately (optional), or define here again
+
+def predict_from_csv(file):
+    df = pd.read_csv(file.name)
+    
+    # Perform the same preprocessing as during training
+    # This must match what you did before model.fit()
+
+    # For demo, let's assume the last N rows have the correct shape
+    input_data = np.expand_dims(df.tail(1).values, axis=0)
+    
+    # Make prediction
+    pred = model.predict(input_data)
+    return f"Prediction: {pred.flatten()[0]}"
+
+# Gradio interface
+gr.Interface(fn=predict_from_csv, inputs="file", outputs="text").launch()

requirements.txt

@@ -0,0 +1,4 @@
+tensorflow
+gradio
+numpy
+pandas

tft_model.keras

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8c855acdbcb6c1771a788bba2a9f398a1ee31deaccf0443555a4e0521ddae054
+size 3704537