源码.txt
立即下载
是吗@
2025-03-24
data
np.
float32
np.array
c1
data
label
False
input
Size
2.7 KB
import numpy as npimport scipy.io as scioimport pandas as pdimport tensorflow as tffrom itertools import chainglobal graph,modelgraph = tf.get_default_graph()mat = 'C://test.mat'data = scio.loadmat(mat)print(list(data))a = data['Data1_AI_1']b = data['Data1_AI_2']c = data['Data1_time_AI_1']d = data['Data1_AI_3']a1=a.astype(np.float32)b1=b.astype(np.float32)c1=c.astype(np.float32)d1=d.astype(np.float32)a2 = np.array(a1)b2 = np.array(b1)c2 = np.array(c1)d0 = np.around(d1)d2_1 = np.array(d0)x = np.stack((a1, b1, c1), axis=1)d2 = list(chain(*d2_1))d3 = pd.get_dummies(d2)def next_batch(): input_queue = tf.train.slice_input_producer([x, d3], shuffle=False, num_epochs=1) data_batch, label_batch = tf.train.batch(input_queue, batch_size=5, num_threads=1, capacity=20, allow_smaller_final_batch=False) return data_batch, label_batch#搭建神经网络numClasses = 138 inputSize = 3 numHiddenUnits = 4 trainingIterations = 10000 batchSize = 100 numHiddenUnitsLa
data/np./float32/np.array/c1/data_batch/label_batch/False/input_queue/Size/
data/np./float32/np.array/c1/data_batch/label_batch/False/input_queue/Size/
-->