From 0e864fb29140a765ee20bbe7ea8afaec9df25d88 Mon Sep 17 00:00:00 2001
From: Tomasz Pietruszka <tomek.pietruszka@gmail.com>
Date: Tue, 22 Jan 2019 20:10:38 +0100
Subject: [PATCH] test_end_to_end now working, added instructions to README,
 sentencepiece dependency

---
 README.md                | 10 ++++++++++
 prepare_imdb.sh          |  0
 requirements.txt         |  3 ++-
 tests/test_end_to_end.py |  3 ++-
 4 files changed, 14 insertions(+), 2 deletions(-)
 mode change 100644 => 100755 prepare_imdb.sh

diff --git a/README.md b/README.md
index a04fca8..41cd8b6 100644
--- a/README.md
+++ b/README.md
@@ -88,3 +88,13 @@ $ git push --set-upstream n-waves ulmfit_multilingual  # to automatically push u
     - `bilm` -- scripts to train biLM ELMo style, Bert style
     - `class`  -- scripts to test classifiers on multiple languages
     - `xnli` -- scripts to test nli 
+
+
+## Running tests
+
+To run the tests, the following data is necessary:
+
+- wikitext-2 (prepared by `./prepare_wiki-en.sh`, along with wikitext-103)
+- imdb (prepared by `./prepare_imdb.sh`)
+
+then simply run tests, e.g. `pytest .`
diff --git a/prepare_imdb.sh b/prepare_imdb.sh
old mode 100644
new mode 100755
diff --git a/requirements.txt b/requirements.txt
index eb003e9..b211a96 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,4 +1,5 @@
 fire>=0.1.3
 cupy>=5.0.0
 scikit-learn>=0.20
-sacremoses>=0.0.5
\ No newline at end of file
+sacremoses>=0.0.5
+sentencepiece
diff --git a/tests/test_end_to_end.py b/tests/test_end_to_end.py
index 3d4d112..266bc3f 100644
--- a/tests/test_end_to_end.py
+++ b/tests/test_end_to_end.py
@@ -25,7 +25,8 @@ def get_test_data():
     imdb = data / "imdb"
 
     test_data = data / "test"
-    shutil.rmtree(test_data)
+    if test_data.exists():
+        shutil.rmtree(test_data)
 
     test_wt = test_data / 'wikitext-s'
     test_imdb = test_data / 'imdb'