diff --git a/src/languedoc/__init__.py b/src/languedoc/__init__.py
new file mode 100644
--- /dev/null
+++ b/src/languedoc/__init__.py
@@ -0,0 +1,1 @@
+from .predict import identify
diff --git a/shared.py b/src/languedoc/predict.py
rename from shared.py
rename to src/languedoc/predict.py
--- a/shared.py
+++ b/src/languedoc/predict.py
@@ -5,7 +5,7 @@ import json
 import gzip
 
 TOP_NGRAM_COUNT = 3000
-MODEL_PATH = os.path.join(os.path.dirname(__file__), "models.json.gz")
+MODEL_PATH = os.path.join(os.path.dirname(__file__), "../../models.json.gz")
 
 
 def preprocess(text):
diff --git a/languedoc.py b/src/languedoc/train.py
rename from languedoc.py
rename to src/languedoc/train.py
--- a/languedoc.py
+++ b/src/languedoc/train.py
@@ -4,7 +4,7 @@ import itertools
 import json
 import gzip
 
-from shared import preprocess, identify, extract_ngram_freqs, rank_ngram_freqs, Sample
+from predict import preprocess, identify, extract_ngram_freqs, rank_ngram_freqs, Sample
 
 random.seed(19181028)
 
@@ -72,9 +72,9 @@ def cross_validate(sample_sets):
 	return score / max_score, (score, max_score)
 
 
-DATA_DIR = os.path.join(os.path.dirname(__file__), "data")
+DATA_DIR = os.path.join(os.path.dirname(__file__), "../../data")
 LANG_DIRS = sorted([x.path for x in os.scandir(DATA_DIR)])
-MODEL_PATH = os.path.join(os.path.dirname(__file__), "models.json.gz")
+MODEL_PATH = os.path.join(os.path.dirname(__file__), "../../models.json.gz")
 
 if __name__ == "__main__":
 	samples = []