✅ ⏪ Revert rasterio to 1.0.13 to fix unit tests

Revert #90 to fix broken unit tests on the get_window_bounds and selective_tile functions in data_prep.ipynb. I've tracked it down to an incorrect affine output from dataset.transform. Seems to be a 'bug' introduced in rasterio version 1.0.14 (see #84), something with how the GDAL env is now set? Also smallfix to remove read_table deprecation warning in pandas 0.24 (see #89), replacing with read_csv instead.
weiji14 · Feb 11, 2019 · 22a1fc2 · 22a1fc2
1 parent 73ad018
commit 22a1fc2
Show file tree

Hide file tree

Showing 4 changed files with 54 additions and 54 deletions.
diff --git a/Pipfile b/Pipfile
@@ -24,7 +24,7 @@ packaging = "==19.0"
 pandas = "==0.24.1"
 pyproj = "==1.9.6"
 quilt = "==2.9.14"
-rasterio = "==1.0.16"
+rasterio = "==1.0.13"
 requests = "==2.21.0"
 scikit-image = "==0.14.2"
 shapely = "==1.7a1"

diff --git a/Pipfile.lock b/Pipfile.lock
diff --git a/data_prep.ipynb b/data_prep.ipynb
@@ -29,7 +29,7 @@
       "Geopandas    : 0.4.0+26.g9e584cc\n",
       "GMT          : 0.0.1a0+16.g7004aa0\n",
       "Numpy        : 1.14.5\n",
-      "Rasterio     : 1.0.16\n",
+      "Rasterio     : 1.0.13\n",
       "Scikit-image : 0.14.2\n",
       "Xarray       : 0.11.3\n"
      ]
@@ -622,7 +622,7 @@
     "    assert len(files) > 0  # check that there are actually files being matched!\n",
     "\n",
     "    df = pd.concat(\n",
-    "        pd.read_table(f, sep=sep, header=skip, names=names, usecols=usecols)\n",
+    "        pd.read_csv(f, sep=sep, header=skip, names=names, usecols=usecols)\n",
     "        for f in files\n",
     "    )\n",
     "    df.reset_index(drop=True, inplace=True)  # reset index after concatenation\n",

diff --git a/data_prep.py b/data_prep.py
@@ -265,7 +265,7 @@ def ascii_to_xyz(pipeline_file: str) -> pd.DataFrame:
     assert len(files) > 0  # check that there are actually files being matched!
 
     df = pd.concat(
-        pd.read_table(f, sep=sep, header=skip, names=names, usecols=usecols)
+        pd.read_csv(f, sep=sep, header=skip, names=names, usecols=usecols)
         for f in files
     )
     df.reset_index(drop=True, inplace=True)  # reset index after concatenation