吴茱萸汤训练的数据

cd /root/wuzhuyu-agent && node src/ml/train.js

2025-05-15 10:28:28.775064: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
开始训练吴茱萸汤医案识别模型...
正在连接MongoDB...
MongoDB连接成功
正在获取医案数据...
成功获取 5 篇医案
正在预处理医案文本...
中医术语词典加载完成
Compiling dictionary
预处理完成，有效医案数据: 5
正在创建词汇表...
词汇表创建完成，共 2114 词
正在准备训练数据...
正在创建残差网络模型...

Layer (type) Input Shape Output shape Param # Receives inputs

input1 (InputLayer) [[null,200]] [null,200] 0

embedding_Embedding1 (Embe [[null,200]] [null,200,128] 270592 input1[0][0]

conv1d_Conv1D1 (Conv1D) [[null,200,128]] [null,100,64] 57408 embedding_Embedding1[0]

batch_normalization_BatchN [[null,100,64]] [null,100,64] 256 conv1d_Conv1D1[0][0]

activation_Activation1 (Ac [[null,100,64]] [null,100,64] 0 batch_normalization_Bat

max_pooling1d_MaxPooling1D [[null,100,64]] [null,50,64] 0 activation_Activation1[

conv1d_Conv1D2 (Conv1D) [[null,50,64]] [null,50,64] 12352 max_pooling1d_MaxPoolin

batch_normalization_BatchN [[null,50,64]] [null,50,64] 256 conv1d_Conv1D2[0][0]

activation_Activation2 (Ac [[null,50,64]] [null,50,64] 0 batch_normalization_Bat

conv1d_Conv1D3 (Conv1D) [[null,50,64]] [null,50,64] 12352 activation_Activation2[

batch_normalization_BatchN [[null,50,64]] [null,50,64] 256 conv1d_Conv1D3[0][0]

add_Add1 (Add) [[null,50,64],[null,50,64]] [null,50,64] 0 batch_normalization_Bat
max_pooling1d_MaxPoolin

activation_Activation3 (Ac [[null,50,64]] [null,50,64] 0 add_Add1[0][0]

conv1d_Conv1D4 (Conv1D) [[null,50,64]] [null,50,64] 12352 activation_Activation3[

batch_normalization_BatchN [[null,50,64]] [null,50,64] 256 conv1d_Conv1D4[0][0]

activation_Activation4 (Ac [[null,50,64]] [null,50,64] 0 batch_normalization_Bat

conv1d_Conv1D5 (Conv1D) [[null,50,64]] [null,50,64] 12352 activation_Activation4[

batch_normalization_BatchN [[null,50,64]] [null,50,64] 256 conv1d_Conv1D5[0][0]

add_Add2 (Add) [[null,50,64],[null,50,64]] [null,50,64] 0 batch_normalization_Bat
activation_Activation3[

activation_Activation5 (Ac [[null,50,64]] [null,50,64] 0 add_Add2[0][0]

conv1d_Conv1D6 (Conv1D) [[null,50,64]] [null,25,128] 24704 activation_Activation5[

batch_normalization_BatchN [[null,25,128]] [null,25,128] 512 conv1d_Conv1D6[0][0]

activation_Activation6 (Ac [[null,25,128]] [null,25,128] 0 batch_normalization_Bat

conv1d_Conv1D7 (Conv1D) [[null,25,128]] [null,25,128] 49280 activation_Activation6[

conv1d_Conv1D8 (Conv1D) [[null,50,64]] [null,25,128] 8320 activation_Activation5[

batch_normalization_BatchN [[null,25,128]] [null,25,128] 512 conv1d_Conv1D7[0][0]

batch_normalization_BatchN [[null,25,128]] [null,25,128] 512 conv1d_Conv1D8[0][0]

add_Add3 (Add) [[null,25,128],[null,25,128 [null,25,128] 0 batch_normalization_Bat
batch_normalization_Bat

activation_Activation7 (Ac [[null,25,128]] [null,25,128] 0 add_Add3[0][0]

conv1d_Conv1D9 (Conv1D) [[null,25,128]] [null,25,128] 49280 activation_Activation7[

batch_normalization_BatchN [[null,25,128]] [null,25,128] 512 conv1d_Conv1D9[0][0]

activation_Activation8 (Ac [[null,25,128]] [null,25,128] 0 batch_normalization_Bat

conv1d_Conv1D10 (Conv1D) [[null,25,128]] [null,25,128] 49280 activation_Activation8[

batch_normalization_BatchN [[null,25,128]] [null,25,128] 512 conv1d_Conv1D10[0][0]

add_Add4 (Add) [[null,25,128],[null,25,128 [null,25,128] 0 batch_normalization_Bat
activation_Activation7[

activation_Activation9 (Ac [[null,25,128]] [null,25,128] 0 add_Add4[0][0]

conv1d_Conv1D11 (Conv1D) [[null,25,128]] [null,13,256] 98560 activation_Activation9[

batch_normalization_BatchN [[null,13,256]] [null,13,256] 1024 conv1d_Conv1D11[0][0]

activation_Activation10 (A [[null,13,256]] [null,13,256] 0 batch_normalization_Bat

conv1d_Conv1D12 (Conv1D) [[null,13,256]] [null,13,256] 196864 activation_Activation10

conv1d_Conv1D13 (Conv1D) [[null,25,128]] [null,13,256] 33024 activation_Activation9[

batch_normalization_BatchN [[null,13,256]] [null,13,256] 1024 conv1d_Conv1D12[0][0]

batch_normalization_BatchN [[null,13,256]] [null,13,256] 1024 conv1d_Conv1D13[0][0]

add_Add5 (Add) [[null,13,256],[null,13,256 [null,13,256] 0 batch_normalization_Bat
batch_normalization_Bat

activation_Activation11 (A [[null,13,256]] [null,13,256] 0 add_Add5[0][0]

conv1d_Conv1D14 (Conv1D) [[null,13,256]] [null,13,256] 196864 activation_Activation11

batch_normalization_BatchN [[null,13,256]] [null,13,256] 1024 conv1d_Conv1D14[0][0]

activation_Activation12 (A [[null,13,256]] [null,13,256] 0 batch_normalization_Bat

conv1d_Conv1D15 (Conv1D) [[null,13,256]] [null,13,256] 196864 activation_Activation12

batch_normalization_BatchN [[null,13,256]] [null,13,256] 1024 conv1d_Conv1D15[0][0]

add_Add6 (Add) [[null,13,256],[null,13,256 [null,13,256] 0 batch_normalization_Bat
activation_Activation11

activation_Activation13 (A [[null,13,256]] [null,13,256] 0 add_Add6[0][0]

conv1d_Conv1D16 (Conv1D) [[null,13,256]] [null,7,512] 393728 activation_Activation13

batch_normalization_BatchN [[null,7,512]] [null,7,512] 2048 conv1d_Conv1D16[0][0]

activation_Activation14 (A [[null,7,512]] [null,7,512] 0 batch_normalization_Bat

conv1d_Conv1D17 (Conv1D) [[null,7,512]] [null,7,512] 786944 activation_Activation14

conv1d_Conv1D18 (Conv1D) [[null,13,256]] [null,7,512] 131584 activation_Activation13

batch_normalization_BatchN [[null,7,512]] [null,7,512] 2048 conv1d_Conv1D17[0][0]

batch_normalization_BatchN [[null,7,512]] [null,7,512] 2048 conv1d_Conv1D18[0][0]

add_Add7 (Add) [[null,7,512],[null,7,512]] [null,7,512] 0 batch_normalization_Bat
batch_normalization_Bat

activation_Activation15 (A [[null,7,512]] [null,7,512] 0 add_Add7[0][0]

conv1d_Conv1D19 (Conv1D) [[null,7,512]] [null,7,512] 786944 activation_Activation15

batch_normalization_BatchN [[null,7,512]] [null,7,512] 2048 conv1d_Conv1D19[0][0]

activation_Activation16 (A [[null,7,512]] [null,7,512] 0 batch_normalization_Bat

conv1d_Conv1D20 (Conv1D) [[null,7,512]] [null,7,512] 786944 activation_Activation16

batch_normalization_BatchN [[null,7,512]] [null,7,512] 2048 conv1d_Conv1D20[0][0]

add_Add8 (Add) [[null,7,512],[null,7,512]] [null,7,512] 0 batch_normalization_Bat
activation_Activation15

activation_Activation17 (A [[null,7,512]] [null,7,512] 0 add_Add8[0][0]

global_average_pooling1d_G [[null,7,512]] [null,512] 0 activation_Activation17

dense_Dense1 (Dense) [[null,512]] [null,256] 131328 global_average_pooling1

wuzhuyu_prediction (Dense) [[null,256]] [null,1] 257 dense_Dense1[0][0]

Total params: 4317377
Trainable params: 4307777
Non-trainable params: 9600

开始训练模型...
Epoch 1 / 20
2025-05-15 10:32:46.061765: W tensorflow/core/framework/cpu_allocator_impl.cc:82] Allocation of 12057600 exceeds 10% of free system memory.
2025-05-15 10:32:46.079703: W tensorflow/core/framework/cpu_allocator_impl.cc:82] Allocation of 12057600 exceeds 10% of free system memory.
2025-05-15 10:32:46.453114: W tensorflow/core/framework/cpu_allocator_impl.cc:82] Allocation of 12057600 exceeds 10% of free system memory.
2025-05-15 10:32:46.509236: W tensorflow/core/framework/cpu_allocator_impl.cc:82] Allocation of 12057600 exceeds 10% of free system memory.
2025-05-15 10:32:46.894626: W tensorflow/core/framework/cpu_allocator_impl.cc:82] Allocation of 12057600 exceeds 10% of free system memory.
eta=0.0 ====================================================================>
49734ms 12433532us/step - acc=0.250 loss=1.34 val_acc=1.00 val_loss=0.679
Epoch 1: loss = 1.3399, accuracy = 0.2500
Epoch 2 / 20
eta=0.0 ====================================================================>
82171ms 20542705us/step - acc=1.00 loss=0.0563 val_acc=1.00 val_loss=0.659
Epoch 2: loss = 0.0563, accuracy = 1.0000
Epoch 3 / 20
eta=0.0 ====================================================================>
57168ms 14292114us/step - acc=1.00 loss=1.68e-3 val_acc=1.00 val_loss=0.636
Epoch 3: loss = 0.0017, accuracy = 1.0000
Epoch 4 / 20
eta=0.0 ====================================================================>
99601ms 24900137us/step - acc=1.00 loss=1.55e-3 val_acc=1.00 val_loss=0.609
Epoch 4: loss = 0.0015, accuracy = 1.0000
Epoch 5 / 20
eta=0.0 ====================================================================>
81431ms 20357799us/step - acc=1.00 loss=1.91e-3 val_acc=1.00 val_loss=0.581
Epoch 5: loss = 0.0019, accuracy = 1.0000
Epoch 6 / 20
eta=0.0 ====================================================================>
109351ms 27337840us/step - acc=1.00 loss=2.01e-3 val_acc=1.00 val_loss=0.550
Epoch 6: loss = 0.0020, accuracy = 1.0000
Epoch 7 / 20
eta=0.0 ====================================================================>
48353ms 12088250us/step - acc=1.00 loss=1.87e-3 val_acc=1.00 val_loss=0.518
Epoch 7: loss = 0.0019, accuracy = 1.0000
Epoch 8 / 20
eta=0.0 ====================================================================>
28527ms 7131650us/step - acc=1.00 loss=1.71e-3 val_acc=1.00 val_loss=0.486
Epoch 8: loss = 0.0017, accuracy = 1.0000
Epoch 9 / 20
eta=0.0 ====================================================================>
27138ms 6784606us/step - acc=1.00 loss=1.55e-3 val_acc=1.00 val_loss=0.453
Epoch 9: loss = 0.0016, accuracy = 1.0000
Epoch 10 / 20
eta=0.0 ====================================================================>
23171ms 5792816us/step - acc=1.00 loss=1.33e-3 val_acc=1.00 val_loss=0.421
Epoch 10: loss = 0.0013, accuracy = 1.0000
Epoch 11 / 20
eta=0.0 ====================================================================>
40645ms 10161174us/step - acc=1.00 loss=1.09e-3 val_acc=1.00 val_loss=0.389
Epoch 11: loss = 0.0011, accuracy = 1.0000
Epoch 12 / 20
eta=0.0 ====================================================================>
65354ms 16338470us/step - acc=1.00 loss=8.74e-4 val_acc=1.00 val_loss=0.359
Epoch 12: loss = 0.0009, accuracy = 1.0000
Epoch 13 / 20
eta=0.0 ====================================================================>
109955ms 27488664us/step - acc=1.00 loss=6.94e-4 val_acc=1.00 val_loss=0.330
Epoch 13: loss = 0.0007, accuracy = 1.0000
Epoch 14 / 20
eta=0.0 ====================================================================>
46097ms 11524321us/step - acc=1.00 loss=5.52e-4 val_acc=1.00 val_loss=0.303
Epoch 14: loss = 0.0006, accuracy = 1.0000
Epoch 15 / 20
eta=0.0 ====================================================================>
29641ms 7410351us/step - acc=1.00 loss=4.47e-4 val_acc=1.00 val_loss=0.277
Epoch 15: loss = 0.0004, accuracy = 1.0000
Epoch 16 / 20
eta=0.0 ====================================================================>
42078ms 10519595us/step - acc=1.00 loss=3.65e-4 val_acc=1.00 val_loss=0.252
Epoch 16: loss = 0.0004, accuracy = 1.0000
Epoch 17 / 20
eta=0.0 ====================================================================>
31305ms 7826247us/step - acc=1.00 loss=2.98e-4 val_acc=1.00 val_loss=0.229
Epoch 17: loss = 0.0003, accuracy = 1.0000
Epoch 18 / 20
eta=0.0 ====================================================================>
20645ms 5161216us/step - acc=1.00 loss=2.42e-4 val_acc=1.00 val_loss=0.208
Epoch 18: loss = 0.0002, accuracy = 1.0000
Epoch 19 / 20
eta=0.0 ====================================================================>
69281ms 17320244us/step - acc=1.00 loss=1.96e-4 val_acc=1.00 val_loss=0.189
Epoch 19: loss = 0.0002, accuracy = 1.0000
Epoch 20 / 20
eta=0.0 ====================================================================>
83496ms 20873937us/step - acc=1.00 loss=1.59e-4 val_acc=1.00 val_loss=0.171
Epoch 20: loss = 0.0002, accuracy = 1.0000
训练完成，正在保存模型...
模型已保存到 /root/wuzhuyu-agent/src/ml/models/wuzhuyu-resnet
模型训练和保存完成！

posted @ 2025-05-15 19:09  昆仑葫芦  阅读(1)  评论(0)    收藏  举报