odify genome for the official release

examples/interpret_visualize/genome_sympy.py

@@ -0,0 +1,37 @@
+import jax, jax.numpy as jnp
+
+from algorithm.neat import *
+from algorithm.neat.genome.dense import DenseInitialize
+from utils.graph import topological_sort_python
+from tensorneat.common import *
+
+if __name__ == "__main__":
+    genome = DenseInitialize(
+        num_inputs=3,
+        num_outputs=1,
+        max_nodes=50,
+        max_conns=500,
+    )
+
+    state = genome.setup()
+
+    randkey = jax.random.PRNGKey(42)
+    nodes, conns = genome.initialize(state, randkey)
+
+    network = genome.network_dict(state, nodes, conns)
+
+    input_idx, output_idx = genome.get_input_idx(), genome.get_output_idx()
+
+    res = genome.sympy_func(state, network, sympy_input_transform=lambda x: 999999999*x, sympy_output_transform=SympyStandardSigmoid)
+    (symbols,
+    args_symbols,
+    input_symbols,
+    nodes_exprs,
+    output_exprs,
+    forward_func,) = res
+
+    print(symbols)
+    print(output_exprs[0].subs(args_symbols))
+
+    inputs = jnp.zeros(3)
+    print(forward_func(inputs))