add missing output in bonus

ch02/03_bonus_embedding-vs-matmul/embeddings-and-linear-layers.ipynb

@@ -19,7 +19,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 1,
    "id": "061720f4-f025-4640-82a0-15098fa94cf9",
    "metadata": {},
    "outputs": [
@@ -27,7 +27,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "PyTorch version: 2.1.0.dev20230825\n"
+      "PyTorch version: 2.1.0\n"
      ]
     }
    ],
@@ -47,7 +47,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 2,
    "id": "cc489ea5-73db-40b9-959e-0d70cae25f40",
    "metadata": {},
    "outputs": [],
@@ -76,7 +76,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 3,
    "id": "60a7c104-36e1-4b28-bd02-a24a1099dc66",
    "metadata": {},
    "outputs": [],
@@ -99,7 +99,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 4,
    "id": "595f603e-8d2a-4171-8f94-eac8106b2e57",
    "metadata": {},
    "outputs": [
@@ -113,7 +113,7 @@
        "        [-2.8400, -0.7849, -1.4096, -0.4076,  0.7953]], requires_grad=True)"
       ]
      },
-     "execution_count": 18,
+     "execution_count": 4,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -132,7 +132,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 5,
    "id": "8bbc0255-4805-4be9-9f4c-1d0d967ef9d5",
    "metadata": {},
    "outputs": [
@@ -143,7 +143,7 @@
        "       grad_fn=<EmbeddingBackward0>)"
       ]
      },
-     "execution_count": 17,
+     "execution_count": 5,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -178,7 +178,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 6,
    "id": "c309266a-c601-4633-9404-2e10b1cdde8c",
    "metadata": {},
    "outputs": [
@@ -189,7 +189,7 @@
        "       grad_fn=<EmbeddingBackward0>)"
       ]
      },
-     "execution_count": 19,
+     "execution_count": 6,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -216,7 +216,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 7,
    "id": "0191aa4b-f6a8-4b0d-9c36-65e82b81d071",
    "metadata": {},
    "outputs": [
@@ -229,7 +229,7 @@
        "       grad_fn=<EmbeddingBackward0>)"
       ]
      },
-     "execution_count": 22,
+     "execution_count": 7,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -274,7 +274,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 8,
    "id": "b5bb56cf-bc73-41ab-b107-91a43f77bdba",
    "metadata": {},
    "outputs": [
@@ -286,7 +286,7 @@
        "        [0, 1, 0, 0]])"
       ]
      },
-     "execution_count": 23,
+     "execution_count": 8,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -306,7 +306,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 28,
+   "execution_count": 9,
    "id": "ae04c1ed-242e-4dd7-b8f7-4b7e4caae383",
    "metadata": {},
    "outputs": [
@@ -321,14 +321,15 @@
        "        [-0.3814,  0.3274, -0.1179,  0.1605]], requires_grad=True)"
       ]
      },
-     "execution_count": 28,
+     "execution_count": 9,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
     "torch.manual_seed(123)\n",
-    "linear = torch.nn.Linear(num_idx, out_dim, bias=False)"
+    "linear = torch.nn.Linear(num_idx, out_dim, bias=False)\n",
+    "linear.weight"
    ]
   },
   {
@@ -341,7 +342,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
+   "execution_count": 10,
    "id": "a3b90d69-761c-486e-bd19-b38a2988fe62",
    "metadata": {},
    "outputs": [],
@@ -359,7 +360,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
+   "execution_count": 11,
    "id": "90d2b0dd-9f1d-4c0f-bb16-1f6ce6b8ac2c",
    "metadata": {},
    "outputs": [
@@ -371,7 +372,7 @@
        "        [ 1.3010,  1.2753, -0.2010, -0.1606, -0.4015]], grad_fn=<MmBackward0>)"
       ]
      },
-     "execution_count": 31,
+     "execution_count": 11,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -390,7 +391,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 32,
+   "execution_count": 12,
    "id": "2b057649-3176-4a54-b58c-fd8fbf818c61",
    "metadata": {},
    "outputs": [
@@ -403,7 +404,7 @@
        "       grad_fn=<EmbeddingBackward0>)"
       ]
      },
-     "execution_count": 32,
+     "execution_count": 12,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -452,14 +453,6 @@
     "- Since all but one index in each one-hot encoded row are 0 (by design), this matrix multiplication is essentially the same as a look-up of the one-hot elements\n",
     "- This use of the matrix multiplication on one-hot encodings is equivalent to the embedding layer look-up but can be inefficient if we work with large embedding matrices, because there are a lot of wasteful multiplications by zero"
    ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "5eacc005-86fc-490c-8f6a-dc37d8a0df7c",
-   "metadata": {},
-   "outputs": [],
-   "source": []
   }
  ],
  "metadata": {
@@ -478,7 +471,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.12"
+   "version": "3.10.6"
   }
  },
  "nbformat": 4,