diff --git a/CHANGES.txt b/CHANGES.txt
index 4891b858..8abcc776 100644
--- a/CHANGES.txt
+++ b/CHANGES.txt
@@ -1,6 +1,8 @@
-NEXT (TBD)
+2.0a5 (2020-05-06)
 ------------------
-- make `rio_tiler.io.landsat8.tile` return Uint16 data and not float32
+- make `rio_tiler.io.landsat8.tile` return Uint16 data and not float32 (#173)
+- `rio_tiler.profiles.img_profiles` item access return `copy` of the items (#177)
+- better colormap docs (#176, author @kylebarron)
 
 2.0a4 (2020-04-08)
 ------------------
diff --git a/README.md b/README.md
index 7d300435..3e5910d4 100644
--- a/README.md
+++ b/README.md
@@ -146,7 +146,7 @@ landsat8.metadata('LC08_L1TP_016037_20170813_20170814_01_RT', pmin=5, pmax=95)
 The primary purpose for calculating minimum and maximum values of an image is to rescale pixel values from their original range (e.g. 0 to 65,535) to the range used by computer screens (i.e. 0 and 255) through a linear transformation.
 This will make images look good on display.
 
-#### Working SpatioTemporal Asset Catalog (STAC) 
+#### Working with SpatioTemporal Asset Catalog (STAC) 
 
 In rio-tiler v2, we added a `rio_tiler.io.stac` submodule to allow tile/metadata fetching of assets withing a STAC item.
 
diff --git a/docs/v2_migration.md b/docs/v2_migration.md
index 5572e54f..9f44fc0c 100644
--- a/docs/v2_migration.md
+++ b/docs/v2_migration.md
@@ -68,7 +68,7 @@ In *rio_tiler==1* most of the magic was happening in [`rio_tiler.utils._tile_rea
 
 To ease the transition we added a `rio_tiler.reader.tile` function.
 
-Note: The new `rio_tiler.reader.part` function enables to perform non-squared data cropping of different.
+Note: The new `rio_tiler.reader.part` function enables to perform non-squared data cropping by passing output width and height (instead of just tilesize).
 
 ```python
 # v1
diff --git a/setup.py b/setup.py
index 1491a32b..df39fdf2 100644
--- a/setup.py
+++ b/setup.py
@@ -22,7 +22,7 @@
 
 setup(
     name="rio-tiler",
-    version="2.0a4",
+    version="2.0a5",
     python_requires=">=3",
     description=u"""Get mercator tile from CloudOptimized GeoTIFF and other cloud hosted raster such as CBERS-4, Sentinel-2, Sentinel-1 and Landsat-8 AWS PDS""",
     long_description=readme,