Upload 6 files

app.py

@@ -0,0 +1,25 @@
+import gradio as gr
+import numpy as np
+from tensorflow.keras.models import load_model
+from tkan import TKAN
+from tkat import TKAT
+from keras.utils import custom_object_scope
+
+# Load the model with custom objects
+with custom_object_scope({"TKAN": TKAN, "TKAT": TKAT}):
+    model = load_model("best_model_TKAN_nahead_1.h5")
+
+# Define predict function
+def predict(pm25, pm10, co, temp):
+    input_data = np.array([[pm25, pm10, co, temp]])
+    output = model.predict(input_data)
+    return float(output[0][0])
+
+# Gradio interface
+interface = gr.Interface(
+    fn=predict,
+    inputs=[gr.Number(), gr.Number(), gr.Number(), gr.Number()],
+    outputs=gr.Number()
+)
+
+interface.launch()

best_model_TKAN_nahead_1.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dbbb02d8bfc51e23d05291151a330448efc27086afb59b10a54060714b7abbb5
+size 856432

requirements.txt

@@ -0,0 +1,5 @@
+gradio
+tensorflow==2.15.0
+numpy==1.26.4
+pandas==2.1.3
+matplotlib==3.9.2

tkan.py

@@ -0,0 +1,18 @@
+import tensorflow as tf
+from tensorflow.keras import layers
+
+class TKAN(tf.keras.layers.Layer):
+    def __init__(self, units, **kwargs):
+        super(TKAN, self).__init__(**kwargs)
+        self.units = units
+        self.dense1 = layers.Dense(units, activation="relu")
+        self.dense2 = layers.Dense(units, activation="relu")
+
+    def call(self, inputs):
+        x = self.dense1(inputs)
+        return self.dense2(x)
+
+    def get_config(self):
+        config = super(TKAN, self).get_config()
+        config.update({"units": self.units})
+        return config

tkat.py

@@ -0,0 +1,16 @@
+import tensorflow as tf
+from tensorflow.keras.layers import Layer, Dense
+
+class AttentionLayer(Layer):
+    def __init__(self, units):
+        super(AttentionLayer, self).__init__()
+        self.W1 = Dense(units)
+        self.W2 = Dense(units)
+        self.V = Dense(1)
+
+    def call(self, query, values):
+        score = self.V(tf.nn.tanh(self.W1(query) + self.W2(values)))
+        attention_weights = tf.nn.softmax(score, axis=1)
+        context_vector = attention_weights * values
+        context_vector = tf.reduce_sum(context_vector, axis=1)
+        return context_vector, attention_weights