diff --git a/paddle_samples/PaddleX/STFPM/graph_hash.txt b/paddle_samples/PaddleX/STFPM/graph_hash.txt
new file mode 100644
index 000000000..8008325f5
--- /dev/null
+++ b/paddle_samples/PaddleX/STFPM/graph_hash.txt
@@ -0,0 +1 @@
+d0f1ca7a396f08edcce5424b609a6a9794e602f78374ed87e6951c1d4dc4fe92
\ No newline at end of file
diff --git a/paddle_samples/PaddleX/STFPM/graph_net.json b/paddle_samples/PaddleX/STFPM/graph_net.json
new file mode 100644
index 000000000..599a5a566
--- /dev/null
+++ b/paddle_samples/PaddleX/STFPM/graph_net.json
@@ -0,0 +1,6 @@
+{
+    "framework": "paddle",
+    "model_name": "STFPM",
+    "num_devices_required": 1,
+    "num_nodes_required": 1
+}
\ No newline at end of file
diff --git a/paddle_samples/PaddleX/STFPM/input_meta.py b/paddle_samples/PaddleX/STFPM/input_meta.py
new file mode 100644
index 000000000..9e78dfcbd
--- /dev/null
+++ b/paddle_samples/PaddleX/STFPM/input_meta.py
@@ -0,0 +1,9 @@
+class Program_weight_tensor_data_0:
+    name = "data_0"
+    shape = [1, 3, 256, 256]
+    dtype = "float32"
+    min_val = float("-1.54422")
+    max_val = float("1.21516")
+    mean = float("-0.201062")
+    std = float("0.553945")
+    data = None
diff --git a/paddle_samples/PaddleX/STFPM/model.py b/paddle_samples/PaddleX/STFPM/model.py
new file mode 100644
index 000000000..476792de1
--- /dev/null
+++ b/paddle_samples/PaddleX/STFPM/model.py
@@ -0,0 +1,911 @@
+import paddle
+
+
+class GraphModule(paddle.nn.Layer):
+    def __init__(self):
+        super().__init__()
+
+    def forward(
+        self,
+        parameter_0,
+        parameter_1,
+        parameter_2,
+        parameter_3,
+        parameter_4,
+        parameter_5,
+        parameter_6,
+        parameter_7,
+        parameter_8,
+        parameter_9,
+        parameter_10,
+        parameter_11,
+        parameter_12,
+        parameter_13,
+        parameter_14,
+        parameter_15,
+        parameter_16,
+        parameter_17,
+        parameter_18,
+        parameter_19,
+        parameter_20,
+        parameter_21,
+        parameter_22,
+        parameter_23,
+        parameter_24,
+        parameter_25,
+        parameter_26,
+        parameter_27,
+        parameter_28,
+        parameter_29,
+        parameter_30,
+        parameter_31,
+        parameter_32,
+        parameter_33,
+        parameter_34,
+        parameter_35,
+        parameter_36,
+        parameter_37,
+        parameter_38,
+        parameter_39,
+        parameter_40,
+        parameter_41,
+        parameter_42,
+        parameter_43,
+        parameter_44,
+        parameter_45,
+        parameter_46,
+        parameter_47,
+        parameter_48,
+        parameter_49,
+        parameter_50,
+        parameter_51,
+        parameter_52,
+        parameter_53,
+        parameter_54,
+        parameter_55,
+        parameter_56,
+        parameter_57,
+        parameter_58,
+        parameter_59,
+        parameter_60,
+        parameter_61,
+        parameter_62,
+        parameter_63,
+        parameter_64,
+        parameter_65,
+        parameter_66,
+        parameter_67,
+        parameter_68,
+        parameter_69,
+        parameter_70,
+        parameter_71,
+        parameter_72,
+        parameter_73,
+        parameter_74,
+        parameter_75,
+        parameter_76,
+        parameter_77,
+        parameter_78,
+        parameter_79,
+        parameter_80,
+        parameter_81,
+        parameter_82,
+        parameter_83,
+        parameter_84,
+        parameter_85,
+        parameter_86,
+        parameter_87,
+        parameter_88,
+        parameter_89,
+        parameter_90,
+        parameter_91,
+        parameter_92,
+        parameter_93,
+        parameter_94,
+        parameter_95,
+        parameter_96,
+        parameter_97,
+        parameter_98,
+        parameter_99,
+        data_0,
+    ):
+        # pd_op.conv2d: (1x64x128x128xf32) <- (1x3x256x256xf32, 64x3x7x7xf32)
+        conv2d_0 = paddle._C_ops.conv2d(
+            data_0, parameter_99, [2, 2], [3, 3], "EXPLICIT", [1, 1], 1, "NCHW"
+        )
+        del data_0, parameter_99
+
+        # pd_op.batch_norm_: (1x64x128x128xf32, 64xf32, 64xf32, 64xf32, 64xf32, -1xui8) <- (1x64x128x128xf32, 64xf32, 64xf32, 64xf32, 64xf32)
+        (
+            batch_norm__0,
+            batch_norm__1,
+            batch_norm__2,
+            batch_norm__3,
+            batch_norm__4,
+            batch_norm__5,
+        ) = (lambda x, f: f(x))(
+            paddle._C_ops.batch_norm(
+                conv2d_0,
+                parameter_98,
+                parameter_97,
+                parameter_96,
+                parameter_95,
+                True,
+                float("0.9"),
+                float("1e-05"),
+                "NCHW",
+                True,
+                False,
+            ),
+            lambda out: out
+            if isinstance(out, (list, tuple))
+            else (out, None, None, None, None, None),
+        )
+        del conv2d_0, parameter_95, parameter_96, parameter_97, parameter_98
+
+        # pd_op.relu: (1x64x128x128xf32) <- (1x64x128x128xf32)
+        relu_1 = paddle._C_ops.relu(batch_norm__0)
+        del batch_norm__0
+
+        # pd_op.full_int_array: (2xi64) <- ()
+        full_int_array_0 = [3, 3]
+
+        # pd_op.pool2d: (1x64x64x64xf32) <- (1x64x128x128xf32, 2xi64)
+        pool2d_0 = paddle._C_ops.pool2d(
+            relu_1,
+            full_int_array_0,
+            [2, 2],
+            [1, 1],
+            False,
+            True,
+            "NCHW",
+            "max",
+            False,
+            False,
+            "EXPLICIT",
+        )
+        del full_int_array_0, relu_1
+
+        # pd_op.conv2d: (1x64x64x64xf32) <- (1x64x64x64xf32, 64x64x3x3xf32)
+        conv2d_1 = paddle._C_ops.conv2d(
+            pool2d_0, parameter_94, [1, 1], [1, 1], "EXPLICIT", [1, 1], 1, "NCHW"
+        )
+        del parameter_94
+
+        # pd_op.batch_norm_: (1x64x64x64xf32, 64xf32, 64xf32, 64xf32, 64xf32, -1xui8) <- (1x64x64x64xf32, 64xf32, 64xf32, 64xf32, 64xf32)
+        (
+            batch_norm__6,
+            batch_norm__7,
+            batch_norm__8,
+            batch_norm__9,
+            batch_norm__10,
+            batch_norm__11,
+        ) = (lambda x, f: f(x))(
+            paddle._C_ops.batch_norm(
+                conv2d_1,
+                parameter_93,
+                parameter_92,
+                parameter_91,
+                parameter_90,
+                True,
+                float("0.9"),
+                float("1e-05"),
+                "NCHW",
+                True,
+                False,
+            ),
+            lambda out: out
+            if isinstance(out, (list, tuple))
+            else (out, None, None, None, None, None),
+        )
+        del conv2d_1, parameter_90, parameter_91, parameter_92, parameter_93
+
+        # pd_op.relu: (1x64x64x64xf32) <- (1x64x64x64xf32)
+        relu_2 = paddle._C_ops.relu(batch_norm__6)
+        del batch_norm__6
+
+        # pd_op.conv2d: (1x64x64x64xf32) <- (1x64x64x64xf32, 64x64x3x3xf32)
+        conv2d_2 = paddle._C_ops.conv2d(
+            relu_2, parameter_89, [1, 1], [1, 1], "EXPLICIT", [1, 1], 1, "NCHW"
+        )
+        del parameter_89, relu_2
+
+        # pd_op.batch_norm_: (1x64x64x64xf32, 64xf32, 64xf32, 64xf32, 64xf32, -1xui8) <- (1x64x64x64xf32, 64xf32, 64xf32, 64xf32, 64xf32)
+        (
+            batch_norm__12,
+            batch_norm__13,
+            batch_norm__14,
+            batch_norm__15,
+            batch_norm__16,
+            batch_norm__17,
+        ) = (lambda x, f: f(x))(
+            paddle._C_ops.batch_norm(
+                conv2d_2,
+                parameter_88,
+                parameter_87,
+                parameter_86,
+                parameter_85,
+                True,
+                float("0.9"),
+                float("1e-05"),
+                "NCHW",
+                True,
+                False,
+            ),
+            lambda out: out
+            if isinstance(out, (list, tuple))
+            else (out, None, None, None, None, None),
+        )
+        del conv2d_2, parameter_85, parameter_86, parameter_87, parameter_88
+
+        # pd_op.add: (1x64x64x64xf32) <- (1x64x64x64xf32, 1x64x64x64xf32)
+        add_0 = paddle._C_ops.add(batch_norm__12, pool2d_0)
+        del batch_norm__12, pool2d_0
+
+        # pd_op.relu: (1x64x64x64xf32) <- (1x64x64x64xf32)
+        relu_3 = paddle._C_ops.relu(add_0)
+        del add_0
+
+        # pd_op.conv2d: (1x64x64x64xf32) <- (1x64x64x64xf32, 64x64x3x3xf32)
+        conv2d_3 = paddle._C_ops.conv2d(
+            relu_3, parameter_84, [1, 1], [1, 1], "EXPLICIT", [1, 1], 1, "NCHW"
+        )
+        del parameter_84
+
+        # pd_op.batch_norm_: (1x64x64x64xf32, 64xf32, 64xf32, 64xf32, 64xf32, -1xui8) <- (1x64x64x64xf32, 64xf32, 64xf32, 64xf32, 64xf32)
+        (
+            batch_norm__18,
+            batch_norm__19,
+            batch_norm__20,
+            batch_norm__21,
+            batch_norm__22,
+            batch_norm__23,
+        ) = (lambda x, f: f(x))(
+            paddle._C_ops.batch_norm(
+                conv2d_3,
+                parameter_83,
+                parameter_82,
+                parameter_81,
+                parameter_80,
+                True,
+                float("0.9"),
+                float("1e-05"),
+                "NCHW",
+                True,
+                False,
+            ),
+            lambda out: out
+            if isinstance(out, (list, tuple))
+            else (out, None, None, None, None, None),
+        )
+        del conv2d_3, parameter_80, parameter_81, parameter_82, parameter_83
+
+        # pd_op.relu: (1x64x64x64xf32) <- (1x64x64x64xf32)
+        relu_4 = paddle._C_ops.relu(batch_norm__18)
+        del batch_norm__18
+
+        # pd_op.conv2d: (1x64x64x64xf32) <- (1x64x64x64xf32, 64x64x3x3xf32)
+        conv2d_4 = paddle._C_ops.conv2d(
+            relu_4, parameter_79, [1, 1], [1, 1], "EXPLICIT", [1, 1], 1, "NCHW"
+        )
+        del parameter_79, relu_4
+
+        # pd_op.batch_norm_: (1x64x64x64xf32, 64xf32, 64xf32, 64xf32, 64xf32, -1xui8) <- (1x64x64x64xf32, 64xf32, 64xf32, 64xf32, 64xf32)
+        (
+            batch_norm__24,
+            batch_norm__25,
+            batch_norm__26,
+            batch_norm__27,
+            batch_norm__28,
+            batch_norm__29,
+        ) = (lambda x, f: f(x))(
+            paddle._C_ops.batch_norm(
+                conv2d_4,
+                parameter_78,
+                parameter_77,
+                parameter_76,
+                parameter_75,
+                True,
+                float("0.9"),
+                float("1e-05"),
+                "NCHW",
+                True,
+                False,
+            ),
+            lambda out: out
+            if isinstance(out, (list, tuple))
+            else (out, None, None, None, None, None),
+        )
+        del conv2d_4, parameter_75, parameter_76, parameter_77, parameter_78
+
+        # pd_op.add: (1x64x64x64xf32) <- (1x64x64x64xf32, 1x64x64x64xf32)
+        add_1 = paddle._C_ops.add(batch_norm__24, relu_3)
+        del batch_norm__24, relu_3
+
+        # pd_op.relu: (1x64x64x64xf32) <- (1x64x64x64xf32)
+        relu_5 = paddle._C_ops.relu(add_1)
+        del add_1
+
+        # pd_op.conv2d: (1x128x32x32xf32) <- (1x64x64x64xf32, 128x64x3x3xf32)
+        conv2d_5 = paddle._C_ops.conv2d(
+            relu_5, parameter_74, [2, 2], [1, 1], "EXPLICIT", [1, 1], 1, "NCHW"
+        )
+        del parameter_74
+
+        # pd_op.batch_norm_: (1x128x32x32xf32, 128xf32, 128xf32, 128xf32, 128xf32, -1xui8) <- (1x128x32x32xf32, 128xf32, 128xf32, 128xf32, 128xf32)
+        (
+            batch_norm__30,
+            batch_norm__31,
+            batch_norm__32,
+            batch_norm__33,
+            batch_norm__34,
+            batch_norm__35,
+        ) = (lambda x, f: f(x))(
+            paddle._C_ops.batch_norm(
+                conv2d_5,
+                parameter_73,
+                parameter_72,
+                parameter_71,
+                parameter_70,
+                True,
+                float("0.9"),
+                float("1e-05"),
+                "NCHW",
+                True,
+                False,
+            ),
+            lambda out: out
+            if isinstance(out, (list, tuple))
+            else (out, None, None, None, None, None),
+        )
+        del conv2d_5, parameter_70, parameter_71, parameter_72, parameter_73
+
+        # pd_op.relu: (1x128x32x32xf32) <- (1x128x32x32xf32)
+        relu_6 = paddle._C_ops.relu(batch_norm__30)
+        del batch_norm__30
+
+        # pd_op.conv2d: (1x128x32x32xf32) <- (1x128x32x32xf32, 128x128x3x3xf32)
+        conv2d_6 = paddle._C_ops.conv2d(
+            relu_6, parameter_69, [1, 1], [1, 1], "EXPLICIT", [1, 1], 1, "NCHW"
+        )
+        del parameter_69, relu_6
+
+        # pd_op.batch_norm_: (1x128x32x32xf32, 128xf32, 128xf32, 128xf32, 128xf32, -1xui8) <- (1x128x32x32xf32, 128xf32, 128xf32, 128xf32, 128xf32)
+        (
+            batch_norm__36,
+            batch_norm__37,
+            batch_norm__38,
+            batch_norm__39,
+            batch_norm__40,
+            batch_norm__41,
+        ) = (lambda x, f: f(x))(
+            paddle._C_ops.batch_norm(
+                conv2d_6,
+                parameter_68,
+                parameter_67,
+                parameter_66,
+                parameter_65,
+                True,
+                float("0.9"),
+                float("1e-05"),
+                "NCHW",
+                True,
+                False,
+            ),
+            lambda out: out
+            if isinstance(out, (list, tuple))
+            else (out, None, None, None, None, None),
+        )
+        del conv2d_6, parameter_65, parameter_66, parameter_67, parameter_68
+
+        # pd_op.conv2d: (1x128x32x32xf32) <- (1x64x64x64xf32, 128x64x1x1xf32)
+        conv2d_7 = paddle._C_ops.conv2d(
+            relu_5, parameter_64, [2, 2], [0, 0], "EXPLICIT", [1, 1], 1, "NCHW"
+        )
+        del parameter_64
+
+        # pd_op.batch_norm_: (1x128x32x32xf32, 128xf32, 128xf32, 128xf32, 128xf32, -1xui8) <- (1x128x32x32xf32, 128xf32, 128xf32, 128xf32, 128xf32)
+        (
+            batch_norm__42,
+            batch_norm__43,
+            batch_norm__44,
+            batch_norm__45,
+            batch_norm__46,
+            batch_norm__47,
+        ) = (lambda x, f: f(x))(
+            paddle._C_ops.batch_norm(
+                conv2d_7,
+                parameter_63,
+                parameter_62,
+                parameter_61,
+                parameter_60,
+                True,
+                float("0.9"),
+                float("1e-05"),
+                "NCHW",
+                True,
+                False,
+            ),
+            lambda out: out
+            if isinstance(out, (list, tuple))
+            else (out, None, None, None, None, None),
+        )
+        del conv2d_7, parameter_60, parameter_61, parameter_62, parameter_63
+
+        # pd_op.add: (1x128x32x32xf32) <- (1x128x32x32xf32, 1x128x32x32xf32)
+        add_2 = paddle._C_ops.add(batch_norm__36, batch_norm__42)
+        del batch_norm__36, batch_norm__42
+
+        # pd_op.relu: (1x128x32x32xf32) <- (1x128x32x32xf32)
+        relu_7 = paddle._C_ops.relu(add_2)
+        del add_2
+
+        # pd_op.conv2d: (1x128x32x32xf32) <- (1x128x32x32xf32, 128x128x3x3xf32)
+        conv2d_8 = paddle._C_ops.conv2d(
+            relu_7, parameter_59, [1, 1], [1, 1], "EXPLICIT", [1, 1], 1, "NCHW"
+        )
+        del parameter_59
+
+        # pd_op.batch_norm_: (1x128x32x32xf32, 128xf32, 128xf32, 128xf32, 128xf32, -1xui8) <- (1x128x32x32xf32, 128xf32, 128xf32, 128xf32, 128xf32)
+        (
+            batch_norm__48,
+            batch_norm__49,
+            batch_norm__50,
+            batch_norm__51,
+            batch_norm__52,
+            batch_norm__53,
+        ) = (lambda x, f: f(x))(
+            paddle._C_ops.batch_norm(
+                conv2d_8,
+                parameter_58,
+                parameter_57,
+                parameter_56,
+                parameter_55,
+                True,
+                float("0.9"),
+                float("1e-05"),
+                "NCHW",
+                True,
+                False,
+            ),
+            lambda out: out
+            if isinstance(out, (list, tuple))
+            else (out, None, None, None, None, None),
+        )
+        del conv2d_8, parameter_55, parameter_56, parameter_57, parameter_58
+
+        # pd_op.relu: (1x128x32x32xf32) <- (1x128x32x32xf32)
+        relu_8 = paddle._C_ops.relu(batch_norm__48)
+        del batch_norm__48
+
+        # pd_op.conv2d: (1x128x32x32xf32) <- (1x128x32x32xf32, 128x128x3x3xf32)
+        conv2d_9 = paddle._C_ops.conv2d(
+            relu_8, parameter_54, [1, 1], [1, 1], "EXPLICIT", [1, 1], 1, "NCHW"
+        )
+        del parameter_54, relu_8
+
+        # pd_op.batch_norm_: (1x128x32x32xf32, 128xf32, 128xf32, 128xf32, 128xf32, -1xui8) <- (1x128x32x32xf32, 128xf32, 128xf32, 128xf32, 128xf32)
+        (
+            batch_norm__54,
+            batch_norm__55,
+            batch_norm__56,
+            batch_norm__57,
+            batch_norm__58,
+            batch_norm__59,
+        ) = (lambda x, f: f(x))(
+            paddle._C_ops.batch_norm(
+                conv2d_9,
+                parameter_53,
+                parameter_52,
+                parameter_51,
+                parameter_50,
+                True,
+                float("0.9"),
+                float("1e-05"),
+                "NCHW",
+                True,
+                False,
+            ),
+            lambda out: out
+            if isinstance(out, (list, tuple))
+            else (out, None, None, None, None, None),
+        )
+        del conv2d_9, parameter_50, parameter_51, parameter_52, parameter_53
+
+        # pd_op.add: (1x128x32x32xf32) <- (1x128x32x32xf32, 1x128x32x32xf32)
+        add_3 = paddle._C_ops.add(batch_norm__54, relu_7)
+        del batch_norm__54, relu_7
+
+        # pd_op.relu: (1x128x32x32xf32) <- (1x128x32x32xf32)
+        relu_9 = paddle._C_ops.relu(add_3)
+        del add_3
+
+        # pd_op.conv2d: (1x256x16x16xf32) <- (1x128x32x32xf32, 256x128x3x3xf32)
+        conv2d_10 = paddle._C_ops.conv2d(
+            relu_9, parameter_49, [2, 2], [1, 1], "EXPLICIT", [1, 1], 1, "NCHW"
+        )
+        del parameter_49
+
+        # pd_op.batch_norm_: (1x256x16x16xf32, 256xf32, 256xf32, 256xf32, 256xf32, -1xui8) <- (1x256x16x16xf32, 256xf32, 256xf32, 256xf32, 256xf32)
+        (
+            batch_norm__60,
+            batch_norm__61,
+            batch_norm__62,
+            batch_norm__63,
+            batch_norm__64,
+            batch_norm__65,
+        ) = (lambda x, f: f(x))(
+            paddle._C_ops.batch_norm(
+                conv2d_10,
+                parameter_48,
+                parameter_47,
+                parameter_46,
+                parameter_45,
+                True,
+                float("0.9"),
+                float("1e-05"),
+                "NCHW",
+                True,
+                False,
+            ),
+            lambda out: out
+            if isinstance(out, (list, tuple))
+            else (out, None, None, None, None, None),
+        )
+        del conv2d_10, parameter_45, parameter_46, parameter_47, parameter_48
+
+        # pd_op.relu: (1x256x16x16xf32) <- (1x256x16x16xf32)
+        relu_10 = paddle._C_ops.relu(batch_norm__60)
+        del batch_norm__60
+
+        # pd_op.conv2d: (1x256x16x16xf32) <- (1x256x16x16xf32, 256x256x3x3xf32)
+        conv2d_11 = paddle._C_ops.conv2d(
+            relu_10, parameter_44, [1, 1], [1, 1], "EXPLICIT", [1, 1], 1, "NCHW"
+        )
+        del parameter_44, relu_10
+
+        # pd_op.batch_norm_: (1x256x16x16xf32, 256xf32, 256xf32, 256xf32, 256xf32, -1xui8) <- (1x256x16x16xf32, 256xf32, 256xf32, 256xf32, 256xf32)
+        (
+            batch_norm__66,
+            batch_norm__67,
+            batch_norm__68,
+            batch_norm__69,
+            batch_norm__70,
+            batch_norm__71,
+        ) = (lambda x, f: f(x))(
+            paddle._C_ops.batch_norm(
+                conv2d_11,
+                parameter_43,
+                parameter_42,
+                parameter_41,
+                parameter_40,
+                True,
+                float("0.9"),
+                float("1e-05"),
+                "NCHW",
+                True,
+                False,
+            ),
+            lambda out: out
+            if isinstance(out, (list, tuple))
+            else (out, None, None, None, None, None),
+        )
+        del conv2d_11, parameter_40, parameter_41, parameter_42, parameter_43
+
+        # pd_op.conv2d: (1x256x16x16xf32) <- (1x128x32x32xf32, 256x128x1x1xf32)
+        conv2d_12 = paddle._C_ops.conv2d(
+            relu_9, parameter_39, [2, 2], [0, 0], "EXPLICIT", [1, 1], 1, "NCHW"
+        )
+        del parameter_39
+
+        # pd_op.batch_norm_: (1x256x16x16xf32, 256xf32, 256xf32, 256xf32, 256xf32, -1xui8) <- (1x256x16x16xf32, 256xf32, 256xf32, 256xf32, 256xf32)
+        (
+            batch_norm__72,
+            batch_norm__73,
+            batch_norm__74,
+            batch_norm__75,
+            batch_norm__76,
+            batch_norm__77,
+        ) = (lambda x, f: f(x))(
+            paddle._C_ops.batch_norm(
+                conv2d_12,
+                parameter_38,
+                parameter_37,
+                parameter_36,
+                parameter_35,
+                True,
+                float("0.9"),
+                float("1e-05"),
+                "NCHW",
+                True,
+                False,
+            ),
+            lambda out: out
+            if isinstance(out, (list, tuple))
+            else (out, None, None, None, None, None),
+        )
+        del conv2d_12, parameter_35, parameter_36, parameter_37, parameter_38
+
+        # pd_op.add: (1x256x16x16xf32) <- (1x256x16x16xf32, 1x256x16x16xf32)
+        add_4 = paddle._C_ops.add(batch_norm__66, batch_norm__72)
+        del batch_norm__66, batch_norm__72
+
+        # pd_op.relu: (1x256x16x16xf32) <- (1x256x16x16xf32)
+        relu_11 = paddle._C_ops.relu(add_4)
+        del add_4
+
+        # pd_op.conv2d: (1x256x16x16xf32) <- (1x256x16x16xf32, 256x256x3x3xf32)
+        conv2d_13 = paddle._C_ops.conv2d(
+            relu_11, parameter_34, [1, 1], [1, 1], "EXPLICIT", [1, 1], 1, "NCHW"
+        )
+        del parameter_34
+
+        # pd_op.batch_norm_: (1x256x16x16xf32, 256xf32, 256xf32, 256xf32, 256xf32, -1xui8) <- (1x256x16x16xf32, 256xf32, 256xf32, 256xf32, 256xf32)
+        (
+            batch_norm__78,
+            batch_norm__79,
+            batch_norm__80,
+            batch_norm__81,
+            batch_norm__82,
+            batch_norm__83,
+        ) = (lambda x, f: f(x))(
+            paddle._C_ops.batch_norm(
+                conv2d_13,
+                parameter_33,
+                parameter_32,
+                parameter_31,
+                parameter_30,
+                True,
+                float("0.9"),
+                float("1e-05"),
+                "NCHW",
+                True,
+                False,
+            ),
+            lambda out: out
+            if isinstance(out, (list, tuple))
+            else (out, None, None, None, None, None),
+        )
+        del conv2d_13, parameter_30, parameter_31, parameter_32, parameter_33
+
+        # pd_op.relu: (1x256x16x16xf32) <- (1x256x16x16xf32)
+        relu_12 = paddle._C_ops.relu(batch_norm__78)
+        del batch_norm__78
+
+        # pd_op.conv2d: (1x256x16x16xf32) <- (1x256x16x16xf32, 256x256x3x3xf32)
+        conv2d_14 = paddle._C_ops.conv2d(
+            relu_12, parameter_29, [1, 1], [1, 1], "EXPLICIT", [1, 1], 1, "NCHW"
+        )
+        del parameter_29, relu_12
+
+        # pd_op.batch_norm_: (1x256x16x16xf32, 256xf32, 256xf32, 256xf32, 256xf32, -1xui8) <- (1x256x16x16xf32, 256xf32, 256xf32, 256xf32, 256xf32)
+        (
+            batch_norm__84,
+            batch_norm__85,
+            batch_norm__86,
+            batch_norm__87,
+            batch_norm__88,
+            batch_norm__89,
+        ) = (lambda x, f: f(x))(
+            paddle._C_ops.batch_norm(
+                conv2d_14,
+                parameter_28,
+                parameter_27,
+                parameter_26,
+                parameter_25,
+                True,
+                float("0.9"),
+                float("1e-05"),
+                "NCHW",
+                True,
+                False,
+            ),
+            lambda out: out
+            if isinstance(out, (list, tuple))
+            else (out, None, None, None, None, None),
+        )
+        del conv2d_14, parameter_25, parameter_26, parameter_27, parameter_28
+
+        # pd_op.add: (1x256x16x16xf32) <- (1x256x16x16xf32, 1x256x16x16xf32)
+        add_5 = paddle._C_ops.add(batch_norm__84, relu_11)
+        del batch_norm__84, relu_11
+
+        # pd_op.relu: (1x256x16x16xf32) <- (1x256x16x16xf32)
+        relu_0 = paddle._C_ops.relu(add_5)
+        del add_5
+
+        # pd_op.conv2d: (1x512x8x8xf32) <- (1x256x16x16xf32, 512x256x3x3xf32)
+        conv2d_15 = paddle._C_ops.conv2d(
+            relu_0, parameter_24, [2, 2], [1, 1], "EXPLICIT", [1, 1], 1, "NCHW"
+        )
+        del parameter_24
+
+        # pd_op.batch_norm_: (1x512x8x8xf32, 512xf32, 512xf32, 512xf32, 512xf32, -1xui8) <- (1x512x8x8xf32, 512xf32, 512xf32, 512xf32, 512xf32)
+        (
+            batch_norm__90,
+            batch_norm__91,
+            batch_norm__92,
+            batch_norm__93,
+            batch_norm__94,
+            batch_norm__95,
+        ) = (lambda x, f: f(x))(
+            paddle._C_ops.batch_norm(
+                conv2d_15,
+                parameter_23,
+                parameter_22,
+                parameter_21,
+                parameter_20,
+                True,
+                float("0.9"),
+                float("1e-05"),
+                "NCHW",
+                True,
+                False,
+            ),
+            lambda out: out
+            if isinstance(out, (list, tuple))
+            else (out, None, None, None, None, None),
+        )
+        del conv2d_15, parameter_20, parameter_21, parameter_22, parameter_23
+
+        # pd_op.relu: (1x512x8x8xf32) <- (1x512x8x8xf32)
+        relu_13 = paddle._C_ops.relu(batch_norm__90)
+        del batch_norm__90
+
+        # pd_op.conv2d: (1x512x8x8xf32) <- (1x512x8x8xf32, 512x512x3x3xf32)
+        conv2d_16 = paddle._C_ops.conv2d(
+            relu_13, parameter_19, [1, 1], [1, 1], "EXPLICIT", [1, 1], 1, "NCHW"
+        )
+        del parameter_19, relu_13
+
+        # pd_op.batch_norm_: (1x512x8x8xf32, 512xf32, 512xf32, 512xf32, 512xf32, -1xui8) <- (1x512x8x8xf32, 512xf32, 512xf32, 512xf32, 512xf32)
+        (
+            batch_norm__96,
+            batch_norm__97,
+            batch_norm__98,
+            batch_norm__99,
+            batch_norm__100,
+            batch_norm__101,
+        ) = (lambda x, f: f(x))(
+            paddle._C_ops.batch_norm(
+                conv2d_16,
+                parameter_18,
+                parameter_17,
+                parameter_16,
+                parameter_15,
+                True,
+                float("0.9"),
+                float("1e-05"),
+                "NCHW",
+                True,
+                False,
+            ),
+            lambda out: out
+            if isinstance(out, (list, tuple))
+            else (out, None, None, None, None, None),
+        )
+        del conv2d_16, parameter_15, parameter_16, parameter_17, parameter_18
+
+        # pd_op.conv2d: (1x512x8x8xf32) <- (1x256x16x16xf32, 512x256x1x1xf32)
+        conv2d_17 = paddle._C_ops.conv2d(
+            relu_0, parameter_14, [2, 2], [0, 0], "EXPLICIT", [1, 1], 1, "NCHW"
+        )
+        del parameter_14, relu_0
+
+        # pd_op.batch_norm_: (1x512x8x8xf32, 512xf32, 512xf32, 512xf32, 512xf32, -1xui8) <- (1x512x8x8xf32, 512xf32, 512xf32, 512xf32, 512xf32)
+        (
+            batch_norm__102,
+            batch_norm__103,
+            batch_norm__104,
+            batch_norm__105,
+            batch_norm__106,
+            batch_norm__107,
+        ) = (lambda x, f: f(x))(
+            paddle._C_ops.batch_norm(
+                conv2d_17,
+                parameter_13,
+                parameter_12,
+                parameter_11,
+                parameter_10,
+                True,
+                float("0.9"),
+                float("1e-05"),
+                "NCHW",
+                True,
+                False,
+            ),
+            lambda out: out
+            if isinstance(out, (list, tuple))
+            else (out, None, None, None, None, None),
+        )
+        del conv2d_17, parameter_10, parameter_11, parameter_12, parameter_13
+
+        # pd_op.add: (1x512x8x8xf32) <- (1x512x8x8xf32, 1x512x8x8xf32)
+        add_6 = paddle._C_ops.add(batch_norm__96, batch_norm__102)
+        del batch_norm__102, batch_norm__96
+
+        # pd_op.relu: (1x512x8x8xf32) <- (1x512x8x8xf32)
+        relu_14 = paddle._C_ops.relu(add_6)
+        del add_6
+
+        # pd_op.conv2d: (1x512x8x8xf32) <- (1x512x8x8xf32, 512x512x3x3xf32)
+        conv2d_18 = paddle._C_ops.conv2d(
+            relu_14, parameter_9, [1, 1], [1, 1], "EXPLICIT", [1, 1], 1, "NCHW"
+        )
+        del parameter_9
+
+        # pd_op.batch_norm_: (1x512x8x8xf32, 512xf32, 512xf32, 512xf32, 512xf32, -1xui8) <- (1x512x8x8xf32, 512xf32, 512xf32, 512xf32, 512xf32)
+        (
+            batch_norm__108,
+            batch_norm__109,
+            batch_norm__110,
+            batch_norm__111,
+            batch_norm__112,
+            batch_norm__113,
+        ) = (lambda x, f: f(x))(
+            paddle._C_ops.batch_norm(
+                conv2d_18,
+                parameter_8,
+                parameter_7,
+                parameter_6,
+                parameter_5,
+                True,
+                float("0.9"),
+                float("1e-05"),
+                "NCHW",
+                True,
+                False,
+            ),
+            lambda out: out
+            if isinstance(out, (list, tuple))
+            else (out, None, None, None, None, None),
+        )
+        del conv2d_18, parameter_5, parameter_6, parameter_7, parameter_8
+
+        # pd_op.relu: (1x512x8x8xf32) <- (1x512x8x8xf32)
+        relu_15 = paddle._C_ops.relu(batch_norm__108)
+        del batch_norm__108
+
+        # pd_op.conv2d: (1x512x8x8xf32) <- (1x512x8x8xf32, 512x512x3x3xf32)
+        conv2d_19 = paddle._C_ops.conv2d(
+            relu_15, parameter_4, [1, 1], [1, 1], "EXPLICIT", [1, 1], 1, "NCHW"
+        )
+        del parameter_4, relu_15
+
+        # pd_op.batch_norm_: (1x512x8x8xf32, 512xf32, 512xf32, 512xf32, 512xf32, -1xui8) <- (1x512x8x8xf32, 512xf32, 512xf32, 512xf32, 512xf32)
+        (
+            batch_norm__114,
+            batch_norm__115,
+            batch_norm__116,
+            batch_norm__117,
+            batch_norm__118,
+            batch_norm__119,
+        ) = (lambda x, f: f(x))(
+            paddle._C_ops.batch_norm(
+                conv2d_19,
+                parameter_3,
+                parameter_2,
+                parameter_1,
+                parameter_0,
+                True,
+                float("0.9"),
+                float("1e-05"),
+                "NCHW",
+                True,
+                False,
+            ),
+            lambda out: out
+            if isinstance(out, (list, tuple))
+            else (out, None, None, None, None, None),
+        )
+        del conv2d_19, parameter_0, parameter_1, parameter_2, parameter_3
+
+        # pd_op.add: (1x512x8x8xf32) <- (1x512x8x8xf32, 1x512x8x8xf32)
+        add_7 = paddle._C_ops.add(batch_norm__114, relu_14)
+        del batch_norm__114, relu_14
+
+        # pd_op.relu: (1x512x8x8xf32) <- (1x512x8x8xf32)
+        relu_16 = paddle._C_ops.relu(add_7)
+        del add_7, relu_5, relu_9
+
+        return relu_16
diff --git a/paddle_samples/PaddleX/STFPM/weight_meta.py b/paddle_samples/PaddleX/STFPM/weight_meta.py
new file mode 100644
index 000000000..ddb7f8a1b
--- /dev/null
+++ b/paddle_samples/PaddleX/STFPM/weight_meta.py
@@ -0,0 +1,1058 @@
+class Program_weight_tensor_parameter_0:
+    name = "parameter_0"
+    shape = [512]
+    dtype = "float32"
+    min_val = float("0.0318808")
+    max_val = float("0.528011")
+    mean = float("0.22751")
+    std = float("0.0630725")
+    data = None
+
+
+class Program_weight_tensor_parameter_1:
+    name = "parameter_1"
+    shape = [512]
+    dtype = "float32"
+    min_val = float("1.4083")
+    max_val = float("2.44238")
+    mean = float("1.74648")
+    std = float("0.121457")
+    data = None
+
+
+class Program_weight_tensor_parameter_2:
+    name = "parameter_2"
+    shape = [512]
+    dtype = "float32"
+    min_val = float("0.00938787")
+    max_val = float("0.0359624")
+    mean = float("0.0140218")
+    std = float("0.00197666")
+    data = None
+
+
+class Program_weight_tensor_parameter_3:
+    name = "parameter_3"
+    shape = [512]
+    dtype = "float32"
+    min_val = float("-0.330013")
+    max_val = float("0.0815704")
+    mean = float("-0.0284101")
+    std = float("0.0258458")
+    data = None
+
+
+class Program_weight_tensor_parameter_4:
+    name = "parameter_4"
+    shape = [512, 512, 3, 3]
+    dtype = "float32"
+    min_val = float("-0.390469")
+    max_val = float("0.251665")
+    mean = float("-0.000136553")
+    std = float("0.0130795")
+    data = None
+
+
+class Program_weight_tensor_parameter_5:
+    name = "parameter_5"
+    shape = [512]
+    dtype = "float32"
+    min_val = float("-0.540901")
+    max_val = float("0.150347")
+    mean = float("-0.240079")
+    std = float("0.092521")
+    data = None
+
+
+class Program_weight_tensor_parameter_6:
+    name = "parameter_6"
+    shape = [512]
+    dtype = "float32"
+    min_val = float("0.127383")
+    max_val = float("0.425453")
+    mean = float("0.287491")
+    std = float("0.0426542")
+    data = None
+
+
+class Program_weight_tensor_parameter_7:
+    name = "parameter_7"
+    shape = [512]
+    dtype = "float32"
+    min_val = float("0.188243")
+    max_val = float("0.892962")
+    mean = float("0.328185")
+    std = float("0.0675297")
+    data = None
+
+
+class Program_weight_tensor_parameter_8:
+    name = "parameter_8"
+    shape = [512]
+    dtype = "float32"
+    min_val = float("-1.67777")
+    max_val = float("1.73262")
+    mean = float("-1.09431")
+    std = float("0.226589")
+    data = None
+
+
+class Program_weight_tensor_parameter_9:
+    name = "parameter_9"
+    shape = [512, 512, 3, 3]
+    dtype = "float32"
+    min_val = float("-0.331881")
+    max_val = float("0.258786")
+    mean = float("-0.00249441")
+    std = float("0.0173289")
+    data = None
+
+
+class Program_weight_tensor_parameter_10:
+    name = "parameter_10"
+    shape = [512]
+    dtype = "float32"
+    min_val = float("-0.376362")
+    max_val = float("0.0591005")
+    mean = float("-0.196825")
+    std = float("0.0614354")
+    data = None
+
+
+class Program_weight_tensor_parameter_11:
+    name = "parameter_11"
+    shape = [512]
+    dtype = "float32"
+    min_val = float("0.0924911")
+    max_val = float("0.518859")
+    mean = float("0.245248")
+    std = float("0.0565846")
+    data = None
+
+
+class Program_weight_tensor_parameter_12:
+    name = "parameter_12"
+    shape = [512]
+    dtype = "float32"
+    min_val = float("0.00425951")
+    max_val = float("0.0404418")
+    mean = float("0.0116802")
+    std = float("0.00391782")
+    data = None
+
+
+class Program_weight_tensor_parameter_13:
+    name = "parameter_13"
+    shape = [512]
+    dtype = "float32"
+    min_val = float("-0.199185")
+    max_val = float("0.608781")
+    mean = float("-0.0179375")
+    std = float("0.0561503")
+    data = None
+
+
+class Program_weight_tensor_parameter_14:
+    name = "parameter_14"
+    shape = [512, 256, 1, 1]
+    dtype = "float32"
+    min_val = float("-0.816973")
+    max_val = float("0.337579")
+    mean = float("-0.000563961")
+    std = float("0.0291861")
+    data = None
+
+
+class Program_weight_tensor_parameter_15:
+    name = "parameter_15"
+    shape = [512]
+    dtype = "float32"
+    min_val = float("-0.376362")
+    max_val = float("0.0591005")
+    mean = float("-0.196825")
+    std = float("0.0614354")
+    data = None
+
+
+class Program_weight_tensor_parameter_16:
+    name = "parameter_16"
+    shape = [512]
+    dtype = "float32"
+    min_val = float("0.223617")
+    max_val = float("1.00544")
+    mean = float("0.531693")
+    std = float("0.0605323")
+    data = None
+
+
+class Program_weight_tensor_parameter_17:
+    name = "parameter_17"
+    shape = [512]
+    dtype = "float32"
+    min_val = float("0.0132799")
+    max_val = float("0.117756")
+    mean = float("0.0236118")
+    std = float("0.00826997")
+    data = None
+
+
+class Program_weight_tensor_parameter_18:
+    name = "parameter_18"
+    shape = [512]
+    dtype = "float32"
+    min_val = float("-0.264202")
+    max_val = float("0.745614")
+    mean = float("-0.122051")
+    std = float("0.0645659")
+    data = None
+
+
+class Program_weight_tensor_parameter_19:
+    name = "parameter_19"
+    shape = [512, 512, 3, 3]
+    dtype = "float32"
+    min_val = float("-0.333323")
+    max_val = float("0.411566")
+    mean = float("-0.00108166")
+    std = float("0.0171628")
+    data = None
+
+
+class Program_weight_tensor_parameter_20:
+    name = "parameter_20"
+    shape = [512]
+    dtype = "float32"
+    min_val = float("-0.699018")
+    max_val = float("0.187542")
+    mean = float("-0.248638")
+    std = float("0.0826924")
+    data = None
+
+
+class Program_weight_tensor_parameter_21:
+    name = "parameter_21"
+    shape = [512]
+    dtype = "float32"
+    min_val = float("0.137705")
+    max_val = float("0.482447")
+    mean = float("0.277886")
+    std = float("0.0342759")
+    data = None
+
+
+class Program_weight_tensor_parameter_22:
+    name = "parameter_22"
+    shape = [512]
+    dtype = "float32"
+    min_val = float("0.075404")
+    max_val = float("0.332755")
+    mean = float("0.13484")
+    std = float("0.0269851")
+    data = None
+
+
+class Program_weight_tensor_parameter_23:
+    name = "parameter_23"
+    shape = [512]
+    dtype = "float32"
+    min_val = float("-1.15621")
+    max_val = float("0.345702")
+    mean = float("-0.325267")
+    std = float("0.175642")
+    data = None
+
+
+class Program_weight_tensor_parameter_24:
+    name = "parameter_24"
+    shape = [512, 256, 3, 3]
+    dtype = "float32"
+    min_val = float("-0.234134")
+    max_val = float("0.319417")
+    mean = float("-0.00139598")
+    std = float("0.0204891")
+    data = None
+
+
+class Program_weight_tensor_parameter_25:
+    name = "parameter_25"
+    shape = [256]
+    dtype = "float32"
+    min_val = float("-0.672112")
+    max_val = float("0.175004")
+    mean = float("-0.207006")
+    std = float("0.144824")
+    data = None
+
+
+class Program_weight_tensor_parameter_26:
+    name = "parameter_26"
+    shape = [256]
+    dtype = "float32"
+    min_val = float("0.0572091")
+    max_val = float("0.621537")
+    mean = float("0.254945")
+    std = float("0.112732")
+    data = None
+
+
+class Program_weight_tensor_parameter_27:
+    name = "parameter_27"
+    shape = [256]
+    dtype = "float32"
+    min_val = float("0.00786664")
+    max_val = float("0.0595664")
+    mean = float("0.0225806")
+    std = float("0.00931339")
+    data = None
+
+
+class Program_weight_tensor_parameter_28:
+    name = "parameter_28"
+    shape = [256]
+    dtype = "float32"
+    min_val = float("-0.637725")
+    max_val = float("0.4241")
+    mean = float("-0.104907")
+    std = float("0.103395")
+    data = None
+
+
+class Program_weight_tensor_parameter_29:
+    name = "parameter_29"
+    shape = [256, 256, 3, 3]
+    dtype = "float32"
+    min_val = float("-0.223175")
+    max_val = float("0.270533")
+    mean = float("-0.00182474")
+    std = float("0.0213125")
+    data = None
+
+
+class Program_weight_tensor_parameter_30:
+    name = "parameter_30"
+    shape = [256]
+    dtype = "float32"
+    min_val = float("-0.760569")
+    max_val = float("0.188128")
+    mean = float("-0.250213")
+    std = float("0.123843")
+    data = None
+
+
+class Program_weight_tensor_parameter_31:
+    name = "parameter_31"
+    shape = [256]
+    dtype = "float32"
+    min_val = float("0.114274")
+    max_val = float("0.514129")
+    mean = float("0.289284")
+    std = float("0.0546789")
+    data = None
+
+
+class Program_weight_tensor_parameter_32:
+    name = "parameter_32"
+    shape = [256]
+    dtype = "float32"
+    min_val = float("0.0858827")
+    max_val = float("0.477435")
+    mean = float("0.16419")
+    std = float("0.0512708")
+    data = None
+
+
+class Program_weight_tensor_parameter_33:
+    name = "parameter_33"
+    shape = [256]
+    dtype = "float32"
+    min_val = float("-1.85871")
+    max_val = float("1.0722")
+    mean = float("-0.395986")
+    std = float("0.348973")
+    data = None
+
+
+class Program_weight_tensor_parameter_34:
+    name = "parameter_34"
+    shape = [256, 256, 3, 3]
+    dtype = "float32"
+    min_val = float("-0.26959")
+    max_val = float("0.265632")
+    mean = float("-0.00185683")
+    std = float("0.0228273")
+    data = None
+
+
+class Program_weight_tensor_parameter_35:
+    name = "parameter_35"
+    shape = [256]
+    dtype = "float32"
+    min_val = float("-0.49799")
+    max_val = float("0.163236")
+    mean = float("-0.0490798")
+    std = float("0.0975837")
+    data = None
+
+
+class Program_weight_tensor_parameter_36:
+    name = "parameter_36"
+    shape = [256]
+    dtype = "float32"
+    min_val = float("0.01382")
+    max_val = float("0.2887")
+    mean = float("0.109583")
+    std = float("0.0481557")
+    data = None
+
+
+class Program_weight_tensor_parameter_37:
+    name = "parameter_37"
+    shape = [256]
+    dtype = "float32"
+    min_val = float("0.00216973")
+    max_val = float("0.0556628")
+    mean = float("0.0213781")
+    std = float("0.010213")
+    data = None
+
+
+class Program_weight_tensor_parameter_38:
+    name = "parameter_38"
+    shape = [256]
+    dtype = "float32"
+    min_val = float("-0.425395")
+    max_val = float("0.372406")
+    mean = float("-0.046856")
+    std = float("0.106813")
+    data = None
+
+
+class Program_weight_tensor_parameter_39:
+    name = "parameter_39"
+    shape = [256, 128, 1, 1]
+    dtype = "float32"
+    min_val = float("-0.405692")
+    max_val = float("0.362475")
+    mean = float("-0.00162062")
+    std = float("0.0387306")
+    data = None
+
+
+class Program_weight_tensor_parameter_40:
+    name = "parameter_40"
+    shape = [256]
+    dtype = "float32"
+    min_val = float("-0.49799")
+    max_val = float("0.163236")
+    mean = float("-0.0490798")
+    std = float("0.0975837")
+    data = None
+
+
+class Program_weight_tensor_parameter_41:
+    name = "parameter_41"
+    shape = [256]
+    dtype = "float32"
+    min_val = float("0.141176")
+    max_val = float("0.569826")
+    mean = float("0.329862")
+    std = float("0.0774057")
+    data = None
+
+
+class Program_weight_tensor_parameter_42:
+    name = "parameter_42"
+    shape = [256]
+    dtype = "float32"
+    min_val = float("0.0439446")
+    max_val = float("0.252004")
+    mean = float("0.0913649")
+    std = float("0.0310039")
+    data = None
+
+
+class Program_weight_tensor_parameter_43:
+    name = "parameter_43"
+    shape = [256]
+    dtype = "float32"
+    min_val = float("-0.675011")
+    max_val = float("0.354508")
+    mean = float("-0.167951")
+    std = float("0.153458")
+    data = None
+
+
+class Program_weight_tensor_parameter_44:
+    name = "parameter_44"
+    shape = [256, 256, 3, 3]
+    dtype = "float32"
+    min_val = float("-0.358815")
+    max_val = float("0.4652")
+    mean = float("-0.00110841")
+    std = float("0.025435")
+    data = None
+
+
+class Program_weight_tensor_parameter_45:
+    name = "parameter_45"
+    shape = [256]
+    dtype = "float32"
+    min_val = float("-0.375005")
+    max_val = float("0.168939")
+    mean = float("-0.141223")
+    std = float("0.0907853")
+    data = None
+
+
+class Program_weight_tensor_parameter_46:
+    name = "parameter_46"
+    shape = [256]
+    dtype = "float32"
+    min_val = float("0.199538")
+    max_val = float("0.480196")
+    mean = float("0.31944")
+    std = float("0.0410408")
+    data = None
+
+
+class Program_weight_tensor_parameter_47:
+    name = "parameter_47"
+    shape = [256]
+    dtype = "float32"
+    min_val = float("0.149106")
+    max_val = float("0.644703")
+    mean = float("0.293793")
+    std = float("0.0770305")
+    data = None
+
+
+class Program_weight_tensor_parameter_48:
+    name = "parameter_48"
+    shape = [256]
+    dtype = "float32"
+    min_val = float("-1.6928")
+    max_val = float("1.3048")
+    mean = float("-0.436802")
+    std = float("0.392002")
+    data = None
+
+
+class Program_weight_tensor_parameter_49:
+    name = "parameter_49"
+    shape = [256, 128, 3, 3]
+    dtype = "float32"
+    min_val = float("-0.339269")
+    max_val = float("0.390058")
+    mean = float("-0.00155937")
+    std = float("0.029771")
+    data = None
+
+
+class Program_weight_tensor_parameter_50:
+    name = "parameter_50"
+    shape = [128]
+    dtype = "float32"
+    min_val = float("-0.563891")
+    max_val = float("0.325674")
+    mean = float("-0.109761")
+    std = float("0.167878")
+    data = None
+
+
+class Program_weight_tensor_parameter_51:
+    name = "parameter_51"
+    shape = [128]
+    dtype = "float32"
+    min_val = float("0.0450367")
+    max_val = float("0.627042")
+    mean = float("0.294486")
+    std = float("0.141658")
+    data = None
+
+
+class Program_weight_tensor_parameter_52:
+    name = "parameter_52"
+    shape = [128]
+    dtype = "float32"
+    min_val = float("0.00875327")
+    max_val = float("0.105119")
+    mean = float("0.0443794")
+    std = float("0.0217019")
+    data = None
+
+
+class Program_weight_tensor_parameter_53:
+    name = "parameter_53"
+    shape = [128]
+    dtype = "float32"
+    min_val = float("-0.663128")
+    max_val = float("0.686416")
+    mean = float("-0.105243")
+    std = float("0.14984")
+    data = None
+
+
+class Program_weight_tensor_parameter_54:
+    name = "parameter_54"
+    shape = [128, 128, 3, 3]
+    dtype = "float32"
+    min_val = float("-0.335569")
+    max_val = float("0.389193")
+    mean = float("-0.0018809")
+    std = float("0.0311068")
+    data = None
+
+
+class Program_weight_tensor_parameter_55:
+    name = "parameter_55"
+    shape = [128]
+    dtype = "float32"
+    min_val = float("-0.696134")
+    max_val = float("0.0644196")
+    mean = float("-0.221818")
+    std = float("0.113874")
+    data = None
+
+
+class Program_weight_tensor_parameter_56:
+    name = "parameter_56"
+    shape = [128]
+    dtype = "float32"
+    min_val = float("0.233854")
+    max_val = float("0.581869")
+    mean = float("0.344386")
+    std = float("0.0526995")
+    data = None
+
+
+class Program_weight_tensor_parameter_57:
+    name = "parameter_57"
+    shape = [128]
+    dtype = "float32"
+    min_val = float("0.104491")
+    max_val = float("0.845752")
+    mean = float("0.282537")
+    std = float("0.123765")
+    data = None
+
+
+class Program_weight_tensor_parameter_58:
+    name = "parameter_58"
+    shape = [128]
+    dtype = "float32"
+    min_val = float("-1.16363")
+    max_val = float("1.28772")
+    mean = float("-0.31756")
+    std = float("0.327098")
+    data = None
+
+
+class Program_weight_tensor_parameter_59:
+    name = "parameter_59"
+    shape = [128, 128, 3, 3]
+    dtype = "float32"
+    min_val = float("-0.330242")
+    max_val = float("0.496327")
+    mean = float("-0.00182635")
+    std = float("0.0354316")
+    data = None
+
+
+class Program_weight_tensor_parameter_60:
+    name = "parameter_60"
+    shape = [128]
+    dtype = "float32"
+    min_val = float("-0.244071")
+    max_val = float("0.210061")
+    mean = float("-0.0105737")
+    std = float("0.0900847")
+    data = None
+
+
+class Program_weight_tensor_parameter_61:
+    name = "parameter_61"
+    shape = [128]
+    dtype = "float32"
+    min_val = float("0.00639295")
+    max_val = float("0.53858")
+    mean = float("0.181035")
+    std = float("0.0949545")
+    data = None
+
+
+class Program_weight_tensor_parameter_62:
+    name = "parameter_62"
+    shape = [128]
+    dtype = "float32"
+    min_val = float("0.00843929")
+    max_val = float("0.287562")
+    mean = float("0.0786078")
+    std = float("0.0558961")
+    data = None
+
+
+class Program_weight_tensor_parameter_63:
+    name = "parameter_63"
+    shape = [128]
+    dtype = "float32"
+    min_val = float("-0.800473")
+    max_val = float("1.31885")
+    mean = float("-0.0430369")
+    std = float("0.273573")
+    data = None
+
+
+class Program_weight_tensor_parameter_64:
+    name = "parameter_64"
+    shape = [128, 64, 1, 1]
+    dtype = "float32"
+    min_val = float("-0.647083")
+    max_val = float("0.771812")
+    mean = float("-0.00173632")
+    std = float("0.0693106")
+    data = None
+
+
+class Program_weight_tensor_parameter_65:
+    name = "parameter_65"
+    shape = [128]
+    dtype = "float32"
+    min_val = float("-0.244071")
+    max_val = float("0.210061")
+    mean = float("-0.0105737")
+    std = float("0.0900847")
+    data = None
+
+
+class Program_weight_tensor_parameter_66:
+    name = "parameter_66"
+    shape = [128]
+    dtype = "float32"
+    min_val = float("0.110028")
+    max_val = float("0.649541")
+    mean = float("0.350825")
+    std = float("0.105627")
+    data = None
+
+
+class Program_weight_tensor_parameter_67:
+    name = "parameter_67"
+    shape = [128]
+    dtype = "float32"
+    min_val = float("0.0278526")
+    max_val = float("0.314263")
+    mean = float("0.101928")
+    std = float("0.050357")
+    data = None
+
+
+class Program_weight_tensor_parameter_68:
+    name = "parameter_68"
+    shape = [128]
+    dtype = "float32"
+    min_val = float("-1.18115")
+    max_val = float("1.04257")
+    mean = float("-0.143823")
+    std = float("0.259063")
+    data = None
+
+
+class Program_weight_tensor_parameter_69:
+    name = "parameter_69"
+    shape = [128, 128, 3, 3]
+    dtype = "float32"
+    min_val = float("-0.370554")
+    max_val = float("0.431965")
+    mean = float("-0.000884545")
+    std = float("0.0360175")
+    data = None
+
+
+class Program_weight_tensor_parameter_70:
+    name = "parameter_70"
+    shape = [128]
+    dtype = "float32"
+    min_val = float("-0.361669")
+    max_val = float("0.467714")
+    mean = float("-0.0826597")
+    std = float("0.10144")
+    data = None
+
+
+class Program_weight_tensor_parameter_71:
+    name = "parameter_71"
+    shape = [128]
+    dtype = "float32"
+    min_val = float("0.189793")
+    max_val = float("0.418438")
+    mean = float("0.340827")
+    std = float("0.0390889")
+    data = None
+
+
+class Program_weight_tensor_parameter_72:
+    name = "parameter_72"
+    shape = [128]
+    dtype = "float32"
+    min_val = float("0.311338")
+    max_val = float("1.74052")
+    mean = float("0.78656")
+    std = float("0.238812")
+    data = None
+
+
+class Program_weight_tensor_parameter_73:
+    name = "parameter_73"
+    shape = [128]
+    dtype = "float32"
+    min_val = float("-1.56294")
+    max_val = float("1.06375")
+    mean = float("-0.360913")
+    std = float("0.430996")
+    data = None
+
+
+class Program_weight_tensor_parameter_74:
+    name = "parameter_74"
+    shape = [128, 64, 3, 3]
+    dtype = "float32"
+    min_val = float("-0.347813")
+    max_val = float("0.354365")
+    mean = float("-0.00150566")
+    std = float("0.0436333")
+    data = None
+
+
+class Program_weight_tensor_parameter_75:
+    name = "parameter_75"
+    shape = [64]
+    dtype = "float32"
+    min_val = float("0")
+    max_val = float("0.5")
+    data = None
+
+
+class Program_weight_tensor_parameter_76:
+    name = "parameter_76"
+    shape = [64]
+    dtype = "float32"
+    min_val = float("0")
+    max_val = float("0.5")
+    data = None
+
+
+class Program_weight_tensor_parameter_77:
+    name = "parameter_77"
+    shape = [64]
+    dtype = "float32"
+    min_val = float("0")
+    max_val = float("0.5")
+    data = None
+
+
+class Program_weight_tensor_parameter_78:
+    name = "parameter_78"
+    shape = [64]
+    dtype = "float32"
+    min_val = float("0")
+    max_val = float("0.5")
+    data = None
+
+
+class Program_weight_tensor_parameter_79:
+    name = "parameter_79"
+    shape = [64, 64, 3, 3]
+    dtype = "float32"
+    min_val = float("-0.380435")
+    max_val = float("0.349747")
+    mean = float("-0.00119679")
+    std = float("0.0449416")
+    data = None
+
+
+class Program_weight_tensor_parameter_80:
+    name = "parameter_80"
+    shape = [64]
+    dtype = "float32"
+    min_val = float("0")
+    max_val = float("0.5")
+    data = None
+
+
+class Program_weight_tensor_parameter_81:
+    name = "parameter_81"
+    shape = [64]
+    dtype = "float32"
+    min_val = float("0")
+    max_val = float("0.5")
+    data = None
+
+
+class Program_weight_tensor_parameter_82:
+    name = "parameter_82"
+    shape = [64]
+    dtype = "float32"
+    min_val = float("0")
+    max_val = float("0.5")
+    data = None
+
+
+class Program_weight_tensor_parameter_83:
+    name = "parameter_83"
+    shape = [64]
+    dtype = "float32"
+    min_val = float("0")
+    max_val = float("0.5")
+    data = None
+
+
+class Program_weight_tensor_parameter_84:
+    name = "parameter_84"
+    shape = [64, 64, 3, 3]
+    dtype = "float32"
+    min_val = float("-0.378201")
+    max_val = float("0.338886")
+    mean = float("-0.00283112")
+    std = float("0.0472125")
+    data = None
+
+
+class Program_weight_tensor_parameter_85:
+    name = "parameter_85"
+    shape = [64]
+    dtype = "float32"
+    min_val = float("0")
+    max_val = float("0.5")
+    data = None
+
+
+class Program_weight_tensor_parameter_86:
+    name = "parameter_86"
+    shape = [64]
+    dtype = "float32"
+    min_val = float("0")
+    max_val = float("0.5")
+    data = None
+
+
+class Program_weight_tensor_parameter_87:
+    name = "parameter_87"
+    shape = [64]
+    dtype = "float32"
+    min_val = float("0")
+    max_val = float("0.5")
+    data = None
+
+
+class Program_weight_tensor_parameter_88:
+    name = "parameter_88"
+    shape = [64]
+    dtype = "float32"
+    min_val = float("0")
+    max_val = float("0.5")
+    data = None
+
+
+class Program_weight_tensor_parameter_89:
+    name = "parameter_89"
+    shape = [64, 64, 3, 3]
+    dtype = "float32"
+    min_val = float("-0.581769")
+    max_val = float("0.404201")
+    mean = float("-0.00159763")
+    std = float("0.0502428")
+    data = None
+
+
+class Program_weight_tensor_parameter_90:
+    name = "parameter_90"
+    shape = [64]
+    dtype = "float32"
+    min_val = float("0")
+    max_val = float("0.5")
+    data = None
+
+
+class Program_weight_tensor_parameter_91:
+    name = "parameter_91"
+    shape = [64]
+    dtype = "float32"
+    min_val = float("0")
+    max_val = float("0.5")
+    data = None
+
+
+class Program_weight_tensor_parameter_92:
+    name = "parameter_92"
+    shape = [64]
+    dtype = "float32"
+    min_val = float("0")
+    max_val = float("0.5")
+    data = None
+
+
+class Program_weight_tensor_parameter_93:
+    name = "parameter_93"
+    shape = [64]
+    dtype = "float32"
+    min_val = float("0")
+    max_val = float("0.5")
+    data = None
+
+
+class Program_weight_tensor_parameter_94:
+    name = "parameter_94"
+    shape = [64, 64, 3, 3]
+    dtype = "float32"
+    min_val = float("-0.78444")
+    max_val = float("0.668266")
+    mean = float("-0.00292819")
+    std = float("0.0603098")
+    data = None
+
+
+class Program_weight_tensor_parameter_95:
+    name = "parameter_95"
+    shape = [64]
+    dtype = "float32"
+    min_val = float("0")
+    max_val = float("0.5")
+    data = None
+
+
+class Program_weight_tensor_parameter_96:
+    name = "parameter_96"
+    shape = [64]
+    dtype = "float32"
+    min_val = float("0")
+    max_val = float("0.5")
+    data = None
+
+
+class Program_weight_tensor_parameter_97:
+    name = "parameter_97"
+    shape = [64]
+    dtype = "float32"
+    min_val = float("0")
+    max_val = float("0.5")
+    data = None
+
+
+class Program_weight_tensor_parameter_98:
+    name = "parameter_98"
+    shape = [64]
+    dtype = "float32"
+    min_val = float("0")
+    max_val = float("0.5")
+    data = None
+
+
+class Program_weight_tensor_parameter_99:
+    name = "parameter_99"
+    shape = [64, 3, 7, 7]
+    dtype = "float32"
+    min_val = float("-0.792412")
+    max_val = float("0.898735")
+    mean = float("-0.000543189")
+    std = float("0.133299")
+    data = None