From b8ffe341e677c84f84836aedb47fc8307a9b15c2 Mon Sep 17 00:00:00 2001
From: hsharif3 <hsharif3@illinois.edu>
Date: Thu, 25 Mar 2021 07:38:00 +0000
Subject: [PATCH] Update Support.md
---
hpvm/projects/keras/docs/Support.md | 19 ++++++++++++++++---
1 file changed, 16 insertions(+), 3 deletions(-)
diff --git a/hpvm/projects/keras/docs/Support.md b/hpvm/projects/keras/docs/Support.md
index e5e7b1a1a2..ba4dafcb92 100644
--- a/hpvm/projects/keras/docs/Support.md
+++ b/hpvm/projects/keras/docs/Support.md
@@ -1,4 +1,3 @@
-
## Supported Keras Operators
The Keras frontend supports `Sequential()` Keras models.
@@ -27,14 +26,28 @@ The list of supported operations is as follows:
* Softmax operator should be the last operation in the CNN pipeline
* Softmax operation must be a separate operator (not specified as activation to another type of Keras operator). Example of what works:
-```
+```python
Activation ("softmax")
```
Example of what is NOT supported:
-```
+```python
Dense(num_classes, activation="softmax")
```
+* For convolutions with stride > 1 `same` convolution is NOT supported. Explicitly add `ZeroPadding2D` layer before `Conv2D` or `DepthwiseConv2D` operators. Example of what does NOT work:
+
+```python
+Conv2D(num_filters, (3,3), strides = (2,2), padding = `same`)
+```
+
+Example of what works instead:
+
+```python
+ZeroPadding2D(padding=(1,1), data_format = `channels_first`) # only support NCHW
+Conv2D(num_filters, (3,3), strides = (2,2), padding = `valid`)
+```
+
+
--
GitLab