import numpy as np import tensorflow as tf #TF 1.1.0rc1 tf.logging.set_verbosity(tf.logging.ERROR) #import matplotlib.pyplot as plt from tsc_model import Model, sample_batch, load_data import os #Set these directories #direc = '/home/rob/Dropbox/ml_projects/LSTM/UCR_TS_Archive_2015' #summaries_dir = '/home/rob/Dropbox/ml_projects/LSTM_TSC/log_tb' """Load the data""" #ratio = np.array([0.8,0.9]) #Ratios where to split the training and validation set X_train, X_test, y_train, y_test = load_data() N, sl = X_train.shape num_classes = len(np.unique(y_train)) """Hyperparamaters""" batch_size = 100 max_iterations = 20000 dropout = 0.5 config = { 'num_layers': 3, #number of layers of stacked RNN's 'hidden_size': 16, #memory cells in a layer 'max_grad_norm': 5, #maximum gradient norm during training 'batch_size': batch_size, 'learning_rate': .005, 'sl': sl, 'num_classes': num_classes } epochs = np.floor(batch_size * max_iterations / N) #print('Train %.0f samples in approximately %d epochs' %(N,epochs))
"""Hyperparamaters""" config = {} #Put all configuration information into the dict config['num_layers'] = 80 #number of layers of stacked RNN's config['hidden_size'] = 40 #memory cells in a layer config['max_grad_norm'] = 100 #maximum gradient norm during training config['batch_size'] = batch_size = 30 config['learning_rate'] = .0001 max_iterations = 5000 dropout = 0.8 ratio = np.array([0.8,0.9]) #Ratios where to split the training and validation set """Load the data""" direc = '/home/kyle/dvlp/ml/LSTM_tsc/UCR_TS_Archive_2015' X_train,X_val,X_test,y_train,y_val,y_test = load_data(direc,ratio,dataset='IXIC') N,sl = X_train.shape config['sl'] = sl = X_train.shape[1] config['num_classes'] = num_classes = len(np.unique(y_train)) # Collect the costs in a numpy fashion epochs = np.floor(batch_size*max_iterations / N) print('Train %.0f samples in approximately %d epochs' %(N,epochs)) perf_collect = np.zeros((4,int(np.floor(max_iterations /100)))) #Instantiate a model model = Model(config) """Session time"""
generates an output to be classified with Softmax """ import numpy as np import tensorflow as tf #TF 1.1.0rc1 tf.logging.set_verbosity(tf.logging.ERROR) import matplotlib.pyplot as plt from tsc_model import Model, sample_batch, load_data, check_test #Set these directories direc = '/home/rob/Dropbox/ml_projects/LSTM/UCR_TS_Archive_2015' summaries_dir = '/home/rob/Dropbox/ml_projects/LSTM_TSC/log_tb' """Load the data""" ratio = np.array([0.8, 0.9]) #Ratios where to split the training and validation set X_train, X_val, X_test, y_train, y_val, y_test = load_data( direc, ratio, dataset='ChlorineConcentration') N, sl = X_train.shape num_classes = len(np.unique(y_train)) """Hyperparamaters""" batch_size = 30 max_iterations = 3000 dropout = 0.8 config = { 'num_layers': 3, #number of layers of stacked RNN's 'hidden_size': 120, #memory cells in a layer 'max_grad_norm': 5, #maximum gradient norm during training 'batch_size': batch_size, 'learning_rate': .005, 'sl': sl, 'num_classes': num_classes }
import numpy as np import tensorflow as tf #TF 1.1.0rc1 tf.logging.set_verbosity(tf.logging.ERROR) import matplotlib.pyplot as plt from tsc_model import Model,sample_batch,load_data ratio = np.array([0.8,0.9]) #Ratios where to split the training and validation set X_train,X_val,X_test,y_train,y_val,y_test = load_data(ratio,dataset='new_features.csv') N,sl = X_train.shape num_classes = len(np.unique(y_train)) batch_size = 30 max_iterations = 3000 dropout = 0.8 config = { 'num_layers' : 3, #number of layers of stacked RNN's 'hidden_size' : 120, #memory cells in a layer 'max_grad_norm' : 5, #maximum gradient norm during training 'batch_size' : batch_size, 'learning_rate' : .005, 'sl': sl, 'num_classes': num_classes} epochs = np.floor(batch_size*max_iterations / N) print('Train %.0f samples in approximately %d epochs' %(N,epochs)) #Instantiate a model model = Model(config)
generates an output to be classified with Softmax """ import numpy as np import tensorflow as tf #TF 1.1.0rc1 tf.logging.set_verbosity(tf.logging.ERROR) import matplotlib.pyplot as plt from tsc_model import Model, sample_batch, load_data #Set these directories direc = 'data_half' summaries_dir = 'log_tb' """Load the data""" ratio = np.array([0.8, 0.9]) #Ratios where to split the training and validation set X_train, X_val, X_test, y_train, y_val, y_test = load_data( direc, ratio, dataset='ArmSensor_s1') N, sl = X_train.shape num_classes = len(np.unique(y_train)) """Hyperparamaters""" batch_size = 30 max_iterations = 5000 dropout = 0.8 config = { 'num_layers': 3, #number of layers of stacked RNN's 'hidden_size': 120, #memory cells in a layer 'max_grad_norm': 5, #maximum gradient norm during training 'batch_size': batch_size, 'learning_rate': .005, 'sl': sl, 'num_classes': num_classes }
generates an output to be classified with Softmax """ import numpy as np import tensorflow as tf #TF 1.1.0rc1 tf.logging.set_verbosity(tf.logging.ERROR) import matplotlib.pyplot as plt from tsc_model import Model,sample_batch,load_data,check_test #Set these directories direc = '/home/rob/Dropbox/ml_projects/LSTM/UCR_TS_Archive_2015' summaries_dir = '/home/rob/Dropbox/ml_projects/LSTM_TSC/log_tb' """Load the data""" ratio = np.array([0.8,0.9]) #Ratios where to split the training and validation set X_train,X_val,X_test,y_train,y_val,y_test = load_data(direc,ratio,dataset='ChlorineConcentration') N,sl = X_train.shape num_classes = len(np.unique(y_train)) """Hyperparamaters""" batch_size = 30 max_iterations = 3000 dropout = 0.8 config = { 'num_layers' : 3, #number of layers of stacked RNN's 'hidden_size' : 120, #memory cells in a layer 'max_grad_norm' : 5, #maximum gradient norm during training 'batch_size' : batch_size, 'learning_rate' : .005, 'sl': sl, 'num_classes': num_classes}