From 636aef63e4a543826248f905306bbfb864bcfe0f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=E2=80=9Coguiza=E2=80=9D?= <“oguiza@gmail.com”>
Date: Fri, 21 Jan 2022 09:22:35 +0100
Subject: [PATCH] fixes #376

---
 nbs/016_data.preprocessing.ipynb | 254 ++++++++++++++++++++-----------
 nbs/017_data.transforms.ipynb    | 115 +++++++++-----
 tsai/_nbdev.py                   |   2 +
 tsai/data/preprocessing.py       |  24 ++-
 tsai/data/transforms.py          |  26 +++-
 5 files changed, 290 insertions(+), 131 deletions(-)

diff --git a/nbs/016_data.preprocessing.ipynb b/nbs/016_data.preprocessing.ipynb
index 316ed3fc9..fd84f73a5 100644
--- a/nbs/016_data.preprocessing.ipynb
+++ b/nbs/016_data.preprocessing.ipynb
@@ -770,7 +770,7 @@
     {
      "data": {
       "text/plain": [
-       "TSTensor([-2.2872467041015625], device=cpu)"
+       "TSTensor([-2.388400077819824], device=cpu)"
       ]
      },
      "execution_count": null,
@@ -1084,15 +1084,15 @@
     {
      "data": {
       "text/plain": [
-       "tensor([[[   nan, 0.4736, 0.3374,    nan, 0.3008, 0.3492,    nan, 0.1158],\n",
-       "         [   nan, 0.4824,    nan,    nan,    nan, 0.3499, 0.4910,    nan],\n",
-       "         [0.4806,    nan,    nan, 0.2372,    nan, 0.1333, 0.3234, 0.4244],\n",
-       "         [1.0000, 2.0000, 1.0000, 1.0000, 2.0000, 1.0000, 1.0000, 2.0000],\n",
-       "         [1.0000, 1.0000, 2.0000, 3.0000, 1.0000, 2.0000, 1.0000, 1.0000],\n",
-       "         [1.0000, 1.0000, 2.0000, 1.0000, 1.0000, 2.0000, 1.0000, 1.0000],\n",
-       "         [3.0000, 2.0000, 1.0000, 2.0000, 1.0000, 1.0000, 1.0000, 1.0000],\n",
+       "tensor([[[   nan, 0.2066, 0.4902,    nan, 0.3808, 0.2035, 0.2886, 0.1228],\n",
+       "         [   nan, 0.0387,    nan, 0.2696,    nan,    nan, 0.0113,    nan],\n",
+       "         [   nan, 0.0680, 0.3021, 0.0375, 0.4763,    nan,    nan,    nan],\n",
+       "         [1.0000, 2.0000, 1.0000, 1.0000, 2.0000, 1.0000, 1.0000, 1.0000],\n",
+       "         [1.0000, 2.0000, 1.0000, 1.0000, 1.0000, 1.0000, 2.0000, 3.0000],\n",
+       "         [1.0000, 1.0000, 2.0000, 1.0000, 1.0000, 1.0000, 1.0000, 1.0000],\n",
+       "         [1.0000, 1.0000, 1.0000, 1.0000, 4.0000, 3.0000, 2.0000, 1.0000],\n",
        "         [1.0000, 1.0000, 1.0000, 1.0000, 1.0000, 1.0000, 1.0000, 1.0000],\n",
-       "         [1.0000, 1.0000, 1.0000, 2.0000, 1.0000, 1.0000, 1.0000, 1.0000]]])"
+       "         [1.0000, 1.0000, 1.0000, 1.0000, 1.0000, 1.0000, 2.0000, 1.0000]]])"
       ]
      },
      "execution_count": null,
@@ -1154,17 +1154,17 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "tensor([[[   nan, 0.4732,    nan, 0.0619, 0.3361,    nan,    nan,    nan],\n",
-      "         [   nan,    nan,    nan,    nan, 0.1014, 0.0027, 0.2076, 0.1597],\n",
-      "         [   nan, 0.1548, 0.5667, 0.5099,    nan, 0.2899,    nan,    nan]]])\n",
-      "tensor([[[0.4732, 0.4732, 0.4732, 0.0619, 0.3361, 0.3361, 0.3361, 0.3361],\n",
-      "         [   nan,    nan,    nan,    nan, 0.1014, 0.0027, 0.2076, 0.1597],\n",
-      "         [0.1548, 0.1548, 0.5667, 0.5099, 0.5099, 0.2899, 0.2899, 0.2899],\n",
-      "         [0.4732, 0.4732, 0.4732, 0.3361, 0.2904, 0.2447, 0.3361, 0.3361],\n",
-      "         [0.1548, 0.1548, 0.2921, 0.4105, 0.5288, 0.4365, 0.3632, 0.2899]]])\n",
-      "tensor([[[0.4732, 0.4732, 0.4732, 0.3361, 0.2904, 0.2447, 0.3361, 0.3361],\n",
-      "         [0.1014, 0.1014, 0.1014, 0.1014, 0.1014, 0.0685, 0.1039, 0.1233],\n",
-      "         [0.1548, 0.1548, 0.2921, 0.4105, 0.5288, 0.4365, 0.3632, 0.2899]]])\n"
+      "tensor([[[   nan,    nan, 0.3284, 0.3478, 0.5412, 0.4566, 0.4781,    nan],\n",
+      "         [0.0451, 0.3167, 0.1816, 0.5600,    nan, 0.4331, 0.3259, 0.5339],\n",
+      "         [0.2252, 0.0482, 0.0429, 0.0438,    nan, 0.5943, 0.3788,    nan]]])\n",
+      "tensor([[[0.3284, 0.3284, 0.3284, 0.3478, 0.5412, 0.4566, 0.4781, 0.4781],\n",
+      "         [0.0451, 0.3167, 0.1816, 0.5600,    nan, 0.4331, 0.3259, 0.5339],\n",
+      "         [0.2252, 0.0482, 0.0429, 0.0438, 0.0438, 0.5943, 0.3788, 0.3788],\n",
+      "         [0.3284, 0.3284, 0.3284, 0.3349, 0.4058, 0.4486, 0.4920, 0.4709],\n",
+      "         [0.2252, 0.1367, 0.1054, 0.0450, 0.0435, 0.2273, 0.3390, 0.4506]]])\n",
+      "tensor([[[0.3284, 0.3284, 0.3284, 0.3349, 0.4058, 0.4486, 0.4920, 0.4709],\n",
+      "         [0.0451, 0.1809, 0.1811, 0.3528, 0.4339, 0.5177, 0.4397, 0.4309],\n",
+      "         [0.2252, 0.1367, 0.1054, 0.0450, 0.0435, 0.2273, 0.3390, 0.4506]]])\n"
      ]
     }
    ],
@@ -1238,6 +1238,43 @@
     "test_eq(TSAdd(1)(t), TSTensor([2,3,4]).float())"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#export\n",
+    "class TSClipByVar(Transform):\n",
+    "    \"\"\"Clip  batch of type `TSTensor` by variable\n",
+    "    \n",
+    "    Args:\n",
+    "        var_min_max: list of tuples containing variable index, min value (or None) and max value (or None)\n",
+    "    \"\"\"\n",
+    "    order = 90\n",
+    "    def __init__(self, var_min_max):\n",
+    "        self.var_min_max = var_min_max\n",
+    "\n",
+    "    def encodes(self, o:TSTensor):\n",
+    "        for v,m,M in self.var_min_max:\n",
+    "            o[:, v] = torch.clamp(o[:, v], m, M)\n",
+    "        return o"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "t = TSTensor(torch.rand(16, 3, 10) * tensor([1,10,100]).reshape(1,-1,1))\n",
+    "max_values = t.max(0).values.max(-1).values.data\n",
+    "max_values2 = TSClipByVar([(1,None,5), (2,10,50)])(t).max(0).values.max(-1).values.data\n",
+    "test_le(max_values2[1], 5)\n",
+    "test_ge(max_values2[2], 10)\n",
+    "test_le(max_values2[2], 50)"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},
@@ -1523,10 +1560,10 @@
        "      <th>0</th>\n",
        "      <td>2022</td>\n",
        "      <td>1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>11</td>\n",
-       "      <td>1</td>\n",
-       "      <td>11</td>\n",
+       "      <td>3</td>\n",
+       "      <td>21</td>\n",
+       "      <td>4</td>\n",
+       "      <td>21</td>\n",
        "      <td>False</td>\n",
        "      <td>False</td>\n",
        "      <td>False</td>\n",
@@ -1538,10 +1575,10 @@
        "      <th>1</th>\n",
        "      <td>2022</td>\n",
        "      <td>1</td>\n",
-       "      <td>2</td>\n",
-       "      <td>12</td>\n",
-       "      <td>2</td>\n",
-       "      <td>12</td>\n",
+       "      <td>3</td>\n",
+       "      <td>22</td>\n",
+       "      <td>5</td>\n",
+       "      <td>22</td>\n",
        "      <td>False</td>\n",
        "      <td>False</td>\n",
        "      <td>False</td>\n",
@@ -1555,8 +1592,8 @@
       ],
       "text/plain": [
        "   _Year  _Month  _Week  _Day  _Dayofweek  _Dayofyear  _Is_month_end  \\\n",
-       "0   2022       1      2    11           1          11          False   \n",
-       "1   2022       1      2    12           2          12          False   \n",
+       "0   2022       1      3    21           4          21          False   \n",
+       "1   2022       1      3    22           5          22          False   \n",
        "\n",
        "   _Is_month_start  _Is_quarter_end  _Is_quarter_start  _Is_year_end  \\\n",
        "0            False            False              False         False   \n",
@@ -1649,90 +1686,90 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
-       "      <td>0.265004</td>\n",
-       "      <td>0.025259</td>\n",
-       "      <td>0.186945</td>\n",
+       "      <td>0.740783</td>\n",
+       "      <td>0.288323</td>\n",
+       "      <td>0.310604</td>\n",
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
-       "      <td>0.032012</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>0.234934</td>\n",
+       "      <td>0.474600</td>\n",
+       "      <td>0.435126</td>\n",
+       "      <td>0.685560</td>\n",
+       "      <td>0</td>\n",
        "      <td>0</td>\n",
-       "      <td>1</td>\n",
        "      <td>0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
-       "      <td>0.113251</td>\n",
-       "      <td>0.575120</td>\n",
-       "      <td>0.695516</td>\n",
+       "      <td>0.071655</td>\n",
+       "      <td>0.332069</td>\n",
+       "      <td>0.082420</td>\n",
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
-       "      <td>0.087097</td>\n",
-       "      <td>0.422338</td>\n",
-       "      <td>0.777238</td>\n",
-       "      <td>0</td>\n",
+       "      <td>0.637574</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>0.201933</td>\n",
        "      <td>0</td>\n",
+       "      <td>1</td>\n",
        "      <td>0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
-       "      <td>0.563594</td>\n",
-       "      <td>0.648365</td>\n",
-       "      <td>0.294030</td>\n",
+       "      <td>0.455716</td>\n",
+       "      <td>0.311572</td>\n",
+       "      <td>0.403498</td>\n",
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>5</th>\n",
-       "      <td>0.087172</td>\n",
-       "      <td>0.156181</td>\n",
-       "      <td>0.312204</td>\n",
-       "      <td>0</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>0.410037</td>\n",
+       "      <td>0.427390</td>\n",
+       "      <td>1</td>\n",
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>6</th>\n",
-       "      <td>NaN</td>\n",
-       "      <td>0.127532</td>\n",
-       "      <td>0.445534</td>\n",
-       "      <td>1</td>\n",
+       "      <td>0.504213</td>\n",
+       "      <td>0.623972</td>\n",
+       "      <td>0.025626</td>\n",
+       "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>7</th>\n",
-       "      <td>0.795420</td>\n",
-       "      <td>0.177605</td>\n",
-       "      <td>NaN</td>\n",
+       "      <td>0.080880</td>\n",
+       "      <td>0.491392</td>\n",
+       "      <td>0.080737</td>\n",
+       "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>0</td>\n",
-       "      <td>1</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>8</th>\n",
-       "      <td>0.246201</td>\n",
-       "      <td>0.096138</td>\n",
        "      <td>NaN</td>\n",
+       "      <td>0.400076</td>\n",
+       "      <td>0.066079</td>\n",
+       "      <td>1</td>\n",
        "      <td>0</td>\n",
        "      <td>0</td>\n",
-       "      <td>1</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>9</th>\n",
-       "      <td>0.794408</td>\n",
-       "      <td>0.464993</td>\n",
-       "      <td>0.299659</td>\n",
+       "      <td>0.162308</td>\n",
+       "      <td>0.739196</td>\n",
+       "      <td>0.669122</td>\n",
        "      <td>0</td>\n",
        "      <td>0</td>\n",
        "      <td>0</td>\n",
@@ -1743,16 +1780,16 @@
       ],
       "text/plain": [
        "          a         b         c  a_missing  b_missing  c_missing\n",
-       "0  0.265004  0.025259  0.186945          0          0          0\n",
-       "1  0.032012       NaN  0.234934          0          1          0\n",
-       "2  0.113251  0.575120  0.695516          0          0          0\n",
-       "3  0.087097  0.422338  0.777238          0          0          0\n",
-       "4  0.563594  0.648365  0.294030          0          0          0\n",
-       "5  0.087172  0.156181  0.312204          0          0          0\n",
-       "6       NaN  0.127532  0.445534          1          0          0\n",
-       "7  0.795420  0.177605       NaN          0          0          1\n",
-       "8  0.246201  0.096138       NaN          0          0          1\n",
-       "9  0.794408  0.464993  0.299659          0          0          0"
+       "0  0.740783  0.288323  0.310604          0          0          0\n",
+       "1  0.474600  0.435126  0.685560          0          0          0\n",
+       "2  0.071655  0.332069  0.082420          0          0          0\n",
+       "3  0.637574       NaN  0.201933          0          1          0\n",
+       "4  0.455716  0.311572  0.403498          0          0          0\n",
+       "5       NaN  0.410037  0.427390          1          0          0\n",
+       "6  0.504213  0.623972  0.025626          0          0          0\n",
+       "7  0.080880  0.491392  0.080737          0          0          0\n",
+       "8       NaN  0.400076  0.066079          1          0          0\n",
+       "9  0.162308  0.739196  0.669122          0          0          0"
       ]
      },
      "execution_count": null,
@@ -1847,7 +1884,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -1892,7 +1929,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -1936,7 +1973,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -1980,7 +2017,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -2024,7 +2061,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -2069,7 +2106,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -2081,7 +2118,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -2140,10 +2177,10 @@
     {
      "data": {
       "text/plain": [
-       "(array(['a', 'd', 'd', 'd', 'c', 'b', 'e', 'e', 'a', 'd', 'a', 'c', 'b',\n",
-       "        'c', 'a', 'e', 'b', 'd', 'e', 'c'], dtype='<U1'),\n",
-       " array(['x', 'z', 'z', 'z', 'y', 'x', 'z', 'z', 'x', 'z', 'x', 'y', 'x',\n",
-       "        'y', 'x', 'z', 'x', 'z', 'z', 'y'], dtype='<U1'))"
+       "(array(['b', 'a', 'c', 'd', 'c', 'e', 'd', 'a', 'e', 'b', 'd', 'e', 'b',\n",
+       "        'c', 'a', 'c', 'e', 'd', 'a', 'e'], dtype='<U1'),\n",
+       " array(['x', 'x', 'y', 'z', 'y', 'z', 'z', 'x', 'z', 'x', 'z', 'z', 'x',\n",
+       "        'y', 'x', 'y', 'z', 'z', 'x', 'z'], dtype='<U1'))"
       ]
      },
      "execution_count": null,
@@ -2164,7 +2201,50 @@
    "cell_type": "code",
    "execution_count": null,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "data": {
+      "application/javascript": [
+       "IPython.notebook.save_checkpoint();"
+      ],
+      "text/plain": [
+       "<IPython.core.display.Javascript object>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "016_data.preprocessing.ipynb saved at 2022-01-21 08:50:28.\n",
+      "Converted 016_data.preprocessing.ipynb.\n",
+      "\n",
+      "\n",
+      "Correct conversion! 😃\n",
+      "Total time elapsed 0.103 s\n",
+      "Friday 21/01/22 08:50:33 CET\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "\n",
+       "                <audio  controls=\"controls\" autoplay=\"autoplay\">\n",
+       "                    <source src=\"data:audio/wav;base64,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\" type=\"audio/wav\" />\n",
+       "                    Your browser does not support the audio element.\n",
+       "                </audio>\n",
+       "              "
+      ],
+      "text/plain": [
+       "<IPython.lib.display.Audio object>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
    "source": [
     "#hide\n",
     "from tsai.imports import create_scripts\n",
diff --git a/nbs/017_data.transforms.ipynb b/nbs/017_data.transforms.ipynb
index 8a0b366a0..72cbfd65b 100644
--- a/nbs/017_data.transforms.ipynb
+++ b/nbs/017_data.transforms.ipynb
@@ -158,7 +158,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -1087,17 +1087,7 @@
    "cell_type": "code",
    "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/Users/nacho/opt/anaconda3/envs/py37/lib/python3.7/site-packages/torch/_tensor.py:579: UserWarning: floor_divide is deprecated, and will be removed in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values.\n",
-      "To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor'). (Triggered internally at  ../aten/src/ATen/native/BinaryOps.cpp:467.)\n",
-      "  return torch.floor_divide(other, self)\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "test_eq(TSMaskOut()(xb, split_idx=0).shape, xb.shape)\n",
     "test_ne(TSMaskOut()(xb, split_idx=0), xb)"
@@ -1457,7 +1447,6 @@
    "outputs": [],
    "source": [
     "#export\n",
-    "\n",
     "class TSRandomConv(RandTransform):\n",
     "    \"\"\"Applies a convolution with a random kernel and random weights with required_grad=False\"\"\"\n",
     "    order = 90\n",
@@ -1489,6 +1478,75 @@
     "    test_eq(o.shape, xb.shape)"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#export\n",
+    "from fastai.vision.augment import RandTransform\n",
+    "class TSAddNan(RandTransform):\n",
+    "    \"Randomly sets selected variables of type `TSTensor` to Nan values\"\n",
+    "    order = 90\n",
+    "    def __init__(self, nan_perc=0.1, sel_vars=None, static=False, **kwargs):\n",
+    "        self.sel_vars = sel_vars if sel_vars is not None else None\n",
+    "        self.nan_perc = nan_perc\n",
+    "        self.static = static\n",
+    "        super().__init__(**kwargs)\n",
+    "\n",
+    "    def encodes(self, o:TSTensor):\n",
+    "        if self.static:\n",
+    "            nan_vals = torch.rand(*o.shape[:-1])\n",
+    "        else:\n",
+    "            nan_vals = torch.rand(*o.shape)\n",
+    "        if self.sel_vars is not None:\n",
+    "            nan_vals[:, ~torch.isin(torch.arange(o.shape[1]), tensor(self.sel_vars))] = 0\n",
+    "        o[nan_vals > 1 - self.nan_perc] = np.nan\n",
+    "        return o"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "tensor([[[1., nan, 1., 1.],\n",
+       "         [1., 1., 1., 1.],\n",
+       "         [1., nan, 1., 1.]],\n",
+       "\n",
+       "        [[nan, nan, 1., nan],\n",
+       "         [1., 1., 1., 1.],\n",
+       "         [nan, nan, nan, nan]]])"
+      ]
+     },
+     "execution_count": null,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "t = TSTensor(torch.ones(2,3,4))\n",
+    "TSAddNan(nan_perc=.5, sel_vars=[0,2])(t, split_idx=0).data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "t = TSTensor(torch.ones(2,3,100))\n",
+    "test_gt(np.isnan(TSAddNan(nan_perc=.5)(t, split_idx=0)).sum().item(), 0)\n",
+    "t = TSTensor(torch.ones(2,3,100))\n",
+    "test_gt(np.isnan(TSAddNan(nan_perc=.5, sel_vars=[0,2])(t, split_idx=0)[:, [0,2]]).sum().item(), 0)\n",
+    "t = TSTensor(torch.ones(2,3,100))\n",
+    "test_eq(np.isnan(TSAddNan(nan_perc=.5, sel_vars=[0,2])(t, split_idx=0)[:, 1]).sum().item(), 0)"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,
@@ -1649,29 +1707,11 @@
    "outputs": [
     {
      "data": {
-      "text/html": [
-       "<img src onerror=\"\n",
-       "        this.nextElementSibling.focus();\n",
-       "        this.dispatchEvent(new KeyboardEvent('keydown', {key:'s', keyCode: 83, metaKey: true}));\n",
-       "        \" style=\"display:none\"><input style=\"width:0;height:0;border:0\">"
-      ],
-      "text/plain": [
-       "<IPython.core.display.HTML object>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "<img src onerror=\"\n",
-       "        this.nextElementSibling.focus();\n",
-       "        this.dispatchEvent(new KeyboardEvent('keydown', {key:'s', keyCode: 83, metaKey: true}));\n",
-       "        \" style=\"display:none\"><input style=\"width:0;height:0;border:0\">"
+      "application/javascript": [
+       "IPython.notebook.save_checkpoint();"
       ],
       "text/plain": [
-       "<IPython.core.display.HTML object>"
+       "<IPython.core.display.Javascript object>"
       ]
      },
      "metadata": {},
@@ -1681,12 +1721,13 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
+      "017_data.transforms.ipynb saved at 2022-01-21 09:07:58.\n",
       "Converted 017_data.transforms.ipynb.\n",
       "\n",
       "\n",
       "Correct conversion! 😃\n",
-      "Total time elapsed 0.647 s\n",
-      "Thursday 30/09/21 20:55:29 CEST\n"
+      "Total time elapsed 0.119 s\n",
+      "Friday 21/01/22 09:08:03 CET\n"
      ]
     },
     {
@@ -1718,7 +1759,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   }
diff --git a/tsai/_nbdev.py b/tsai/_nbdev.py
index ea46479ad..5eedf6212 100644
--- a/tsai/_nbdev.py
+++ b/tsai/_nbdev.py
@@ -277,6 +277,7 @@
          "TSRollingMean": "016_data.preprocessing.ipynb",
          "TSLogReturn": "016_data.preprocessing.ipynb",
          "TSAdd": "016_data.preprocessing.ipynb",
+         "TSClipByVar": "016_data.preprocessing.ipynb",
          "TSShrinkDataFrame": "016_data.preprocessing.ipynb",
          "TSOneHotEncoder": "016_data.preprocessing.ipynb",
          "TSCategoricalEncoder": "016_data.preprocessing.ipynb",
@@ -340,6 +341,7 @@
          "TSDownUpScale": "017_data.transforms.ipynb",
          "TSRandomDownUpScale": "017_data.transforms.ipynb",
          "TSRandomConv": "017_data.transforms.ipynb",
+         "TSAddNan": "017_data.transforms.ipynb",
          "all_TS_randaugs": "017_data.transforms.ipynb",
          "RandAugment": "017_data.transforms.ipynb",
          "TestTfm": "017_data.transforms.ipynb",
diff --git a/tsai/data/preprocessing.py b/tsai/data/preprocessing.py
index 8ad6bcddc..54b681e33 100644
--- a/tsai/data/preprocessing.py
+++ b/tsai/data/preprocessing.py
@@ -2,10 +2,10 @@
 
 __all__ = ['ToNumpyCategory', 'OneHot', 'TSNan2Value', 'Nan2Value', 'TSStandardize', 'TSNormalize', 'TSClipOutliers',
            'TSClip', 'TSRobustScale', 'TSDiff', 'TSLog', 'TSCyclicalPosition', 'TSLinearPosition', 'TSPosition',
-           'TSMissingness', 'TSPositionGaps', 'TSRollingMean', 'TSLogReturn', 'TSAdd', 'TSShrinkDataFrame',
-           'TSOneHotEncoder', 'TSCategoricalEncoder', 'TSDateTimeEncoder', 'default_date_attr', 'TSMissingnessEncoder',
-           'Preprocessor', 'StandardScaler', 'RobustScaler', 'Normalizer', 'BoxCox', 'YeoJohnshon', 'Quantile',
-           'ReLabeler']
+           'TSMissingness', 'TSPositionGaps', 'TSRollingMean', 'TSLogReturn', 'TSAdd', 'TSClipByVar',
+           'TSShrinkDataFrame', 'TSOneHotEncoder', 'TSCategoricalEncoder', 'TSDateTimeEncoder', 'default_date_attr',
+           'TSMissingnessEncoder', 'Preprocessor', 'StandardScaler', 'RobustScaler', 'Normalizer', 'BoxCox',
+           'YeoJohnshon', 'Quantile', 'ReLabeler']
 
 # Cell
 from ..imports import *
@@ -546,6 +546,22 @@ def encodes(self, o:TSTensor):
         return torch.add(o, self.add)
     def __repr__(self): return f'{self.__class__.__name__}(lag={self.lag}, pad={self.pad})'
 
+# Cell
+class TSClipByVar(Transform):
+    """Clip  batch of type `TSTensor` by variable
+
+    Args:
+        var_min_max: list of tuples containing variable index, min value (or None) and max value (or None)
+    """
+    order = 90
+    def __init__(self, var_min_max):
+        self.var_min_max = var_min_max
+
+    def encodes(self, o:TSTensor):
+        for v,m,M in self.var_min_max:
+            o[:, v] = torch.clamp(o[:, v], m, M)
+        return o
+
 # Cell
 from sklearn.base import BaseEstimator, TransformerMixin
 from fastai.data.transforms import CategoryMap
diff --git a/tsai/data/transforms.py b/tsai/data/transforms.py
index 5b077e44f..c8530e7bf 100644
--- a/tsai/data/transforms.py
+++ b/tsai/data/transforms.py
@@ -8,8 +8,8 @@
            'TSRandomFreqNoise', 'TSRandomResizedLookBack', 'TSRandomLookBackOut', 'TSVarOut', 'TSCutOut',
            'TSTimeStepOut', 'TSRandomCropPad', 'TSMaskOut', 'TSInputDropout', 'TSTranslateX', 'TSRandomShift',
            'TSHorizontalFlip', 'TSRandomTrend', 'TSRandomRotate', 'TSVerticalFlip', 'TSResize', 'TSRandomSize',
-           'TSRandomLowRes', 'TSDownUpScale', 'TSRandomDownUpScale', 'TSRandomConv', 'all_TS_randaugs', 'RandAugment',
-           'TestTfm', 'get_tfm_name']
+           'TSRandomLowRes', 'TSDownUpScale', 'TSRandomDownUpScale', 'TSRandomConv', 'TSAddNan', 'all_TS_randaugs',
+           'RandAugment', 'TestTfm', 'get_tfm_name']
 
 # Cell
 from ..imports import *
@@ -763,7 +763,6 @@ def encodes(self, o: TSTensor):
         return output
 
 # Cell
-
 class TSRandomConv(RandTransform):
     """Applies a convolution with a random kernel and random weights with required_grad=False"""
     order = 90
@@ -783,6 +782,27 @@ def encodes(self, o: TSTensor):
         if self.ex is not None: output[...,self.ex,:] = o[...,self.ex,:]
         return output
 
+# Cell
+from fastai.vision.augment import RandTransform
+class TSAddNan(RandTransform):
+    "Randomly sets selected variables of type `TSTensor` to Nan values"
+    order = 90
+    def __init__(self, nan_perc=0.1, sel_vars=None, static=False, **kwargs):
+        self.sel_vars = sel_vars if sel_vars is not None else None
+        self.nan_perc = nan_perc
+        self.static = static
+        super().__init__(**kwargs)
+
+    def encodes(self, o:TSTensor):
+        if self.static:
+            nan_vals = torch.rand(*o.shape[:-1])
+        else:
+            nan_vals = torch.rand(*o.shape)
+        if self.sel_vars is not None:
+            nan_vals[:, ~torch.isin(torch.arange(o.shape[1]), tensor(self.sel_vars))] = 0
+        o[nan_vals > 1 - self.nan_perc] = np.nan
+        return o
+
 # Cell
 all_TS_randaugs = [