-
Notifications
You must be signed in to change notification settings - Fork 0
/
sentiment_classification.txt
65 lines (65 loc) · 3.61 KB
/
sentiment_classification.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
[nltk_data] Downloading package stopwords to /root/nltk_data...
[nltk_data] Package stopwords is already up-to-date!
[nltk_data] Downloading package punkt to /root/nltk_data...
[nltk_data] Package punkt is already up-to-date!
-- AIT726 Homework 2 from Julia Jeng, Shu Wang, and Arman Anwar --
Load TrainSet: 1181/3000 positive/negative samples.
Vocabulary: 6636 items.
Vocabulary (stem): 5090 items.
Load TestSet: 1182/3000 positive/negative samples.
[Epoch 100] loss: 0.715, train acc: 28.499%, valid acc: 27.240%
[Epoch 200] loss: 0.715, train acc: 28.499%, valid acc: 27.240%
[Epoch 300] loss: 0.201, train acc: 71.501%, valid acc: 72.760%
[Epoch 400] loss: 0.193, train acc: 73.325%, valid acc: 74.671%
[Epoch 500] loss: 0.184, train acc: 74.701%, valid acc: 77.419%
[Epoch 600] loss: 0.177, train acc: 75.807%, valid acc: 78.136%
[Epoch 700] loss: 0.170, train acc: 77.303%, valid acc: 78.853%
[Epoch 800] loss: 0.161, train acc: 78.589%, valid acc: 79.809%
[Epoch 900] loss: 0.150, train acc: 80.562%, valid acc: 81.243%
[Epoch 1000] loss: 0.138, train acc: 82.386%, valid acc: 81.959%
[Epoch 1100] loss: 0.126, train acc: 84.510%, valid acc: 83.274%
[Epoch 1200] loss: 0.112, train acc: 86.364%, valid acc: 85.185%
[Epoch 1300] loss: 0.100, train acc: 88.457%, valid acc: 85.783%
[Epoch 1400] loss: 0.089, train acc: 89.862%, valid acc: 86.380%
[Epoch 1500] loss: 0.082, train acc: 90.789%, valid acc: 86.619%
[Epoch 1600] loss: 0.076, train acc: 91.717%, valid acc: 86.858%
[Epoch 1700] loss: 0.071, train acc: 92.255%, valid acc: 87.097%
[Epoch 1800] loss: 0.067, train acc: 92.703%, valid acc: 87.455%
[Epoch 1900] loss: 0.064, train acc: 93.032%, valid acc: 87.455%
[Epoch 2000] loss: 0.061, train acc: 93.391%, valid acc: 87.455%
[Epoch 2100] loss: 0.059, train acc: 93.750%, valid acc: 87.814%
[Epoch 2200] loss: 0.058, train acc: 93.870%, valid acc: 88.053%
-------------------------------------------
Feed Forward Neural Network | Stem
test accuracy : 86.70%
confusion matrix : (actual)
Neg Pos
(predicted) Neg 2811(TN) 367 (FN)
Pos 189 (FP) 815 (TP)
-------------------------------------------
[Epoch 100] loss: 0.716, train acc: 28.170%, valid acc: 28.554%
[Epoch 200] loss: 0.716, train acc: 28.170%, valid acc: 28.554%
[Epoch 300] loss: 0.196, train acc: 72.937%, valid acc: 72.282%
[Epoch 400] loss: 0.175, train acc: 75.927%, valid acc: 75.508%
[Epoch 500] loss: 0.160, train acc: 78.888%, valid acc: 76.105%
[Epoch 600] loss: 0.143, train acc: 82.177%, valid acc: 78.256%
[Epoch 700] loss: 0.123, train acc: 85.138%, valid acc: 79.928%
[Epoch 800] loss: 0.111, train acc: 87.022%, valid acc: 81.720%
[Epoch 900] loss: 0.090, train acc: 89.145%, valid acc: 82.437%
[Epoch 1000] loss: 0.079, train acc: 90.580%, valid acc: 83.751%
[Epoch 1100] loss: 0.071, train acc: 91.687%, valid acc: 83.871%
[Epoch 1200] loss: 0.066, train acc: 92.195%, valid acc: 84.110%
[Epoch 1300] loss: 0.060, train acc: 93.122%, valid acc: 84.946%
[Epoch 1400] loss: 0.056, train acc: 93.511%, valid acc: 85.305%
[Epoch 1500] loss: 0.053, train acc: 93.840%, valid acc: 85.424%
[Epoch 1600] loss: 0.050, train acc: 94.258%, valid acc: 85.424%
[Epoch 1700] loss: 0.048, train acc: 94.557%, valid acc: 86.260%
[Epoch 1800] loss: 0.047, train acc: 94.587%, valid acc: 86.260%
-------------------------------------------
Feed Forward Neural Network | noStem
test accuracy : 86.35%
confusion matrix : (actual)
Neg Pos
(predicted) Neg 2782(TN) 353 (FN)
Pos 218 (FP) 829 (TP)
-------------------------------------------