Tensorflow之models中lstm测试

在Tensorflow中测试lstm
首先要下载数据数据存储在\simple-examples\data中
然后下载代码，代码存储在models-master\tutorials\rnn\ptb中
在python中运行ptb_word_lm.py，报编码错误时候删除decode(‘utf-8’)
然后运行ptb_word_lm.py
运行命令为
cd models/tutorials/rnn/ptb
python ptb_word_lm.py –data_path=/tmp/simple-examples/data/ –model=small
实验结果如下：

2017-06-02 17:13:20.158843: W c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\platform\cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE instructions, but these are available on your machine and could speed up CPU computations.2017-06-02 17:13:20.159843: W c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\platform\cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE2 instructions, but these are available on your machine and could speed up CPU computations.2017-06-02 17:13:20.159843: W c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\platform\cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE3 instructions, but these are available on your machine and could speed up CPU computations.2017-06-02 17:13:20.159843: W c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\platform\cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations.2017-06-02 17:13:20.160843: W c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\platform\cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations.2017-06-02 17:13:20.160843: W c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\platform\cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.2017-06-02 17:13:20.161844: W c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\platform\cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations.2017-06-02 17:13:20.161844: W c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\platform\cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use FMA instructions, but these are available on your machine and could speed up CPU computations.WARNING:tensorflow:Standard services need a 'logdir' passed to the SessionManagerEpoch: 1 Learning rate: 1.0000.004 perplexity: 5871.056 speed: 1602 wps0.104 perplexity: 851.786 speed: 2284 wps0.204 perplexity: 631.999 speed: 2318 wps0.304 perplexity: 510.089 speed: 2330 wps0.404 perplexity: 439.275 speed: 2316 wps0.504 perplexity: 393.267 speed: 2303 wps0.604 perplexity: 353.944 speed: 2302 wps0.703 perplexity: 327.084 speed: 2308 wps0.803 perplexity: 305.817 speed: 2313 wps0.903 perplexity: 286.216 speed: 2314 wpsEpoch: 1 Train Perplexity: 271.636Epoch: 1 Valid Perplexity: 179.177Epoch: 2 Learning rate: 1.0000.004 perplexity: 213.710 speed: 2302 wps0.104 perplexity: 152.220 speed: 2246 wps0.204 perplexity: 159.639 speed: 2252 wps0.304 perplexity: 154.315 speed: 2268 wps0.404 perplexity: 151.337 speed: 2278 wps0.504 perplexity: 148.721 speed: 2284 wps0.604 perplexity: 144.037 speed: 2287 wps0.703 perplexity: 141.835 speed: 2289 wps0.803 perplexity: 139.766 speed: 2292 wps0.903 perplexity: 136.026 speed: 2295 wpsEpoch: 2 Train Perplexity: 133.924Epoch: 2 Valid Perplexity: 142.841Epoch: 3 Learning rate: 1.0000.004 perplexity: 147.581 speed: 2340 wps0.104 perplexity: 105.235 speed: 2280 wps0.204 perplexity: 114.468 speed: 2305 wps0.304 perplexity: 111.488 speed: 2321 wps0.404 perplexity: 110.555 speed: 2330 wps0.504 perplexity: 109.632 speed: 2334 wps0.604 perplexity: 106.963 speed: 2314 wps0.703 perplexity: 106.316 speed: 2314 wps0.803 perplexity: 105.629 speed: 2316 wps0.903 perplexity: 103.380 speed: 2321 wpsEpoch: 3 Train Perplexity: 102.406Epoch: 3 Valid Perplexity: 132.434Epoch: 4 Learning rate: 1.0000.004 perplexity: 116.187 speed: 2358 wps0.104 perplexity: 85.211 speed: 2358 wps0.204 perplexity: 93.739 speed: 2359 wps0.304 perplexity: 91.477 speed: 2357 wps0.404 perplexity: 91.094 speed: 2358 wps0.504 perplexity: 90.708 speed: 2358 wps0.604 perplexity: 88.812 speed: 2359 wps0.703 perplexity: 88.559 speed: 2359 wps0.803 perplexity: 88.274 speed: 2358 wps0.903 perplexity: 86.596 speed: 2358 wpsEpoch: 4 Train Perplexity: 86.016Epoch: 4 Valid Perplexity: 127.497Epoch: 5 Learning rate: 0.5000.004 perplexity: 98.114 speed: 2364 wps0.104 perplexity: 71.122 speed: 2361 wps0.204 perplexity: 77.194 speed: 2360 wps0.304 perplexity: 74.387 speed: 2359 wps0.404 perplexity: 73.335 speed: 2359 wps0.504 perplexity: 72.389 speed: 2360 wps0.604 perplexity: 70.266 speed: 2360 wps0.703 perplexity: 69.468 speed: 2360 wps0.803 perplexity: 68.668 speed: 2360 wps0.903 perplexity: 66.766 speed: 2359 wpsEpoch: 5 Train Perplexity: 65.747Epoch: 5 Valid Perplexity: 120.065Epoch: 6 Learning rate: 0.2500.004 perplexity: 80.613 speed: 2357 wps0.104 perplexity: 58.697 speed: 2360 wps0.204 perplexity: 63.925 speed: 2355 wps0.304 perplexity: 61.566 speed: 2355 wps0.404 perplexity: 60.633 speed: 2355 wps0.504 perplexity: 59.727 speed: 2355 wps0.604 perplexity: 57.836 speed: 2356 wps0.703 perplexity: 57.067 speed: 2356 wps0.803 perplexity: 56.263 speed: 2357 wps0.903 perplexity: 54.525 speed: 2357 wpsEpoch: 6 Train Perplexity: 53.516Epoch: 6 Valid Perplexity: 119.491Epoch: 7 Learning rate: 0.1250.004 perplexity: 71.696 speed: 2363 wps0.104 perplexity: 52.006 speed: 2311 wps0.204 perplexity: 56.792 speed: 2334 wps0.304 perplexity: 54.663 speed: 2343 wps0.404 perplexity: 53.846 speed: 2344 wps0.504 perplexity: 53.023 speed: 2348 wps0.604 perplexity: 51.302 speed: 2350 wps0.703 perplexity: 50.576 speed: 2350 wps0.803 perplexity: 49.791 speed: 2352 wps0.903 perplexity: 48.191 speed: 2353 wpsEpoch: 7 Train Perplexity: 47.237Epoch: 7 Valid Perplexity: 120.102Epoch: 8 Learning rate: 0.0620.004 perplexity: 67.321 speed: 2368 wps0.104 perplexity: 48.715 speed: 2366 wps0.204 perplexity: 53.230 speed: 2366 wps0.304 perplexity: 51.194 speed: 2362 wps0.404 perplexity: 50.431 speed: 2356 wps0.504 perplexity: 49.684 speed: 2337 wps0.604 perplexity: 48.034 speed: 2335 wps0.703 perplexity: 47.328 speed: 2335 wps0.803 perplexity: 46.551 speed: 2339 wps0.903 perplexity: 45.012 speed: 2339 wpsEpoch: 8 Train Perplexity: 44.086Epoch: 8 Valid Perplexity: 120.569Epoch: 9 Learning rate: 0.0310.004 perplexity: 64.806 speed: 2223 wps0.104 perplexity: 46.999 speed: 2231 wps0.204 perplexity: 51.382 speed: 2219 wps0.304 perplexity: 49.378 speed: 2256 wps0.404 perplexity: 48.638 speed: 2277 wps0.504 perplexity: 47.938 speed: 2289 wps0.604 perplexity: 46.331 speed: 2294 wps0.703 perplexity: 45.626 speed: 2297 wps0.803 perplexity: 44.855 speed: 2294 wps0.903 perplexity: 43.347 speed: 2288 wpsEpoch: 9 Train Perplexity: 42.432Epoch: 9 Valid Perplexity: 120.513Epoch: 10 Learning rate: 0.0160.004 perplexity: 63.348 speed: 2329 wps