AI model

app.py

@@ -0,0 +1,42 @@
+import gradio as gr
+import numpy as np
+from tensorflow import keras
+
+# Load the pre-trained model
+model = keras.models.load_model('omni_rnn_0.h5')
+
+def predict(year, proton_density, temperature):
+    try:
+        # Ensure the input data has the correct types and ranges
+        year = int(year)
+        proton_density = float(proton_density)
+        temperature = float(temperature)
+
+        # Ensure the input data is within valid ranges
+        if year < 0 or proton_density < 0 or temperature < 0:
+            return {"error": "Input values should be non-negative"}
+
+        # Convert input data to a NumPy array
+        input_data = np.array([year, proton_density, temperature]).reshape(1, -1)
+
+        # Make predictions using the loaded model
+        predictions = model.predict(input_data)
+
+        # Format the predictions as a dictionary
+        result = {
+            "predicted_speed": float(predictions[0, 0]),
+            "predicted_field_magnitude": float(predictions[0, 1])
+        }
+
+        return result
+
+    except Exception as e:
+        return {"error": str(e)}
+
+iface = gr.Interface(
+    fn=predict,
+    inputs=["text", "text", "text"],  # Year, Proton Density, Temperature
+    outputs="json"
+)
+
+iface.launch()

omni_rnn_0.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48024055b87eaac2c44ed186d9b21cf8d7acf6bce6b7fb81ea9e4f5f7b02570d
+size 154352

requirements.txt

@@ -0,0 +1,3 @@
+flask 
+flask-cors
+tensorflow