pull from remote

Author: ShusenTang

code/chapter10_natural-language-processing/10.7_sentiment-analysis-rnn.ipynb

@@ -21,7 +21,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "1.0.0 cuda\n"
+      "1.1.0 cuda\n"
      ]
     }
    ],
@@ -39,10 +39,10 @@
     "sys.path.append(\"..\") \n",
     "import d2lzh_pytorch as d2l\n",
     "\n",
-    "os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"7\"\n",
+    "os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"2\"\n",
     "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
     "\n",
-    "DATA_ROOT = \"/S1/CSCL/tangss/Datasets\"\n",
+    "DATA_ROOT = \"/data1/tangss/Datasets\"\n",
     "\n",
     "print(torch.__version__, device)"
    ]
@@ -88,10 +88,10 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "100%|██████████| 12500/12500 [00:04<00:00, 2930.03it/s]\n",
-      "100%|██████████| 12500/12500 [00:04<00:00, 3008.48it/s]\n",
-      "100%|██████████| 12500/12500 [00:03<00:00, 3365.08it/s]\n",
-      "100%|██████████| 12500/12500 [00:03<00:00, 3305.63it/s]\n"
+      "100%|██████████| 12500/12500 [00:00<00:00, 34211.42it/s]\n",
+      "100%|██████████| 12500/12500 [00:00<00:00, 38506.48it/s]\n",
+      "100%|██████████| 12500/12500 [00:00<00:00, 31316.61it/s]\n",
+      "100%|██████████| 12500/12500 [00:00<00:00, 29664.72it/s]\n"
      ]
     }
    ],
@@ -108,7 +108,8 @@
     "    random.shuffle(data)\n",
     "    return data\n",
     "\n",
-    "train_data, test_data = read_imdb('train'), read_imdb('test')"
+    "data_root = os.path.join(DATA_ROOT, \"aclImdb\")\n",
+    "train_data, test_data = read_imdb('train', data_root), read_imdb('test', data_root)"
    ]
   },
   {
@@ -152,7 +153,7 @@
     {
      "data": {
       "text/plain": [
-       "('# words in vocab:', 46151)"
+       "('# words in vocab:', 46152)"
       ]
      },
      "execution_count": 5,
@@ -330,8 +331,7 @@
     "ExecuteTime": {
      "end_time": "2019-07-03T04:26:47.895604Z",
      "start_time": "2019-07-03T04:26:47.685801Z"
-    },
-    "collapsed": true
+    }
    },
    "outputs": [],
    "source": [
@@ -345,10 +345,17 @@
     "ExecuteTime": {
      "end_time": "2019-07-03T04:26:48.102388Z",
      "start_time": "2019-07-03T04:26:47.897582Z"
-    },
-    "collapsed": true
+    }
    },
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "There are 21202 oov words.\n"
+     ]
+    }
+   ],
    "source": [
     "def load_pretrained_embedding(words, pretrained_vocab):\n",
     "    \"\"\"从预训练好的vocab中提取出words对应的词向量\"\"\"\n",
@@ -359,9 +366,9 @@
     "            idx = pretrained_vocab.stoi[word]\n",
     "            embed[i, :] = pretrained_vocab.vectors[idx]\n",
     "        except KeyError:\n",
-    "            oov_count += 0\n",
+    "            oov_count += 1\n",
     "    if oov_count > 0:\n",
-    "        print(\"There are %d oov words.\")\n",
+    "        print(\"There are %d oov words.\" % oov_count)\n",
     "    return embed\n",
     "\n",
     "net.embedding.weight.data.copy_(load_pretrained_embedding(vocab.itos, glove_vocab))\n",
@@ -390,11 +397,11 @@
      "output_type": "stream",
      "text": [
       "training on  cuda\n",
-      "epoch 1, loss 0.5759, train acc 0.666, test acc 0.832, time 250.8 sec\n",
-      "epoch 2, loss 0.1785, train acc 0.842, test acc 0.852, time 253.3 sec\n",
-      "epoch 3, loss 0.1042, train acc 0.866, test acc 0.856, time 253.7 sec\n",
-      "epoch 4, loss 0.0682, train acc 0.888, test acc 0.868, time 254.2 sec\n",
-      "epoch 5, loss 0.0483, train acc 0.901, test acc 0.862, time 251.4 sec\n"
+      "epoch 1, loss 0.5415, train acc 0.719, test acc 0.819, time 48.7 sec\n",
+      "epoch 2, loss 0.1897, train acc 0.837, test acc 0.852, time 53.0 sec\n",
+      "epoch 3, loss 0.1105, train acc 0.857, test acc 0.844, time 51.6 sec\n",
+      "epoch 4, loss 0.0719, train acc 0.881, test acc 0.865, time 52.1 sec\n",
+      "epoch 5, loss 0.0519, train acc 0.894, test acc 0.852, time 51.2 sec\n"
      ]
     }
    ],
@@ -488,9 +495,9 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python [conda env:py36]",
+   "display_name": "Python [conda env:py36_pytorch]",
    "language": "python",
-   "name": "conda-env-py36-py"
+   "name": "conda-env-py36_pytorch-py"
   },
   "language_info": {
    "codemirror_mode": {
@@ -502,7 +509,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.3"
+   "version": "3.6.2"
   },
   "varInspector": {
    "cols": {

code/d2lzh_pytorch/utils.py

@@ -1203,9 +1203,9 @@ def load_pretrained_embedding(words, pretrained_vocab):
             idx = pretrained_vocab.stoi[word]
             embed[i, :] = pretrained_vocab.vectors[idx]
         except KeyError:
-            oov_count += 0
+            oov_count += 1
     if oov_count > 0:
-        print("There are %d oov words.")
+        print("There are %d oov words." % oov_count)
     return embed
 
 def predict_sentiment(net, vocab, sentence):

docs/chapter05_CNN/5.8_nin.md

@@ -43,6 +43,7 @@ NiN是在AlexNet问世不久后提出的。它们的卷积层设定有类似之
 
 ``` python
 # 已保存在d2lzh_pytorch
+import torch.nn.functional as F
 class GlobalAvgPool2d(nn.Module):
     # 全局平均池化层可通过将池化窗口形状设置成输入的高和宽实现
     def __init__(self):

docs/chapter10_natural-language-processing/10.7_sentiment-analysis-rnn.md

@@ -195,15 +195,19 @@ def load_pretrained_embedding(words, pretrained_vocab):
             idx = pretrained_vocab.stoi[word]
             embed[i, :] = pretrained_vocab.vectors[idx]
         except KeyError:
-            oov_count += 0
+            oov_count += 1
     if oov_count > 0:
-        print("There are %d oov words.")
+        print("There are %d oov words." % oov_count)
     return embed
 
 net.embedding.weight.data.copy_(
     load_pretrained_embedding(vocab.itos, glove_vocab))
 net.embedding.weight.requires_grad = False # 直接加载预训练好的, 所以不需要更新它
 ```
+输出:
+```
+There are 21202 oov words.
+```
 
 ### 10.7.2.2 训练并评价模型