♻️ set default cache directory as .cache

.gitignore

@@ -1,3 +1 @@
-ratings_test.txt
-ratings_train.txt
-*.egg-info
+.cache

kobert/mxnet_kobert.py

@@ -32,7 +32,7 @@ def get_mxnet_kobert_model(
     use_decoder=True,
     use_classifier=True,
     ctx=mx.cpu(0),
-    cachedir="~/kobert/",
+    cachedir=".cache",
 ):
     # download model
     model_info = mxnet_kobert

kobert/pytorch_kobert.py

@@ -31,7 +31,7 @@
 }
 
 
-def get_pytorch_kobert_model(ctx="cpu", cachedir="~/kobert/"):
+def get_pytorch_kobert_model(ctx="cpu", cachedir=".cache"):
     # download model
     model_info = pytorch_kobert
     model_down = _download(

kobert/utils.py

@@ -31,7 +31,7 @@
 }
 
 
-def download(url, filename, chksum, cachedir="~/kobert/"):
+def download(url, filename, chksum, cachedir=".cache"):
     f_cachedir = os.path.expanduser(cachedir)
     os.makedirs(f_cachedir, exist_ok=True)
     file_path = os.path.join(f_cachedir, filename)
@@ -54,7 +54,7 @@ def download(url, filename, chksum, cachedir="~/kobert/"):
                 downloaded += len(data)
                 f.write(data)
                 done = int(50 * downloaded / total)
-                sys.stdout.write("\r[{}{}]".format("█" * done, "." * (50 - done)))
+                sys.stdout.write("\r{}[{}{}]".format(file_path, "█" * done, "." * (50 - done)))
                 sys.stdout.flush()
     sys.stdout.write("\n")
     assert (
@@ -63,15 +63,15 @@ def download(url, filename, chksum, cachedir="~/kobert/"):
     return file_path
 
 
-def get_onnx(cachedir="~/kobert/"):
+def get_onnx(cachedir=".cache"):
     """Get KoBERT ONNX file path after downloading"""
     model_info = onnx_kobert
     return download(
         model_info["url"], model_info["fname"], model_info["chksum"], cachedir=cachedir
     )
 
 
-def get_tokenizer(cachedir="~/kobert/"):
+def get_tokenizer(cachedir=".cache"):
     """Get KoBERT Tokenizer file path after downloading"""
     model_info = tokenizer
     return download(

scripts/NSMC/naver_review_classifications_gluon_kobert.ipynb

@@ -84,7 +84,7 @@
    },
    "outputs": [],
    "source": [
-    "bert_base, vocab = get_mxnet_kobert_model(use_decoder=False, use_classifier=False, ctx=ctx)"
+    "bert_base, vocab = get_mxnet_kobert_model(use_decoder=False, use_classifier=False, ctx=ctx, cachedir=\".cache\")"
    ]
   },
   {
@@ -146,8 +146,8 @@
    },
    "outputs": [],
    "source": [
-    "!wget -O ratings_train.txt https://www.dropbox.com/s/374ftkec978br3d/ratings_train.txt?dl=1\n",
-    "!wget -O ratings_test.txt https://www.dropbox.com/s/977gbwh542gdy94/ratings_test.txt?dl=1"
+    "!wget -O .cache/ratings_train.txt https://www.dropbox.com/s/374ftkec978br3d/ratings_train.txt?dl=1\n",
+    "!wget -O .cache/ratings_test.txt https://www.dropbox.com/s/977gbwh542gdy94/ratings_test.txt?dl=1"
    ]
   },
   {
@@ -160,8 +160,8 @@
    },
    "outputs": [],
    "source": [
-    "dataset_train = nlp.data.TSVDataset(\"ratings_train.txt\", field_indices=[1,2], num_discard_samples=1)\n",
-    "dataset_test = nlp.data.TSVDataset(\"ratings_test.txt\", field_indices=[1,2], num_discard_samples=1)"
+    "dataset_train = nlp.data.TSVDataset(\".cache/ratings_train.txt\", field_indices=[1,2], num_discard_samples=1)\n",
+    "dataset_test = nlp.data.TSVDataset(\".cache/ratings_test.txt\", field_indices=[1,2], num_discard_samples=1)"
    ]
   },
   {
@@ -469,7 +469,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.5"
+   "version": "3.7.0"
   },
   "toc": {
    "nav_menu": {},

scripts/NSMC/naver_review_classifications_pytorch_kobert.ipynb

@@ -72,7 +72,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "bertmodel, vocab = get_pytorch_kobert_model()"
+    "bertmodel, vocab = get_pytorch_kobert_model(cachedir=\".cache\")"
    ]
   },
   {
@@ -81,8 +81,8 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "!wget -O ratings_train.txt https://www.dropbox.com/s/374ftkec978br3d/ratings_train.txt?dl=1\n",
-    "!wget -O ratings_test.txt https://www.dropbox.com/s/977gbwh542gdy94/ratings_test.txt?dl=1"
+    "!wget -O .cache/ratings_train.txt https://www.dropbox.com/s/374ftkec978br3d/ratings_train.txt?dl=1\n",
+    "!wget -O .cache/ratings_test.txt https://www.dropbox.com/s/977gbwh542gdy94/ratings_test.txt?dl=1"
    ]
   },
   {
@@ -91,8 +91,8 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "dataset_train = nlp.data.TSVDataset(\"ratings_train.txt\", field_indices=[1,2], num_discard_samples=1)\n",
-    "dataset_test = nlp.data.TSVDataset(\"ratings_test.txt\", field_indices=[1,2], num_discard_samples=1)"
+    "dataset_train = nlp.data.TSVDataset(\".cache/ratings_train.txt\", field_indices=[1,2], num_discard_samples=1)\n",
+    "dataset_test = nlp.data.TSVDataset(\".cache/ratings_test.txt\", field_indices=[1,2], num_discard_samples=1)"
    ]
   },
   {