{ "metadata": { "name": "stacked_autoencoders" }, "nbformat": 3, "nbformat_minor": 0, "worksheets": [ { "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Stacked Autoencoders\n", "by [Mehdi Mirza](http://www-etud.iro.umontreal.ca/~mirzamom/)\n", "\n", "## Introduction\n", "This notebook will show you how to perform layer-wise pre-training using denoising autoencoders (DAEs), and subsequently stack the layers to form a multilayer perceptron (MLP) which can be fine-tuned using supervised training. You can also look at this [more detailed tutorial of training DAEs using Theano](http://deeplearning.net/tutorial/dA.html#daa) as well as [this tutorial](http://deeplearning.net/tutorial/SdA.html#sda) which covers the stacked version.\n", "\n", "The methods used here can easily be adapted to other models such as contractive auto-encoders (CAEs) or restricted Boltzmann machines (RBMs) with only small modifications." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## First layer\n", "\n", "The first layer and its training algorithm are defined in the file `dae_l1.yaml`. Here we load the model and set some of its hypyerparameters." ] }, { "cell_type": "code", "collapsed": false, "input": [ "layer1_yaml = open('dae_l1.yaml', 'r').read()\n", "hyper_params_l1 = {'train_stop' : 50000,\n", " 'batch_size' : 100,\n", " 'monitoring_batches' : 5,\n", " 'nhid' : 500,\n", " 'max_epochs' : 10,\n", " 'save_path' : '.'}\n", "layer1_yaml = layer1_yaml % (hyper_params_l1)\n", "print layer1_yaml" ], "language": "python", "metadata": {}, "outputs": [ { "output_type": "stream", "stream": "stdout", "text": [ "!obj:pylearn2.train.Train {\n", " dataset: &train !obj:pylearn2.datasets.mnist.MNIST {\n", " which_set: 'train',\n", " start: 0,\n", " stop: 50000\n", " },\n", " model: !obj:pylearn2.models.autoencoder.DenoisingAutoencoder {\n", " nvis : 784,\n", " nhid : 500,\n", " irange : 0.05,\n", " corruptor: !obj:pylearn2.corruption.BinomialCorruptor {\n", " corruption_level: .2,\n", " },\n", " act_enc: \"tanh\",\n", " act_dec: null, # Linear activation on the decoder side.\n", " },\n", " algorithm: !obj:pylearn2.training_algorithms.sgd.SGD {\n", " learning_rate : 1e-3,\n", " batch_size : 100,\n", " monitoring_batches : 5,\n", " monitoring_dataset : *train,\n", " cost : !obj:pylearn2.costs.autoencoder.MeanSquaredReconstructionError {},\n", " termination_criterion : !obj:pylearn2.termination_criteria.EpochCounter {\n", " max_epochs: 10,\n", " },\n", " },\n", " save_path: \"./dae_l1.pkl\",\n", " save_freq: 1\n", "}\n", "\n" ] } ], "prompt_number": 4 }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now we can train the model using the YAML string in the same way as the previous tutorials:" ] }, { "cell_type": "code", "collapsed": false, "input": [ "from pylearn2.config import yaml_parse\n", "train = yaml_parse.load(layer1_yaml)\n", "train.main_loop()" ], "language": "python", "metadata": {}, "outputs": [ { "output_type": "stream", "stream": "stdout", "text": [ "Parameter and initial learning rate summary:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvb: 0.0010000000475\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\thb: 0.0010000000475\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tW: 0.0010000000475\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tWprime: 0.0010000000475\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Compiling sgd_update...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Compiling sgd_update done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "compiling begin_record_entry...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "compiling begin_record_entry done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitored channels: \n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Compiling accum...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "graph size: 23\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Compiling accum done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 85.4375915527\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 1\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 50000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 29.1613636017\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 2\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 1000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 100000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 21.9736881256\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 3\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 1500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 150000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 18.4479560852\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 4\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 2000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 200000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 16.2897148132\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 5\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 2500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 250000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 14.8111886978\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 6\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 3000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 300000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 13.6504278183\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 7\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 3500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 350000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 12.9274587631\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 8\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 4000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 400000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 12.2765922546\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 9\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 4500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 450000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 11.7446937561\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 10\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 5000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 500000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 11.4141273499\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l1.pkl done. Time elapsed: 1.000000 seconds\n" ] } ], "prompt_number": 5 }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Second layer\n", "The second layer takes the output of the first layer as its input. Hence we must first apply the first layer's transformations to the raw data using `datasets.transformer_dataset.TransformerDataset`. This class takes two arguments:\n", "\n", " - `raw`: the raw data\n", " - `transformer`: a Pylearn2 block that transforms the raw data, which in our case is the `dae_l1.pkl` file from the previous step\n", "\n", "To train the second layer, we load the YAML file as before and set the hyperparameters before starting the training loop." ] }, { "cell_type": "code", "collapsed": false, "input": [ "layer2_yaml = open('dae_l2.yaml', 'r').read()\n", "hyper_params_l2 = {'train_stop' : 50000,\n", " 'batch_size' : 100,\n", " 'monitoring_batches' : 5,\n", " 'nvis' : hyper_params_l1['nhid'],\n", " 'nhid' : 500,\n", " 'max_epochs' : 10,\n", " 'save_path' : '.'}\n", "layer2_yaml = layer2_yaml % (hyper_params_l2)\n", "print layer2_yaml" ], "language": "python", "metadata": {}, "outputs": [ { "output_type": "stream", "stream": "stdout", "text": [ "!obj:pylearn2.train.Train {\n", " dataset: &train !obj:pylearn2.datasets.transformer_dataset.TransformerDataset {\n", " raw: !obj:pylearn2.datasets.mnist.MNIST {\n", " which_set: 'train',\n", " start: 0,\n", " stop: 50000\n", " },\n", " transformer: !pkl: \"./dae_l1.pkl\"\n", " },\n", " model: !obj:pylearn2.models.autoencoder.DenoisingAutoencoder {\n", " nvis : 500,\n", " nhid : 500,\n", " irange : 0.05,\n", " corruptor: !obj:pylearn2.corruption.BinomialCorruptor {\n", " corruption_level: .3,\n", " },\n", " act_enc: \"tanh\",\n", " act_dec: null, # Linear activation on the decoder side.\n", " },\n", " algorithm: !obj:pylearn2.training_algorithms.sgd.SGD {\n", " learning_rate : 1e-3,\n", " batch_size : 100,\n", " monitoring_batches : 5,\n", " monitoring_dataset : *train,\n", " cost : !obj:pylearn2.costs.autoencoder.MeanSquaredReconstructionError {},\n", " termination_criterion : !obj:pylearn2.termination_criteria.EpochCounter {\n", " max_epochs: 10,\n", " },\n", " },\n", " save_path: \"./dae_l2.pkl\",\n", " save_freq: 1\n", "}\n", "\n" ] } ], "prompt_number": 6 }, { "cell_type": "code", "collapsed": false, "input": [ "train = yaml_parse.load(layer2_yaml)\n", "train.main_loop()" ], "language": "python", "metadata": {}, "outputs": [ { "output_type": "stream", "stream": "stdout", "text": [ "Parameter and initial learning rate summary:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvb: 0.0010000000475\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\thb: 0.0010000000475\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tW: 0.0010000000475\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tWprime: 0.0010000000475\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Compiling sgd_update...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Compiling sgd_update done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "compiling begin_record_entry...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "compiling begin_record_entry done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitored channels: \n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Compiling accum...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "graph size: 23\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Compiling accum done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 51.0506210327\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 1\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 50000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 20.0142116547\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 2\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 1000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 100000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 12.8833475113\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 3\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 1500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 150000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 9.65194129944\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 4\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 2000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 200000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 7.71482992172\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 5\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 2500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 250000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 6.5238275528\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 6\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 3000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 300000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 5.69179153442\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 7\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 3500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 350000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 5.15888118744\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 8\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 4000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 400000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 4.75159025192\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 9\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 4500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 450000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 4.38682460785\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 10\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 5000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 500000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.00100000016391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tobjective: 4.21171569824\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Saving to ./dae_l2.pkl done. Time elapsed: 0.000000 seconds\n" ] } ], "prompt_number": 7 }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Supervised fine-tuning\n", "Now that we have two pre-trained layers, we can stack them to form an MLP which can be trained in a supervised fashion. We use the MLP class as usual for this, except that we now use `models.mlp.PretrainedLayer` for the different layers so that we can pass our pre-trained layers (as pickle files) using the `layer_content` argument." ] }, { "cell_type": "code", "collapsed": false, "input": [ "mlp_yaml = open('dae_mlp.yaml', 'r').read()\n", "hyper_params_mlp = {'train_stop' : 50000,\n", " 'valid_stop' : 60000,\n", " 'batch_size' : 100,\n", " 'max_epochs' : 50,\n", " 'save_path' : '.'}\n", "mlp_yaml = mlp_yaml % (hyper_params_mlp)\n", "print mlp_yaml" ], "language": "python", "metadata": {}, "outputs": [ { "output_type": "stream", "stream": "stdout", "text": [ "!obj:pylearn2.train.Train {\n", " dataset: &train !obj:pylearn2.datasets.mnist.MNIST {\n", " which_set: 'train',\n", " start: 0,\n", " stop: 50000\n", " },\n", " model: !obj:pylearn2.models.mlp.MLP {\n", " batch_size: 100,\n", " layers: [\n", " !obj:pylearn2.models.mlp.PretrainedLayer {\n", " layer_name: 'h1',\n", " layer_content: !pkl: \"./dae_l1.pkl\"\n", " },\n", " !obj:pylearn2.models.mlp.PretrainedLayer {\n", " layer_name: 'h2',\n", " layer_content: !pkl: \"./dae_l2.pkl\"\n", " },\n", " !obj:pylearn2.models.mlp.Softmax {\n", " max_col_norm: 1.9365,\n", " layer_name: 'y',\n", " n_classes: 10,\n", " irange: .005\n", " }\n", " ],\n", " nvis: 784\n", " },\n", " algorithm: !obj:pylearn2.training_algorithms.sgd.SGD {\n", " learning_rate: .05,\n", " learning_rule: !obj:pylearn2.training_algorithms.learning_rule.Momentum {\n", " init_momentum: .5,\n", " },\n", " monitoring_dataset:\n", " {\n", " 'valid' : !obj:pylearn2.datasets.mnist.MNIST {\n", " which_set: 'train',\n", " start: 50000,\n", " stop: 60000\n", " },\n", " },\n", " cost: !obj:pylearn2.costs.mlp.Default {},\n", " termination_criterion: !obj:pylearn2.termination_criteria.And {\n", " criteria: [\n", " !obj:pylearn2.termination_criteria.MonitorBased {\n", " channel_name: \"valid_y_misclass\",\n", " prop_decrease: 0.,\n", " N: 100\n", " },\n", " !obj:pylearn2.termination_criteria.EpochCounter {\n", " max_epochs: 50\n", " }\n", " ]\n", " },\n", " update_callbacks: !obj:pylearn2.training_algorithms.sgd.ExponentialDecay {\n", " decay_factor: 1.00004,\n", " min_lr: .000001\n", " }\n", " },\n", " extensions: [\n", " !obj:pylearn2.training_algorithms.learning_rule.MomentumAdjustor {\n", " start: 1,\n", " saturate: 250,\n", " final_momentum: .7\n", " }\n", " ]\n", "}\n", "\n" ] } ], "prompt_number": 8 }, { "cell_type": "code", "collapsed": false, "input": [ "train = yaml_parse.load(mlp_yaml)\n", "train.main_loop()" ], "language": "python", "metadata": {}, "outputs": [ { "output_type": "stream", "stream": "stdout", "text": [ "Parameter and initial learning rate summary:\n" ] }, { "output_type": "stream", "stream": "stderr", "text": [ "/data/lisa/exp/mirzamom/pylearn2/pylearn2/models/mlp.py:41: UserWarning: MLP changing the recursion limit.\n", " warnings.warn(\"MLP changing the recursion limit.\")\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvb: 0.0500000007451\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\thb: 0.0500000007451\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tW: 0.0500000007451\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tWprime: 0.0500000007451\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvb: 0.0500000007451\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\thb: 0.0500000007451\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tW: 0.0500000007451\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tWprime: 0.0500000007451\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tsoftmax_b: 0.0500000007451\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tsoftmax_W: 0.0500000007451\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Compiling sgd_update...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Compiling sgd_update done. Time elapsed: 51.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "compiling begin_record_entry...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "compiling begin_record_entry done. Time elapsed: 0.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitored channels: \n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Compiling accum...\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "graph size: 75\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Compiling accum done. Time elapsed: 31.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0500000119209\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.499999672174\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.0\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 2.30245757103\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 0.0650026649237\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 0.0641745403409\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 0.0624679774046\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.10553213954\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.102753870189\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.101059176028\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.903100371361\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 2.30245757103\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.0125483665615\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.00897720176727\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.00411556242034\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 1\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 50000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0490099266171\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.499999672174\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.285481214523\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.37920033932\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.25995886326\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.10580408573\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999643802643\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.891385912895\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.366638094187\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0814000219107\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.285481214523\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.306006103754\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.173898175359\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0752066597342\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 2\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 1000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 100000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0480394884944\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.500803589821\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.247136443853\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.53969144821\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.40233445168\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.25563120842\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999809861183\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.914137363434\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.396682620049\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.069399997592\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.247136443853\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.348902791739\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.193130522966\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0754316821694\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 3\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 1500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 150000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0470883138478\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.501606047153\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.209606900811\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.67392218113\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.51739025116\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.41721081734\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999855041504\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.925868034363\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.405808866024\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.06040000543\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.209606900811\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.398027926683\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.20821505785\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0778625309467\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 4\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 2000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 200000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0461559444666\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.502409934998\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.181997314095\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.88737154007\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.62824416161\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.44828641415\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999894917011\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.934701681137\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.424763649702\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0520000010729\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.181997314095\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.444758623838\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.222617387772\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0790278464556\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 5\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 2500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 250000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0452419146895\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.50321239233\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.159930184484\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.9364978075\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.72115278244\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.47439146042\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.99988681078\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.940866410732\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.426196664572\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0440000146627\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.159930184484\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.464266389608\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.234414324164\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0797937735915\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 6\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 3000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 300000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0443461276591\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.504016280174\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.143035233021\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93126213551\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.79720795155\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.52031481266\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999934792519\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.948293268681\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.448669195175\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0376999974251\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.143035233021\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.501182496548\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.244007915258\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0815980285406\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 7\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 3500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 350000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0434680506587\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.504818737507\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.128972783685\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93631577492\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.84140181541\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.56303739548\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.99993532896\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.952728152275\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.457730174065\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0372999943793\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.128972783685\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.52207928896\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.249332204461\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0810364559293\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 8\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 4000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 400000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0426072925329\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.505622982979\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.123533077538\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.9364978075\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.87326931953\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.61571848392\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999963104725\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.954613864422\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.463554471731\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0348999910057\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.123533077538\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.525155007839\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.253258258104\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0812314674258\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 9\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 4500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 450000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0417636223137\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.506425499916\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.119187682867\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.89299559593\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.65916585922\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999965846539\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.95620149374\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.464787423611\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0323999859393\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.119187682867\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.534337043762\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.255589127541\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0810972675681\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 10\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 5000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 500000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0409367084503\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.50722938776\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.107577241957\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.9364978075\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.90345025063\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.70279407501\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999951183796\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.960036695004\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.468458265066\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0300999823958\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.107577241957\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.542799532413\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.256767898798\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0823005959392\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 11\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 5500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 550000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0401261113584\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.508031845093\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.107919149101\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.9149273634\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.76190459728\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999973893166\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.959668278694\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.47409799695\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0300999861211\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.107919149101\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.550510644913\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.258117824793\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0835975408554\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 12\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 6000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 600000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0393316075206\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.508835673332\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0998769327998\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.92226481438\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.80892860889\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999977052212\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.964593172073\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.500402808189\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0274999812245\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0998769327998\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.559845209122\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.259052544832\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0850235819817\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 13\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 6500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 650000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0385527797043\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.509638190269\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0978430137038\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.9364978075\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.92790329456\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.8607878685\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999977111816\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.964517354965\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.493009746075\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0281999818981\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0978430137038\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.565926074982\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.259754091501\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0865102484822\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 14\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 7000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 700000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0377893745899\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.510442078114\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0951417461038\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93095195293\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.90848994255\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999983549118\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.965570628643\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.493750423193\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0279999841005\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0951417461038\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.57372456789\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.26015779376\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0874916240573\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 15\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 7500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 750000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0370411500335\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.511244595051\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0946910232306\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.9364978075\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93529140949\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93278777599\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999984383583\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.966287732124\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.497158616781\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0266999825835\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0946910232306\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.576683402061\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.260768920183\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0881127864122\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 16\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 8000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 800000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0363076739013\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.512048363686\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.089107722044\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.9364978075\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.9356637001\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93223702908\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999985218048\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.96820807457\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.502092540264\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0256999880075\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.089107722044\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.57947987318\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.260900110006\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0890503451228\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 17\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 8500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 850000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0355887822807\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.512850999832\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0881613865495\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.9364978075\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93482923508\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93224895\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999977946281\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.968540728092\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.502689242363\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0259999874979\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0881613865495\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.581995129585\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.260852187872\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0897700637579\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 18\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 9000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 900000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.034884031862\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.513654768467\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0850231051445\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93542182446\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93202567101\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999984383583\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.969747781754\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.50995349884\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0240999888629\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0850231051445\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.582888245583\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.261032491922\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.091475315392\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 19\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 9500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 950000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0341933257878\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.514457404613\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0860132724047\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.9364978075\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93475472927\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93239700794\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999982178211\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.968550920486\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.500067353249\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.024499990046\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0860132724047\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.585309565067\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.261046379805\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0925423651934\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 20\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 10000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 1000000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0335162654519\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.515261173248\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.082815758884\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.9364978075\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93560194969\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93234658241\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999988675117\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.970959126949\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.511843323708\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0254999864846\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.082815758884\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.587334752083\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.261245340109\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0929176732898\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 21\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 10500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 1050000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0328526012599\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.516063690186\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0818511173129\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93555438519\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93388450146\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999989688396\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.972750782967\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.53289026022\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0240999888629\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0818511173129\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.584912240505\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.261357337236\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0945193096995\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 22\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 11000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 1100000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0322021208704\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.51686757803\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0818284451962\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93543183804\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93241846561\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999989748001\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.971523821354\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.512000918388\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0234999898821\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0818284451962\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.585798323154\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.261481463909\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0941896960139\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 23\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 11500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 1150000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0315644294024\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.517670154572\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0783765390515\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.9352645874\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93146395683\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999990105629\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.972935736179\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.526244282722\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0227999929339\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0783765390515\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.584616363049\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.261561661959\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0957764536142\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 24\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 12000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 1200000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0309394672513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.518473863602\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0788094773889\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93574666977\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93321406841\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999989807606\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.973162353039\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.517908155918\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0223999936134\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0788094773889\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.585705161095\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.261755138636\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0961646363139\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 2.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 25\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 12500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 1250000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0303268413991\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.519276380539\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 1.9999986887\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0773832127452\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93531489372\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93346488476\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999991297722\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.973752617836\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.529482901096\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0232999920845\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0773832127452\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.58470761776\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.26182243228\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0976147502661\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 26\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 13000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 1300000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0297263283283\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.520080327988\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0760994702578\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.9364978075\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93572044373\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93307471275\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999990880489\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.974325656891\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.535016596317\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0222999919206\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0760994702578\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.584625601768\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.262017458677\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0985018312931\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 27\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 13500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 1350000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0291377287358\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.520884275436\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0745258107781\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.9364978075\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93582701683\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93377733231\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999992668629\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.974617183208\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.528786301613\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0223999880254\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0745258107781\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.582556009293\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.262156039476\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.098942771554\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 28\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 14000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 1400000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0285607334226\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.521686851978\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0740825012326\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.9364978075\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93625664711\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93552911282\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999993383884\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.975292444229\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.532864153385\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0215999912471\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0740825012326\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.582411289215\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.262312680483\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.0992849618196\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 29\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 14500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 1450000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0279952250421\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.522490501404\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0735178291798\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93584036827\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93419444561\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999993622303\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.975481748581\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.530311584473\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0224999897182\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0735178291798\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.581294953823\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.262381464243\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.099995970726\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 30\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 15000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 1500000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.027440899983\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.523293077946\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0742838978767\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93606865406\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93509995937\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999993681908\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.975233256817\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.529448211193\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.021299989894\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0742838978767\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.579390466213\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.262531936169\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.101199530065\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 31\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 15500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 1550000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0268975384533\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.52409696579\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0728998035192\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.9364978075\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93537724018\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.9334436655\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999993503094\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.975039601326\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.530442178249\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0211999937892\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0728998035192\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.577543079853\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.262552529573\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.101914271712\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 32\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 16000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 1600000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0263649839908\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.524899542332\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0729000940919\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93582475185\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93382787704\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999994158745\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.976245224476\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.523617684841\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0215999912471\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0729000940919\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.575929939747\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.262723714113\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.102134265006\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 33\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 16500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 1650000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0258428994566\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.525703251362\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0711924284697\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.9364978075\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93601918221\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93474292755\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999995052814\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.976920008659\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.53466886282\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0214999932796\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0711924284697\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.575305998325\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.262870043516\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.102859780192\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 34\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 17000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 1700000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.025331215933\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.526505768299\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0699434652925\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93592369556\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93427860737\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999995589256\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.976588606834\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.526415586472\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0207999944687\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0699434652925\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.573732554913\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.262948900461\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.103134132922\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 35\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 17500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 1750000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0248296167701\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.527309715748\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0703471377492\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93556249142\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93350601196\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999995827675\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.977201640606\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.54014390707\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0216999910772\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0703471377492\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.569660007954\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.263015538454\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.103172667325\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 36\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 18000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 1800000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.024337939918\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.528112351894\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0702705159783\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93602788448\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.9353749752\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.99999576807\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.977866590023\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.538337528706\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0210999920964\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0702705159783\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.570420324802\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.263186216354\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.103417083621\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 37\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 18500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 1850000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0238560270518\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.52891600132\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0700398087502\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93587827682\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93338406086\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999996423721\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.977614223957\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.5420165658\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0208999942988\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0700398087502\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.5683183074\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.263258725405\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.102311193943\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 38\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 19000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 1900000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.023383660242\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.529718577862\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0708458870649\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93592023849\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93505263329\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999996066093\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.978010952473\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.540901720524\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0209999922663\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0708458870649\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.56648504734\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.263367444277\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.102654665709\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 39\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 19500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 1950000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0229206457734\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.530522465706\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0704958662391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93585407734\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.9334679842\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999996304512\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.978026509285\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.547220349312\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0213999915868\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0704958662391\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.563822031021\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.263476461172\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.102645337582\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 40\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 20000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 2000000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0224668364972\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.531325042248\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.069045573473\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.9364978075\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93611621857\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93514800072\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999996364117\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.978076577187\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.548255085945\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0201999917626\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.069045573473\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.561926782131\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.263601183891\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.102276921272\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 41\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 20500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 2050000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0220219288021\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.532128691673\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0694609582424\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93617403507\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93555283546\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999996304512\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.97806340456\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.5496789217\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0206999927759\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0694609582424\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.559349894524\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.263698577881\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.102876082063\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 42\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 21000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 2100000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0215858761221\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.532931268215\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0682552531362\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.9364978075\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93583071232\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93494808674\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999996244907\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.978460967541\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.536799430847\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0206999927759\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0682552531362\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.558497548103\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.263731598854\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.102174289525\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 43\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 21500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 2150000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0211584754288\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.533735215664\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.068164549768\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93625628948\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93569409847\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999996840954\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.978974223137\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.545736849308\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.020799992606\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.068164549768\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.55761551857\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.263863831758\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.102318763733\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 44\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 22000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 2200000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0207395013422\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.534537792206\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0678072869778\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93612587452\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93490695953\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999996721745\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.978856146336\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.54448735714\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0202999915928\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0678072869778\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.557243168354\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.263932317495\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.102473787963\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 45\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 22500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 2250000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0203288514167\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.535341382027\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0676843225956\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93607878685\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93488621712\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999997019768\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.979203939438\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.541955649853\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0211999919266\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0676843225956\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.554672718048\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.263984143734\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.102155432105\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 46\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 23000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 2300000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0199263226241\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.536144316196\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0666035562754\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.9364978075\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93638789654\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93599748611\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999997377396\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.979231536388\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.552667915821\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0199999958277\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0666035562754\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.552026212215\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.264121174812\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.102099023759\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 47\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 23500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 2350000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0195317566395\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.536948263645\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0681700259447\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.9364978075\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93598556519\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93464744091\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999997317791\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.979587137699\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.54142510891\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0199999958277\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0681700259447\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.550442278385\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.264142274857\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.101656988263\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 48\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 24000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 2400000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0191450119019\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.537750899792\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.069911248982\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93574678898\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.9335873127\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999997794628\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.979529380798\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.541589438915\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0213999953121\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.069911248982\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.548533499241\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.264166146517\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.101313956082\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 49\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 24500\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 2450000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0187659449875\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.538554370403\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0663670599461\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93616163731\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93532729149\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999997615814\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.979907333851\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.55751311779\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.0203999932855\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0663670599461\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.54831713438\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.26429900527\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.101693704724\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Time this epoch: 1.000000 seconds\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "Monitoring step:\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tEpochs seen: 50\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tBatches seen: 25000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tExamples seen: 2500000\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tlearning_rate: 0.0183943510056\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmomentum: 0.53935700655\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tmonitor_seconds_per_epoch: 0.999999344349\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_objective: 0.0667693391442\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_max: 1.93649816513\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_mean: 1.93614006042\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_col_norms_min: 1.93519842625\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_max_max_class: 0.999997913837\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_mean_max_class: 0.980073690414\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_min_max_class: 0.548149049282\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_misclass: 0.019999993965\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_nll: 0.0667693391442\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_max: 0.546491324902\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_mean: 0.26435393095\n" ] }, { "output_type": "stream", "stream": "stdout", "text": [ "\tvalid_y_row_norms_min: 0.10142172128\n" ] } ], "prompt_number": 9 } ], "metadata": {} } ] }