Json

yala · Jan 28, 2017 · 072e267 · 072e267
1 parent c797c5d
commit 072e267
Show file tree

Hide file tree

Showing 4 changed files with 13 additions and 6 deletions.
diff --git a/6.S191 Lab 2.ipynb b/6.S191 Lab 2.ipynb
@@ -13,7 +13,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 1,
    "metadata": {
     "collapsed": true
    },
@@ -23,7 +23,8 @@
     "import pickle as p\n",
     "import tensorflow as tf\n",
     "import numpy as np\n",
-    "import utils"
+    "import utils\n",
+    "import json"
    ]
   },
   {
@@ -296,7 +297,10 @@
     "tf.global_variables_initializer().run(session=session)\n",
     "\n",
     "# load our data and separate it into tweets and labels\n",
-    "train_data = p.load(open('data/trainTweets_preprocessed.p','rb'))\n",
+    "train_data = json.load(open('data/trainTweets_preprocessed.json', 'rb'))\n",
+    "train_data = map(lambda row:(np.array(row[0],dtype=np.int32),str(row[1])),train_data3)\n",
+    "\n",
+    "#train_data = p.load(open('data/trainTweets_preprocessed.p','rb'))\n",
     "train_tweets = np.array([t[0] for t in train_data])\n",
     "train_labels = np.array([int(t[1]) for t in train_data])\n",
     "\n",
@@ -446,7 +450,8 @@
     "tf.global_variables_initializer().run(session=session)\n",
     "\n",
     "# load our data and separate it into tweets and labels\n",
-    "train_data = p.load(open('data/trainTweets_preprocessed.p','rb'))\n",
+    "train_data = json.load(open('data/trainTweets_preprocessed.json', 'rb'))\n",
+    "train_data = map(lambda row:(np.array(row[0],dtype=np.int32),str(row[1])),train_data3)\n",
     "train_tweets = np.array([t[0] for t in train_data])\n",
     "train_labels = np.array([int(t[1]) for t in train_data])\n",
     "\n",

diff --git a/data/testTweets_preprocessed.json b/data/testTweets_preprocessed.json
diff --git a/data/trainTweets_preprocessed.json b/data/trainTweets_preprocessed.json
diff --git a/lab1_utils.py b/lab1_utils.py
@@ -2,7 +2,7 @@
 import pandas as pd
 import re
 import special_tokens
-import np
+import numpy as np
 import random
 import tensorflow as tf
 import matplotlib.pyplot as plt
@@ -181,7 +181,7 @@ def classify_and_plot(data, labels, x, out, session):
         # print('ec', int(decision), int(label))
         # print('ec2', decision, label)
         # print(int(decision) is int(label))
-        m_text = 'g' if int(decision) is int(label) else 'r'
+        m_text = 'g' if int(decision) == int(label) else 'r'
         m_text += '_' if label == 0 else '+'
         plt.plot(x_input[0], x_input[1], m_text, markersize=10)