diff --git a/blueoil/converter/templates/src/func/average_pool.cpp b/blueoil/converter/templates/src/func/average_pool.cpp
index 2aa8943a7..291f4ff21 100644
--- a/blueoil/converter/templates/src/func/average_pool.cpp
+++ b/blueoil/converter/templates/src/func/average_pool.cpp
@@ -25,7 +25,7 @@ void func_AveragePool(const TensorView<T_FLOAT, MemoryLayout::NHWC>& input,
     struct avg_pooling_parameters app) {
   Measurement::Start("AveragePool");
 
-  assert (app.kernel_depth == 1 && "kernel depth 1 is not supported.");
+  assert (app.kernel_depth == 1 && "kernel depth != 1 is not supported.");
   assert (app.input_depth == app.kernel_depth * app.output_channels && \
           "input_depth must equal kernel_depth * output_channels.");
 
diff --git a/blueoil/converter/templates/src/func/max_pool.cpp b/blueoil/converter/templates/src/func/max_pool.cpp
index 62f51c686..167da64d9 100644
--- a/blueoil/converter/templates/src/func/max_pool.cpp
+++ b/blueoil/converter/templates/src/func/max_pool.cpp
@@ -27,7 +27,7 @@ void max_pooling(
     struct max_pooling_parameters p)
 {
 
-  assert (p.kernel_depth == 1 && "kernel depth 1 is not supported.");
+  assert (p.kernel_depth == 1 && "kernel depth != 1 is not supported.");
   assert (p.input_depth == p.kernel_depth * p.output_channels && \
           "input_depth must equal kernel_depth * output_channels.");
 
diff --git a/docs/converter/supported_ops.md b/docs/converter/supported_ops.md
index fad38c7c0..6a7fcde0b 100644
--- a/docs/converter/supported_ops.md
+++ b/docs/converter/supported_ops.md
@@ -13,10 +13,9 @@
 ### Tensorflow Ops with Limitations
 - **[tf.layers.AveragePooling2D](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/layers/AveragePooling2D)**
     - Currently, support only `2D`.
-    - Do ***not*** support `kernel depth = 1`.
+    - Do ***not*** support `kernel depth != 1`.
 - **[tf.concat](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/concat)**
     - Do ***not*** support concat of mixed data types (e.g., quantized values and float values).
-    - All tensor channels must be equal. 
     - If inputs are quantized, requires `Each input channel size = multiple of 32`.
 - **[tf.layers.Conv2D](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/layers/Conv2D)**
     - Support only convolution `2D`.
@@ -32,13 +31,13 @@
 - **[tf.linalg.matmul](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/linalg/matmul)**
     - Do ***not*** support `scalar`.
 - **[tf.layers.max_pooling2d](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/layers/max_pooling2d)**
-     - Currently, support only `2D`.
+    - Currently, support only `2D`.
+    - Do ***not*** support `kernel depth != 1`.
 - **[tf.pad](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/pad)**
     - Supports only `channel-wise paddings`.
 - **[tf.nn.space_to_depth](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/nn/space_to_depth)**
     - For quantized tensor, requires `output depth = (multiple of block_size^2 * 32)` or `(block_size^2 * {8, 16})`.
 - **[tf.split](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/split)**
-    - Currently, all of output tensors must have `same` shape.
     - For quantized tensor, requires `number of channel of each output tensor = multiple of 32`.
 
 ###  Tensorflow Ops without Limitations