目录
基于《wonderland爱丽丝梦游仙境记》小说数据集利用LSTM算法(层加深，基于keras)对单个character字符预测
基于《wonderland爱丽丝梦游仙境记》小说数据集利用LSTM算法(层加深，基于keras)对单个character字符预测
设计思路
数据集下载：https://download.csdn.net/download/qq_41185868/13767751
输出结果
1. Using TensorFlow backend. 2. F:\Program Files\Python\Python36\lib\site-packages\tensorflow\python\framework\dtypes.py:523: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'. 3. _np_qint8 = np.dtype([("qint8", np.int8, 1)]) 4. F:\Program Files\Python\Python36\lib\site-packages\tensorflow\python\framework\dtypes.py:524: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'. 5. _np_quint8 = np.dtype([("quint8", np.uint8, 1)]) 6. F:\Program Files\Python\Python36\lib\site-packages\tensorflow\python\framework\dtypes.py:525: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'. 7. _np_qint16 = np.dtype([("qint16", np.int16, 1)]) 8. F:\Program Files\Python\Python36\lib\site-packages\tensorflow\python\framework\dtypes.py:526: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'. 9. _np_quint16 = np.dtype([("quint16", np.uint16, 1)]) 10. F:\Program Files\Python\Python36\lib\site-packages\tensorflow\python\framework\dtypes.py:527: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'. 11. _np_qint32 = np.dtype([("qint32", np.int32, 1)]) 12. F:\Program Files\Python\Python36\lib\site-packages\tensorflow\python\framework\dtypes.py:532: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'. 13. np_resource = np.dtype([("resource", np.ubyte, 1)]) 14. [nltk_data] Error loading punkt: <urlopen error [Errno 11004] 15. [nltk_data] getaddrinfo failed> 16. raw_text[:10] : alice's ad 17. Total Characters: 144413 18. chars ['\n', ' ', '!', '"', "'", '(', ')', '*', ',', '-', '.', '0', '3', ':', ';', '?', '[', ']', '_', 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z'] 19. Total Vocab: 45 20. sentences 1625 ["alice's adventures in wonderland\n\nlewis carroll\n\nthe millennium fulcrum edition 3.0\n\nchapter i. down the rabbit-hole\n\nalice was beginning to get very tired of sitting by her sister on the\nbank, and of having nothing to do: once or twice she had peeped into the\nbook her sister was reading, but it had no pictures or conversations in\nit, 'and what is the use of a book,' thought alice 'without pictures or\nconversations?'", 'so she was considering in her own mind (as well as she could, for the\nhot day made her feel very sleepy and stupid), whether the pleasure\nof making a daisy-chain would be worth the trouble of getting up and\npicking the daisies, when suddenly a white rabbit with pink eyes ran\nclose by her.', "there was nothing so very remarkable in that; nor did alice think it so\nvery much out of the way to hear the rabbit say to itself, 'oh dear!", 'oh dear!', "i shall be late!'"] 21. lengths (1625,) [420 289 140 ... 636 553 7] 22. CharMapInt_dict 45 {'\n': 0, ' ': 1, '!': 2, '"': 3, "'": 4, '(': 5, ')': 6, '*': 7, ',': 8, '-': 9, '.': 10, '0': 11, '3': 12, ':': 13, ';': 14, '?': 15, '[': 16, ']': 17, '_': 18, 'a': 19, 'b': 20, 'c': 21, 'd': 22, 'e': 23, 'f': 24, 'g': 25, 'h': 26, 'i': 27, 'j': 28, 'k': 29, 'l': 30, 'm': 31, 'n': 32, 'o': 33, 'p': 34, 'q': 35, 'r': 36, 's': 37, 't': 38, 'u': 39, 'v': 40, 'w': 41, 'x': 42, 'y': 43, 'z': 44} 23. IntMapChar_dict 45 {0: '\n', 1: ' ', 2: '!', 3: '"', 4: "'", 5: '(', 6: ')', 7: '*', 8: ',', 9: '-', 10: '.', 11: '0', 12: '3', 13: ':', 14: ';', 15: '?', 16: '[', 17: ']', 18: '_', 19: 'a', 20: 'b', 21: 'c', 22: 'd', 23: 'e', 24: 'f', 25: 'g', 26: 'h', 27: 'i', 28: 'j', 29: 'k', 30: 'l', 31: 'm', 32: 'n', 33: 'o', 34: 'p', 35: 'q', 36: 'r', 37: 's', 38: 't', 39: 'u', 40: 'v', 41: 'w', 42: 'x', 43: 'y', 44: 'z'} 24. dataX： 144313 100 [[19, 30, 27, 21, 23, 4, 37, 1, 19, 22, 40, 23, 32, 38, 39, 36, 23, 37, 1, 27, 32, 1, 41, 33, 32, 22, 23, 36, 30, 19, 32, 22, 0, 0, 30, 23, 41, 27, 37, 1, 21, 19, 36, 36, 33, 30, 30, 0, 0, 38, 26, 23, 1, 31, 27, 30, 30, 23, 32, 32, 27, 39, 31, 1, 24, 39, 30, 21, 36, 39, 31, 1, 23, 22, 27, 38, 27, 33, 32, 1, 12, 10, 11, 0, 0, 21, 26, 19, 34, 38, 23, 36, 1, 27, 10, 1, 22, 33, 41, 32], [30, 27, 21, 23, 4, 37, 1, 19, 22, 40, 23, 32, 38, 39, 36, 23, 37, 1, 27, 32, 1, 41, 33, 32, 22, 23, 36, 30, 19, 32, 22, 0, 0, 30, 23, 41, 27, 37, 1, 21, 19, 36, 36, 33, 30, 30, 0, 0, 38, 26, 23, 1, 31, 27, 30, 30, 23, 32, 32, 27, 39, 31, 1, 24, 39, 30, 21, 36, 39, 31, 1, 23, 22, 27, 38, 27, 33, 32, 1, 12, 10, 11, 0, 0, 21, 26, 19, 34, 38, 23, 36, 1, 27, 10, 1, 22, 33, 41, 32, 1], [27, 21, 23, 4, 37, 1, 19, 22, 40, 23, 32, 38, 39, 36, 23, 37, 1, 27, 32, 1, 41, 33, 32, 22, 23, 36, 30, 19, 32, 22, 0, 0, 30, 23, 41, 27, 37, 1, 21, 19, 36, 36, 33, 30, 30, 0, 0, 38, 26, 23, 1, 31, 27, 30, 30, 23, 32, 32, 27, 39, 31, 1, 24, 39, 30, 21, 36, 39, 31, 1, 23, 22, 27, 38, 27, 33, 32, 1, 12, 10, 11, 0, 0, 21, 26, 19, 34, 38, 23, 36, 1, 27, 10, 1, 22, 33, 41, 32, 1, 38], [21, 23, 4, 37, 1, 19, 22, 40, 23, 32, 38, 39, 36, 23, 37, 1, 27, 32, 1, 41, 33, 32, 22, 23, 36, 30, 19, 32, 22, 0, 0, 30, 23, 41, 27, 37, 1, 21, 19, 36, 36, 33, 30, 30, 0, 0, 38, 26, 23, 1, 31, 27, 30, 30, 23, 32, 32, 27, 39, 31, 1, 24, 39, 30, 21, 36, 39, 31, 1, 23, 22, 27, 38, 27, 33, 32, 1, 12, 10, 11, 0, 0, 21, 26, 19, 34, 38, 23, 36, 1, 27, 10, 1, 22, 33, 41, 32, 1, 38, 26], [23, 4, 37, 1, 19, 22, 40, 23, 32, 38, 39, 36, 23, 37, 1, 27, 32, 1, 41, 33, 32, 22, 23, 36, 30, 19, 32, 22, 0, 0, 30, 23, 41, 27, 37, 1, 21, 19, 36, 36, 33, 30, 30, 0, 0, 38, 26, 23, 1, 31, 27, 30, 30, 23, 32, 32, 27, 39, 31, 1, 24, 39, 30, 21, 36, 39, 31, 1, 23, 22, 27, 38, 27, 33, 32, 1, 12, 10, 11, 0, 0, 21, 26, 19, 34, 38, 23, 36, 1, 27, 10, 1, 22, 33, 41, 32, 1, 38, 26, 23]] 25. dataY： 144313 [1, 38, 26, 23, 1] 26. Total patterns: 144313 27. X_train.shape (144313, 100, 1) 28. Y_train.shape (144313, 45) 29. Init data，after read_out, chars: 30. 144313 alice's adventures in wonderland 31. 32. lewis carroll 33. 34. tge millennium fulcrum edition 3.0 35. 36. cgapter i. down 37. _________________________________________________________________ 38. Layer (type) Output Shape Param # 39. ================================================================= 40. F:\File_Jupyter\实用代码\NeuralNetwork（神经网络）\CharacterLanguageLSTM.py:135: UserWarning: The `nb_epoch` argument in `fit` has been renamed `epochs`. 41. LSTM_Model.fit(X_train[:train_index], Y_train[:train_index], nb_epoch=10, batch_size=64, callbacks=callbacks_list) 42. lstm_1 (LSTM) (None, 256) 264192 43. _________________________________________________________________ 44. dropout_1 (Dropout) (None, 256) 0 45. _________________________________________________________________ 46. dense_1 (Dense) (None, 45) 11565 47. ================================================================= 48. Total params: 275,757 49. Trainable params: 275,757 50. Non-trainable params: 0 51. _________________________________________________________________ 52. LSTM_Model 53. None 54. Epoch 1/10 55. 2020-12-23 23:42:07.919094: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 56. 57. 64/1000 [>.............................] - ETA: 29s - loss: 3.8086 58. 128/1000 [==>...........................] - ETA: 15s - loss: 3.7953 59. 192/1000 [====>.........................] - ETA: 11s - loss: 3.7823 60. 256/1000 [======>.......................] - ETA: 8s - loss: 3.7692 61. 320/1000 [========>.....................] - ETA: 7s - loss: 3.7552 62. 384/1000 [==========>...................] - ETA: 5s - loss: 3.7372 63. 448/1000 [============>.................] - ETA: 4s - loss: 3.7026 64. 512/1000 [==============>...............] - ETA: 4s - loss: 3.6552 65. 576/1000 [================>.............] - ETA: 3s - loss: 3.5955 66. 640/1000 [==================>...........] - ETA: 2s - loss: 3.5678 67. 704/1000 [====================>.........] - ETA: 2s - loss: 3.5116 68. 768/1000 [======================>.......] - ETA: 1s - loss: 3.4778 69. 832/1000 [=======================>......] - ETA: 1s - loss: 3.4441 70. 896/1000 [=========================>....] - ETA: 0s - loss: 3.4278 71. 960/1000 [===========================>..] - ETA: 0s - loss: 3.4092 72. 1000/1000 [==============================] - 7s 7ms/step - loss: 3.3925 73. 74. Epoch 00001: loss improved from inf to 3.39249, saving model to hdf5/weights-improvement-01-3.3925.hdf5 75. Epoch 2/10 76. 77. 64/1000 [>.............................] - ETA: 4s - loss: 3.1429 78. 128/1000 [==>...........................] - ETA: 4s - loss: 3.1370 79. 192/1000 [====>.........................] - ETA: 3s - loss: 3.1034 80. 256/1000 [======>.......................] - ETA: 3s - loss: 3.1038 81. 320/1000 [========>.....................] - ETA: 3s - loss: 3.0962 82. 384/1000 [==========>...................] - ETA: 2s - loss: 3.1055 83. 448/1000 [============>.................] - ETA: 2s - loss: 3.0986 84. 512/1000 [==============>...............] - ETA: 2s - loss: 3.0628 85. 576/1000 [================>.............] - ETA: 2s - loss: 3.0452 86. 640/1000 [==================>...........] - ETA: 1s - loss: 3.0571 87. 704/1000 [====================>.........] - ETA: 1s - loss: 3.0684 88. 768/1000 [======================>.......] - ETA: 1s - loss: 3.0606 89. 832/1000 [=======================>......] - ETA: 0s - loss: 3.0596 90. 896/1000 [=========================>....] - ETA: 0s - loss: 3.0529 91. 960/1000 [===========================>..] - ETA: 0s - loss: 3.0484 92. 1000/1000 [==============================] - 5s 5ms/step - loss: 3.0371 93. 94. Epoch 00002: loss improved from 3.39249 to 3.03705, saving model to hdf5/weights-improvement-02-3.0371.hdf5 95. Epoch 3/10 96. 97. 64/1000 [>.............................] - ETA: 4s - loss: 3.1671 98. 128/1000 [==>...........................] - ETA: 4s - loss: 3.0008 99. 192/1000 [====>.........................] - ETA: 4s - loss: 3.0159 100. 256/1000 [======>.......................] - ETA: 4s - loss: 3.0019 101. 320/1000 [========>.....................] - ETA: 3s - loss: 3.0056 102. 384/1000 [==========>...................] - ETA: 3s - loss: 3.0156 103. 448/1000 [============>.................] - ETA: 2s - loss: 3.0392 104. 512/1000 [==============>...............] - ETA: 2s - loss: 3.0243 105. 576/1000 [================>.............] - ETA: 2s - loss: 3.0226 106. 640/1000 [==================>...........] - ETA: 1s - loss: 3.0162 107. 704/1000 [====================>.........] - ETA: 1s - loss: 3.0238 108. 768/1000 [======================>.......] - ETA: 1s - loss: 3.0195 109. 832/1000 [=======================>......] - ETA: 0s - loss: 3.0286 110. 896/1000 [=========================>....] - ETA: 0s - loss: 3.0272 111. 960/1000 [===========================>..] - ETA: 0s - loss: 3.0214 112. 1000/1000 [==============================] - 6s 6ms/step - loss: 3.0225 113. 114. Epoch 00003: loss improved from 3.03705 to 3.02249, saving model to hdf5/weights-improvement-03-3.0225.hdf5 115. Epoch 4/10 116. 117. 64/1000 [>.............................] - ETA: 5s - loss: 2.7843 118. 128/1000 [==>...........................] - ETA: 5s - loss: 2.8997 119. 192/1000 [====>.........................] - ETA: 4s - loss: 2.9975 120. 256/1000 [======>.......................] - ETA: 4s - loss: 3.0150 121. 320/1000 [========>.....................] - ETA: 3s - loss: 3.0025 122. 384/1000 [==========>...................] - ETA: 3s - loss: 3.0442 123. 448/1000 [============>.................] - ETA: 3s - loss: 3.0494 124. 512/1000 [==============>...............] - ETA: 2s - loss: 3.0398 125. 576/1000 [================>.............] - ETA: 2s - loss: 3.0170 126. 640/1000 [==================>...........] - ETA: 2s - loss: 3.0421 127. 704/1000 [====================>.........] - ETA: 1s - loss: 3.0366 128. 768/1000 [======================>.......] - ETA: 1s - loss: 3.0339 129. 832/1000 [=======================>......] - ETA: 0s - loss: 3.0316 130. 896/1000 [=========================>....] - ETA: 0s - loss: 3.0361 131. 960/1000 [===========================>..] - ETA: 0s - loss: 3.0326 132. 1000/1000 [==============================] - 6s 6ms/step - loss: 3.0352 133. 134. Epoch 00004: loss did not improve from 3.02249 135. Epoch 5/10 136. 137. 64/1000 [>.............................] - ETA: 4s - loss: 2.8958 138. 128/1000 [==>...........................] - ETA: 4s - loss: 2.9239 139. 192/1000 [====>.........................] - ETA: 4s - loss: 2.9044 140. 256/1000 [======>.......................] - ETA: 4s - loss: 2.9417 141. 320/1000 [========>.....................] - ETA: 3s - loss: 2.9674 142. 384/1000 [==========>...................] - ETA: 3s - loss: 2.9646 143. 448/1000 [============>.................] - ETA: 3s - loss: 2.9629 144. 512/1000 [==============>...............] - ETA: 2s - loss: 2.9707 145. 576/1000 [================>.............] - ETA: 2s - loss: 2.9699 146. 640/1000 [==================>...........] - ETA: 1s - loss: 2.9594 147. 704/1000 [====================>.........] - ETA: 1s - loss: 2.9830 148. 768/1000 [======================>.......] - ETA: 1s - loss: 2.9773 149. 832/1000 [=======================>......] - ETA: 0s - loss: 2.9774 150. 896/1000 [=========================>....] - ETA: 0s - loss: 2.9891 151. 960/1000 [===========================>..] - ETA: 0s - loss: 3.0070 152. 1000/1000 [==============================] - 5s 5ms/step - loss: 3.0120 153. 154. Epoch 00005: loss improved from 3.02249 to 3.01205, saving model to hdf5/weights-improvement-05-3.0120.hdf5 155. Epoch 6/10 156. 157. 64/1000 [>.............................] - ETA: 4s - loss: 3.0241 158. 128/1000 [==>...........................] - ETA: 4s - loss: 3.0463 159. 192/1000 [====>.........................] - ETA: 3s - loss: 3.0364 160. 256/1000 [======>.......................] - ETA: 3s - loss: 2.9712 161. 320/1000 [========>.....................] - ETA: 3s - loss: 2.9840 162. 384/1000 [==========>...................] - ETA: 3s - loss: 2.9887 163. 448/1000 [============>.................] - ETA: 2s - loss: 2.9785 164. 512/1000 [==============>...............] - ETA: 2s - loss: 2.9852 165. 576/1000 [================>.............] - ETA: 2s - loss: 2.9893 166. 640/1000 [==================>...........] - ETA: 1s - loss: 2.9931 167. 704/1000 [====================>.........] - ETA: 1s - loss: 2.9790 168. 768/1000 [======================>.......] - ETA: 1s - loss: 2.9962 169. 832/1000 [=======================>......] - ETA: 0s - loss: 3.0166 170. 896/1000 [=========================>....] - ETA: 0s - loss: 3.0213 171. 960/1000 [===========================>..] - ETA: 0s - loss: 3.0143 172. 1000/1000 [==============================] - 5s 5ms/step - loss: 3.0070 173. 174. Epoch 00006: loss improved from 3.01205 to 3.00701, saving model to hdf5/weights-improvement-06-3.0070.hdf5 175. Epoch 7/10 176. 177. 64/1000 [>.............................] - ETA: 5s - loss: 3.0738 178. 128/1000 [==>...........................] - ETA: 5s - loss: 3.0309 179. 192/1000 [====>.........................] - ETA: 4s - loss: 2.9733 180. 256/1000 [======>.......................] - ETA: 4s - loss: 2.9728 181. 320/1000 [========>.....................] - ETA: 4s - loss: 2.9422 182. 384/1000 [==========>...................] - ETA: 3s - loss: 2.9496 183. 448/1000 [============>.................] - ETA: 3s - loss: 2.9548 184. 512/1000 [==============>...............] - ETA: 3s - loss: 2.9635 185. 576/1000 [================>.............] - ETA: 2s - loss: 2.9614 186. 640/1000 [==================>...........] - ETA: 2s - loss: 2.9537 187. 704/1000 [====================>.........] - ETA: 1s - loss: 2.9454 188. 768/1000 [======================>.......] - ETA: 1s - loss: 2.9649 189. 832/1000 [=======================>......] - ETA: 1s - loss: 2.9814 190. 896/1000 [=========================>....] - ETA: 0s - loss: 2.9955 191. 960/1000 [===========================>..] - ETA: 0s - loss: 2.9948 192. 1000/1000 [==============================] - 6s 6ms/step - loss: 2.9903 193. 194. Epoch 00007: loss improved from 3.00701 to 2.99027, saving model to hdf5/weights-improvement-07-2.9903.hdf5 195. Epoch 8/10 196. 197. 64/1000 [>.............................] - ETA: 5s - loss: 2.9248 198. 128/1000 [==>...........................] - ETA: 4s - loss: 2.9293 199. 192/1000 [====>.........................] - ETA: 4s - loss: 2.9820 200. 256/1000 [======>.......................] - ETA: 4s - loss: 3.0261 201. 320/1000 [========>.....................] - ETA: 3s - loss: 2.9989 202. 384/1000 [==========>...................] - ETA: 3s - loss: 3.0101 203. 448/1000 [============>.................] - ETA: 3s - loss: 3.0050 204. 512/1000 [==============>...............] - ETA: 2s - loss: 3.0155 205. 576/1000 [================>.............] - ETA: 2s - loss: 3.0414 206. 640/1000 [==================>...........] - ETA: 2s - loss: 3.0180 207. 704/1000 [====================>.........] - ETA: 1s - loss: 3.0295 208. 768/1000 [======================>.......] - ETA: 1s - loss: 2.9996 209. 832/1000 [=======================>......] - ETA: 0s - loss: 3.0151 210. 896/1000 [=========================>....] - ETA: 0s - loss: 3.0201 211. 960/1000 [===========================>..] - ETA: 0s - loss: 3.0063 212. 1000/1000 [==============================] - 6s 6ms/step - loss: 3.0064 213. 214. Epoch 00008: loss did not improve from 2.99027 215. Epoch 9/10 216. 217. 64/1000 [>.............................] - ETA: 4s - loss: 2.8417 218. 128/1000 [==>...........................] - ETA: 4s - loss: 2.9652 219. 192/1000 [====>.........................] - ETA: 4s - loss: 2.9907 220. 256/1000 [======>.......................] - ETA: 3s - loss: 3.0133 221. 320/1000 [========>.....................] - ETA: 3s - loss: 3.0092 222. 384/1000 [==========>...................] - ETA: 3s - loss: 3.0139 223. 448/1000 [============>.................] - ETA: 2s - loss: 3.0453 224. 512/1000 [==============>...............] - ETA: 2s - loss: 3.0481 225. 576/1000 [================>.............] - ETA: 2s - loss: 3.0434 226. 640/1000 [==================>...........] - ETA: 1s - loss: 3.0158 227. 704/1000 [====================>.........] - ETA: 1s - loss: 3.0141 228. 768/1000 [======================>.......] - ETA: 1s - loss: 3.0203 229. 832/1000 [=======================>......] - ETA: 0s - loss: 3.0068 230. 896/1000 [=========================>....] - ETA: 0s - loss: 2.9980 231. 960/1000 [===========================>..] - ETA: 0s - loss: 3.0016 232. 1000/1000 [==============================] - 5s 5ms/step - loss: 2.9944 233. 234. Epoch 00009: loss did not improve from 2.99027 235. Epoch 10/10 236. 237. 64/1000 [>.............................] - ETA: 4s - loss: 3.0100 238. 128/1000 [==>...........................] - ETA: 4s - loss: 3.0620 239. 192/1000 [====>.........................] - ETA: 4s - loss: 3.0169 240. 256/1000 [======>.......................] - ETA: 3s - loss: 3.0289 241. 320/1000 [========>.....................] - ETA: 3s - loss: 3.0060 242. 384/1000 [==========>...................] - ETA: 3s - loss: 2.9940 243. 448/1000 [============>.................] - ETA: 2s - loss: 2.9823 244. 512/1000 [==============>...............] - ETA: 2s - loss: 2.9686 245. 576/1000 [================>.............] - ETA: 2s - loss: 2.9699 246. 640/1000 [==================>...........] - ETA: 1s - loss: 2.9710 247. 704/1000 [====================>.........] - ETA: 1s - loss: 2.9625 248. 768/1000 [======================>.......] - ETA: 1s - loss: 2.9748 249. 832/1000 [=======================>......] - ETA: 0s - loss: 2.9794 250. 896/1000 [=========================>....] - ETA: 0s - loss: 2.9788 251. 960/1000 [===========================>..] - ETA: 0s - loss: 2.9802 252. 1000/1000 [==============================] - 5s 5ms/step - loss: 2.9963 253. 254. Epoch 00010: loss did not improve from 2.99027 255. LSTM_Pre_word.shape: 256. (3, 45) 257. after LSTM read_out, chars: 258. 3 ["\n,\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n!\n\n '\n\n!!\n\n\n\n !\n\n! ' \n\n\n\n\n", "\n,\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n!\n\n '\n\n!!\n\n\n\n !\n\n! ' \n\n\n\n\n", '\n,\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n!\n\n "\n\n!!\n\n \n !\n\n! \' \n\n\n\n\n'] 259. LSTM_Model，Seed: 260. " ent down its head to hide a smile: some of the other birds 261. tittered audibly. 262. 263. 'what i was going to s " 264. 199 100 265. 266. Generated Sequence: 267. 268. 269. Done. 270. _________________________________________________________________ 271. Layer (type) Output Shape Param # 272. ================================================================= 273. lstm_2 (LSTM) (None, 100, 256) 264192 274. _________________________________________________________________ 275. dropout_2 (Dropout) (None, 100, 256) 0 276. _________________________________________________________________ 277. lstm_3 (LSTM) (None, 64) 82176 278. _________________________________________________________________ 279. dropout_3 (Dropout) (None, 64) 0 280. _________________________________________________________________ 281. dense_2 (Dense) (None, 45) 2925 282. ================================================================= 283. Total params: 349,293 284. Trainable params: 349,293 285. Non-trainable params: 0 286. _________________________________________________________________ 287. DeepLSTM_Model 288. None 289. F:\File_Jupyter\实用代码\NeuralNetwork（神经网络）\CharacterLanguageLSTM.py:246: UserWarning: The `nb_epoch` argument in `fit` has been renamed `epochs`. 290. DeepLSTM_Model.fit(X_train[:train_index], Y_train[:train_index], nb_epoch=2, batch_size=256, callbacks=callbacks_list) 291. Epoch 1/2 292. 293. 256/1000 [======>.......................] - ETA: 11s - loss: 3.8128 294. 512/1000 [==============>...............] - ETA: 5s - loss: 3.8058 295. 768/1000 [======================>.......] - ETA: 2s - loss: 3.7976 296. 1000/1000 [==============================] - 10s 10ms/step - loss: 3.7883 297. 298. Epoch 00001: loss improved from inf to 3.78827, saving model to hdf5/weights-improvement-01-3.7883.hdf5 299. Epoch 2/2 300. 301. 256/1000 [======>.......................] - ETA: 5s - loss: 3.7167 302. 512/1000 [==============>...............] - ETA: 4s - loss: 3.6880 303. 768/1000 [======================>.......] - ETA: 1s - loss: 3.6622 304. 1000/1000 [==============================] - 8s 8ms/step - loss: 3.6151 305. 306. Epoch 00002: loss improved from 3.78827 to 3.61512, saving model to hdf5/weights-improvement-02-3.6151.hdf5 307. DeepLSTM_Pre_word.shape: 308. (3, 45) 309. after DeepLSTM read_out, chars: 310. 3 ["\n,\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n!\n\n '\n\n!!\n\n\n\n !\n\n! ' \n\n\n\n\n", "\n,\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n!\n\n '\n\n!!\n\n\n\n !\n\n! ' \n\n\n\n\n", '\n,\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n!\n\n "\n\n!!\n\n \n !\n\n! \' \n\n\n\n\n']
核心代码
1. LSTM_Model = Sequential() 2. LSTM_Model.add(LSTM(256, input_shape=(X_train.shape[1], X_train.shape[2]))) 3. LSTM_Model.add(Dropout(0.2)) 4. LSTM_Model.add(Dense(Y_train.shape[1], activation='softmax')) 5. LSTM_Model.compile(loss='categorical_crossentropy', optimizer='adam') 6. print('LSTM_Model \n',LSTM_Model.summary()) 7. 8. 9. 10. 11. embedding_vector_length = 32 12. LSTMWithE_Model = Sequential() 13. LSTMWithE_Model.add(Embedding(chars_len, embedding_vector_length, input_length=seq_length)) 14. LSTMWithE_Model.add(LSTM(256)) 15. LSTMWithE_Model.add(Dropout(0.2)) 16. LSTMWithE_Model.add(Dense(Y_train.shape[1], activation='softmax')) 17. LSTMWithE_Model.compile(loss='categorical_crossentropy', optimizer='adam') 18. print (LSTMWithE_Model.summary()) 19. 20. 21. 22. DeepLSTM_Model = Sequential() 23. DeepLSTM_Model.add(LSTM(256, input_shape=(X_train.shape[1], X_train.shape[2]), return_sequences=True)) 24. DeepLSTM_Model.add(Dropout(0.2)) 25. DeepLSTM_Model.add(LSTM(64)) 26. DeepLSTM_Model.add(Dropout(0.2)) 27. DeepLSTM_Model.add(Dense(Y_train.shape[1], activation='softmax')) 28. DeepLSTM_Model.compile(loss='categorical_crossentropy', optimizer='adam') 29. print('DeepLSTM_Model \n',DeepLSTM_Model.summary())
