Completes UNB-2546 - Don't download dataset if it already exists locally

Author: gustavocidornelas

examples/text-classification/fasttext/fasttext.ipynb

@@ -47,7 +47,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 2,
    "id": "42de6fd6",
    "metadata": {},
    "outputs": [],
@@ -66,9 +66,6 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current\n",
-      "                                 Dload  Upload   Total   Spent    Left  Speed\n",
-      "100  446k  100  446k    0     0   194k      0  0:00:02  0:00:02 --:--:--  194k\n",
       "x cooking.stackexchange.id\n",
       "x cooking.stackexchange.txt\n",
       "x readme.txt\n"
@@ -82,7 +79,11 @@
     "    mkdir ./data\n",
     "fi\n",
     "\n",
-    "curl https://dl.fbaipublicfiles.com/fasttext/data/cooking.stackexchange.tar.gz --output data/cooking.stackexchange.tar.gz && tar xvzf data/cooking.stackexchange.tar.gz -C data\n",
+    "if [ ! -f data/cooking.stackexchange.tar.gz ]; then\n",
+    "    curl https://dl.fbaipublicfiles.com/fasttext/data/cooking.stackexchange.tar.gz --output data/cooking.stackexchange.tar.gz \n",
+    "fi \n",
+    "\n",
+    "tar xvzf data/cooking.stackexchange.tar.gz -C data\n",
     "head -n 12404 data/cooking.stackexchange.txt > data/cooking.train\n",
     "tail -n 3000 data/cooking.stackexchange.txt > data/cooking.valid"
    ]
@@ -108,7 +109,7 @@
       "Read 0M words\n",
       "Number of words:  14543\n",
       "Number of labels: 735\n",
-      "Progress: 100.0% words/sec/thread: 1863090 lr:  0.000000 avg.loss:  5.498450 ETA:   0h 0m 0s\n"
+      "Progress: 100.0% words/sec/thread: 1836491 lr:  0.000000 avg.loss:  5.229637 ETA:   0h 0m 0s-0.000011 avg.loss:  5.229637 ETA:   0h 0m 0s\n"
      ]
     }
    ],
@@ -125,7 +126,7 @@
     {
      "data": {
       "text/plain": [
-       "(3000, 0.49633333333333335, 0.2146461006198645)"
+       "(3000, 0.5046666666666667, 0.2182499639613666)"
       ]
      },
      "execution_count": 6,