UserWarning: Method on_batch_end() is slow compared to the batch update. Check your callbacks.

Keras训练verbose参数问题解决
最新推荐文章于 2022-11-01 15:28:28 发布
最新推荐文章于 2022-11-01 15:28:28 发布
阅读量4k 收藏 7
点赞数 4
CC 4.0 BY-SA版权
文章标签： 人工智能 python
原文链接：http://www.cnblogs.com/BlueBlueSea/p/11063712.html
博客提到GitHub上Keras相关问题（issue/2603），原因是berbose=1时，训练一个batch打印进度条和结果到控制台的时间比训练模型时间长而出现警告。解决办法是将verbose设为0，不打印训练过程信息；设为2，每个epoch打印一次。

https://github.com/keras-team/keras/issues/2603

history=model.fit(X_train,y_train,epochs=30,batch_size=32,validation_split=0.2,verbose=2)

出现的原因是一开始berbose=1（默认），在一个epoch中训练一个batch就会打印进度条和结果，但是打印到控制台所需时间比训练模型时间更长，所以有了警告。

将verbose=0，使得一点训练过程信息也不打印；verbose=2，每一个epoch打印一次。

 

转载于:https://www.cnblogs.com/BlueBlueSea/p/11063712.html

keras和tensorflow 报错解决：UserWarning: Method on_batch_end() is slow compared to the batch update Check
个人博客
06-07 8229
早上在使用keras时报错如下： C:\Users\peter\Anaconda3\lib\site-packages\keras\callbacks.py:122: UserWarning: Method on_batch_end() is slow compared to the batch update (0.115551). Check your callbacks. 这是一个警告，可以...
TensorFlow性能分析
weixin_26737625的博客
08-25 3341
In previous posts (here and here), I told you about how our team uses the Amazon SageMaker and Amazon s3 services to train our deep neural networks on large quantities of data. 在之前的文章( 此处和此处 )中，我告诉您我们...
解决WARNING:tensorflow:Callbacks method `on_train_batch_end` is slow compared to the batch time的可能方法
weixin_44107946的博客
10-13 1万+
修改代码keras.callbacks.ModelCheckpoint(filepath, monitor=‘val_loss’, verbose=0, save_best_only=False, save_weights_only=False, mode=‘auto’, period=1) 添加verbose=0（如果有则修改）， 注：不要看这篇： Callbacks method on_test_batch_end is slow ...
TensorFlow - 自定义 callback
无聊的豆奶
09-20 1160
callback 是一个强大的工具，可以在训练、评估和推断期间自定义 Keras 模型的行为。例如，可以使用 TensorBoard 可视化训练进度和结果，可以在训练时定期保存模型。下面介绍介绍什么是 callback，它可以做什么，以及如何创建自己的 callback。
e-packages\keras\src\layers\convolutional\base_conv.py:107: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead. super().__init__(activity_regularizer=activity_regularizer, **kwargs) 2025-04-17 10:15:21.483852: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations. To enable the following instructions: SSE3 SSE4.1 SSE4.2 AVX AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags. 2025-04-17 10:15:21.646762: W tensorflow/core/framework/op_kernel.cc:1857] OP_REQUIRES failed at summary_kernels.cc:65 : FAILED_PRECONDITION: logs/cnn_1D is not a directory 2025-04-17 10:15:21.646787: I tensorflow/core/framework/local_rendezvous.cc:407] Local rendezvous is aborting with status: FAILED_PRECONDITION: logs/cnn_1D
04-18
tensorboard_callback = keras.callbacks.TensorBoard(log_dir=r"logs/cnn_1D") # 使用原始字符串避免转义问题 ``` #### 三、完整代码示例 ```python from tensorflow import keras import numpy as np # 生成...
Code 分享 Notebook 保存成功 Python 3 (ipykernel) import os import numpy as np import pandas as pd import matplotlib.pyplot as plt from sklearn.preprocessing import MinMaxScaler from tensorflow.keras import layers,losses,optimizers, Sequential from tensorflow.keras.layers import InputLayer, Dense, LSTM, Dropout from tensorflow.keras.models import load_model from tensorflow.keras import Model 0秒 + Code + Markdown stock_data = pd.read_csv('IBM_stock_data.csv') 0秒 + Code + Markdown stock_data.head() 0秒 date Open High Low Close Volume Price Change % 0 1999/11/1 98.50 98.81 96.37 96.75 9551800 0.000000 1 1999/11/2 96.75 96.81 93.69 94.81 11105400 -2.005168 2 1999/11/3 95.87 95.94 93.50 94.37 10369100 -0.464086 3 1999/11/4 94.44 94.44 90.00 91.56 16697600 -2.977641 4 1999/11/5 92.75 92.94 90.19 90.25 13737600 -1.430756 + Code + Markdown scaler = MinMaxScaler(feature_range=(0, 1)) scaled_data = scaler.fit_transform(stock_data['Close'].values.reshape(-1, 1)) 0秒 + Code + Markdown scaled_data[0] 0秒 array([0.23131139]) + Code + Markdown len(scaled_data) 0秒 6293 + Code + Markdown training_data_len = int(np.ceil(len(scaled_data) * 0.8)) #向上取整 0秒 + Code + Markdown train_data = scaled_data[0:training_data_len] X_train, y_train = [], [] time_step = 10 # 时间窗口，模型基于前10个时间步长进行预测。可以尝试不同长度（如20、30）并观察效果变化。 0秒 + Code + Markdown for i in range(len(train_data) - time_step - 1): X_train.append(train_data[i:(i + time_step), 0]) y_train.append(train_data[i + time_step, 0]) 0秒 + Code + Markdown X_train, y_train = np.array(X_train), np.array(y_train) X_train = X_train.reshape(X_train.shape[0], X_train.shape[1], 1) 0秒 + Code + Markdown X_train.shape 0秒 (5024, 10, 1) + Code + Markdown X_test, y_test = [], [] test_data = scaled_data[training_data_len - time_step:] ​ for i in range(len(test_data) - time_step): X_test.append(test_data[i:(i + time_step), 0]) y_test.append(test_data[i + time_step, 0]) ​ X_test, y_test = np.array(X_test), np.array(y_test) X_test = X_test.reshape(X_test.shape[0], X_test.shape[1], 1) 0秒 + Code + Markdown model = Sequential() model.add(InputLayer(input_shape=(X_train.shape[1], 1))) model.add(LSTM(units=64, return_sequences=True)) # 调整Dropout： 当前设置为0.3，可以尝试在不同的层上使用不同的Dropout值，例如0.2~0.5之间。Dropout的作用是防止过拟合。 model.add(Dropout(0.3)) model.add(LSTM(units=64, return_sequences=True)) model.add(Dropout(0.3)) model.add(LSTM(units=32)) model.add(Dropout(0.2)) # 增加回归层： 如果希望更高的拟合精度，可以添加多个Dense层，例如在输出前再增加一层Dense。 model.add(Dense(units=1)) 3秒 2025-06-21 16:21:14.205566: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcuda.so.1'; dlerror: libcuda.so.1: cannot open shared object file: No such file or directory 2025-06-21 16:21:14.205665: W tensorflow/stream_executor/cuda/cuda_driver.cc:269] failed call to cuInit: UNKNOWN ERROR (303) 2025-06-21 16:21:14.205686: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:156] kernel driver does not appear to be running on this host (dsw-1161427-779497c87d-j6j54): /proc/driver/nvidia/version does not exist 2025-06-21 16:21:14.206190: I tensorflow/core/platform/cpu_feature_guard.cc:151] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 AVX512F FMA To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags. + Code + Markdown def MyRNN(): model = Sequential([ layers.InputLayer(input_shape=(X_train.shape[1],1)), layers.SimpleRNN(units=64, dropout=0.5, return_sequences=True, unroll=True), layers.SimpleRNN(units=64, dropout=0.5, unroll=True), layers.Dense(1)] ) return model 0秒 + Code + Markdown model = MyRNN() model.compile(optimizer='adam', loss='mean_squared_error') 0秒 + Code + Markdown history = model.fit( X_train, y_train, epochs=2, # 批大小（batch_size）： 尝试不同的batch_size，例如16、32、64，以找到训练稳定性和准确性之间的平衡 batch_size=32, #callbacks=[early_stopping, lr_scheduler] ) 12秒 Epoch 1/2 WARNING:tensorflow:AutoGraph could not transform <function Model.make_train_function.<locals>.train_function at 0x7f02c8692040> and will run it as-is. Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: module 'gast' has no attribute 'Constant' To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert WARNING: AutoGraph could not transform <function Model.make_train_function.<locals>.train_function at 0x7f02c8692040> and will run it as-is. Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: module 'gast' has no attribute 'Constant' To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert 157/157 [==============================] - 9s 20ms/step - loss: 0.0396 Epoch 2/2 157/157 [==============================] - 3s 21ms/step - loss: 0.0266 + Code + Markdown model.save('my_model.keras') 0秒 + Code + Markdown train_loss = model.evaluate(X_train, y_train, verbose=0) test_loss = model.evaluate(X_test, y_test, verbose=0) 5秒 WARNING:tensorflow:AutoGraph could not transform <function Model.make_test_function.<locals>.test_function at 0x7f02cc2b5dc0> and will run it as-is. Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: module 'gast' has no attribute 'Constant' To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert WARNING: AutoGraph could not transform <function Model.make_test_function.<locals>.test_function at 0x7f02cc2b5dc0> and will run it as-is. Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: module 'gast' has no attribute 'Constant' To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert + Code + Markdown # 训练集上计算的损失值，数值越小表示模型在训练数据上的拟合效果越好 print(f"Training Loss: {train_loss:.4f}") # 测试集上计算的损失值，反映了模型在未见过的数据上的表现。测试损失略高于训练损失，但差距不大，说明模型在新数据上的表现依然良好。 print(f"Testing Loss: {test_loss:.4f}") 0秒 Training Loss: 0.0412 Testing Loss: 0.0480 + Code + Markdown model = load_model('my_model.keras') predictions = model.predict(X_test) predictions = scaler.inverse_transform(predictions) # 反归一化预测值 2秒 WARNING:tensorflow:AutoGraph could not transform <function Model.make_predict_function.<locals>.predict_function at 0x7f0276f14040> and will run it as-is. Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: module 'gast' has no attribute 'Constant' To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert WARNING: AutoGraph could not transform <function Model.make_predict_function.<locals>.predict_function at 0x7f0276f14040> and will run it as-is. Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: module 'gast' has no attribute 'Constant' To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert + Code + Markdown import matplotlib.pyplot as plt plt.rcParams['font.sans-serif'] = ['SimHei'] # 设置中文字体为黑体 plt.rcParams['axes.unicode_minus'] = False # 正确显示负号 0秒 + Code + Markdown %matplotlib inline 0秒 + Code + Markdown Code train = stock_data[:training_data_len] valid = stock_data[training_data_len:] valid.loc[:, 'Predictions'] = predictions ​ ​ # 绘制图像 plt.figure(figsize=(14, 5)) plt.title('股票价格预测', fontsize=20) # 图表标题改为中文 plt.xlabel('日期', fontsize=14) # X 轴标签改为中文 plt.ylabel('收盘价', fontsize=14) # Y 轴标签改为中文 plt.plot(train['date'], train['Close'], label='训练数据', color='blue') # 训练数据标签改为中文 plt.plot(valid['date'], valid['Close'], label='真实价格', color='green') # 真实价格标签改为中文 plt.plot(valid['date'], valid['Predictions'], label='预测价格', color='red') # 预测价格标签改为中文 plt.legend() # 添加图例 # 添加保存图像的代码 plt.savefig('stock_price_predictions.png') # 保存图像 plt.show() ​ # 计算和输出评估指标 rmse = np.sqrt(np.mean(np.square(predictions - y_test))) mae = np.mean(np.abs(predictions - y_test)) print(f'均方根误差 (RMSE): {rmse}, 平均绝对误差 (MAE): {mae}') # 输出信息改为中文 4分钟58秒 /opt/conda/lib/python3.8/site-packages/pandas/core/indexing.py:1667: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy self.obj[key] = value findfont: Font family ['sans-serif'] not found. Falling back to DejaVu Sans. findfont: Generic family 'sans-serif' not found because none of the following families were found: SimHei /tmp/ipykernel_18835/4137420046.py:16: UserWarning: Glyph 32929 (\N{CJK UNIFIED IDEOGRAPH-80A1}) missing from current font. plt.savefig('stock_price_predictions.png') # 保存图像 /tmp/ipykernel_18835/4137420046.py:16: UserWarning: Glyph 31080 (\N{CJK UNIFIED IDEOGRAPH-7968}) missing from current font. plt.savefig('stock_price_predictions.png') # 保存图像 /tmp/ipykernel_18835/4137420046.py:16: UserWarning: Glyph 20215 (\N{CJK UNIFIED IDEOGRAPH-4EF7}) missing from current font. plt.savefig('stock_price_predictions.png') # 保存图像 /tmp/ipykernel_18835/4137420046.py:16: UserWarning: Glyph 26684 (\N{CJK UNIFIED IDEOGRAPH-683C}) missing from current font. plt.savefig('stock_price_predictions.png') # 保存图像 /tmp/ipykernel_18835/4137420046.py:16: UserWarning: Glyph 39044 (\N{CJK UNIFIED IDEOGRAPH-9884}) missing from current font. plt.savefig('stock_price_predictions.png') # 保存图像 /tmp/ipykernel_18835/4137420046.py:16: UserWarning: Glyph 27979 (\N{CJK UNIFIED IDEOGRAPH-6D4B}) missing from current font. plt.savefig('stock_price_predictions.png') # 保存图像 findfont: Font family ['sans-serif'] not found. Falling back to DejaVu Sans. findfont: Generic family 'sans-serif' not found because none of the following families were found: SimHei findfont: Font family ['sans-serif'] not found. Falling back to DejaVu Sans. findfont: Generic family 'sans-serif' not found because none of the following families were found: SimHei /tmp/ipykernel_18835/4137420046.py:16: UserWarning: Glyph 26085 (\N{CJK UNIFIED IDEOGRAPH-65E5}) missing from current font. plt.savefig('stock_price_predictions.png') # 保存图像 /tmp/ipykernel_18835/4137420046.py:16: UserWarning: Glyph 26399 (\N{CJK UNIFIED IDEOGRAPH-671F}) missing from current font. plt.savefig('stock_price_predictions.png') # 保存图像 /tmp/ipykernel_18835/4137420046.py:16: UserWarning: Glyph 25910 (\N{CJK UNIFIED IDEOGRAPH-6536}) missing from current font. plt.savefig('stock_price_predictions.png') # 保存图像 /tmp/ipykernel_18835/4137420046.py:16: UserWarning: Glyph 30424 (\N{CJK UNIFIED IDEOGRAPH-76D8}) missing from current font. plt.savefig('stock_price_predictions.png') # 保存图像 /tmp/ipykernel_18835/4137420046.py:16: UserWarning: Glyph 35757 (\N{CJK UNIFIED IDEOGRAPH-8BAD}) missing from current font. plt.savefig('stock_price_predictions.png') # 保存图像 /tmp/ipykernel_18835/4137420046.py:16: UserWarning: Glyph 32451 (\N{CJK UNIFIED IDEOGRAPH-7EC3}) missing from current font. plt.savefig('stock_price_predictions.png') # 保存图像 /tmp/ipykernel_18835/4137420046.py:16: UserWarning: Glyph 25968 (\N{CJK UNIFIED IDEOGRAPH-6570}) missing from current font. plt.savefig('stock_price_predictions.png') # 保存图像 /tmp/ipykernel_18835/4137420046.py:16: UserWarning: Glyph 25454 (\N{CJK UNIFIED IDEOGRAPH-636E}) missing from current font. plt.savefig('stock_price_predictions.png') # 保存图像 /tmp/ipykernel_18835/4137420046.py:16: UserWarning: Glyph 30495 (\N{CJK UNIFIED IDEOGRAPH-771F}) missing from current font. plt.savefig('stock_price_predictions.png') # 保存图像 /tmp/ipykernel_18835/4137420046.py:16: UserWarning: Glyph 23454 (\N{CJK UNIFIED IDEOGRAPH-5B9E}) missing from current font. plt.savefig('stock_price_predictions.png') # 保存图像 /opt/conda/lib/python3.8/site-packages/IPython/core/pylabtools.py:152: UserWarning: Glyph 26085 (\N{CJK UNIFIED IDEOGRAPH-65E5}) missing from current font. fig.canvas.print_figure(bytes_io, **kw) /opt/conda/lib/python3.8/site-packages/IPython/core/pylabtools.py:152: UserWarning: Glyph 26399 (\N{CJK UNIFIED IDEOGRAPH-671F}) missing from current font. fig.canvas.print_figure(bytes_io, **kw) /opt/conda/lib/python3.8/site-packages/IPython/core/pylabtools.py:152: UserWarning: Glyph 25910 (\N{CJK UNIFIED IDEOGRAPH-6536}) missing from current font. fig.canvas.print_figure(bytes_io, **kw) /opt/conda/lib/python3.8/site-packages/IPython/core/pylabtools.py:152: UserWarning: Glyph 30424 (\N{CJK UNIFIED IDEOGRAPH-76D8}) missing from current font. fig.canvas.print_figure(bytes_io, **kw) /opt/conda/lib/python3.8/site-packages/IPython/core/pylabtools.py:152: UserWarning: Glyph 20215 (\N{CJK UNIFIED IDEOGRAPH-4EF7}) missing from current font. fig.canvas.print_figure(bytes_io, **kw) /opt/conda/lib/python3.8/site-packages/IPython/core/pylabtools.py:152: UserWarning: Glyph 32929 (\N{CJK UNIFIED IDEOGRAPH-80A1}) missing from current font. fig.canvas.print_figure(bytes_io, **kw) /opt/conda/lib/python3.8/site-packages/IPython/core/pylabtools.py:152: UserWarning: Glyph 31080 (\N{CJK UNIFIED IDEOGRAPH-7968}) missing from current font. fig.canvas.print_figure(bytes_io, **kw) /opt/conda/lib/python3.8/site-packages/IPython/core/pylabtools.py:152: UserWarning: Glyph 26684 (\N{CJK UNIFIED IDEOGRAPH-683C}) missing from current font. fig.canvas.print_figure(bytes_io, **kw) /opt/conda/lib/python3.8/site-packages/IPython/core/pylabtools.py:152: UserWarning: Glyph 39044 (\N{CJK UNIFIED IDEOGRAPH-9884}) missing from current font. fig.canvas.print_figure(bytes_io, **kw) /opt/conda/lib/python3.8/site-packages/IPython/core/pylabtools.py:152: UserWarning: Glyph 27979 (\N{CJK UNIFIED IDEOGRAPH-6D4B}) missing from current font. fig.canvas.print_figure(bytes_io, **kw) /opt/conda/lib/python3.8/site-packages/IPython/core/pylabtools.py:152: UserWarning: Glyph 35757 (\N{CJK UNIFIED IDEOGRAPH-8BAD}) missing from current font. fig.canvas.print_figure(bytes_io, **kw) /opt/conda/lib/python3.8/site-packages/IPython/core/pylabtools.py:152: UserWarning: Glyph 32451 (\N{CJK UNIFIED IDEOGRAPH-7EC3}) missing from current font. fig.canvas.print_figure(bytes_io, **kw) /opt/conda/lib/python3.8/site-packages/IPython/core/pylabtools.py:152: UserWarning: Glyph 25968 (\N{CJK UNIFIED IDEOGRAPH-6570}) missing from current font. fig.canvas.print_figure(bytes_io, **kw) /opt/conda/lib/python3.8/site-packages/IPython/core/pylabtools.py:152: UserWarning: Glyph 25454 (\N{CJK UNIFIED IDEOGRAPH-636E}) missing from current font. fig.canvas.print_figure(bytes_io, **kw) /opt/conda/lib/python3.8/site-packages/IPython/core/pylabtools.py:152: UserWarning: Glyph 30495 (\N{CJK UNIFIED IDEOGRAPH-771F}) missing from current font. fig.canvas.print_figure(bytes_io, **kw) /opt/conda/lib/python3.8/site-packages/IPython/core/pylabtools.py:152: UserWarning: Glyph 23454 (\N{CJK UNIFIED IDEOGRAPH-5B9E}) missing from current font. fig.canvas.print_figure(bytes_io, **kw) 均方根误差 (RMSE): 105.03158114646604, 平均绝对误差 (MAE): 104.21013102460681误差好大，还有中文显示不成功
06-22
callbacks=callbacks, verbose=1 ) return history # 4. 使用完整代码流程 # 加载数据 data = pd.read_csv('IBM_stock_data.csv') # 准备数据 time_step = 30 X_train, X_test, y_train, y_test, scaler = ...
Training Begin e:/B19060708share/ENDING/spectrum_anomaly_detection-master/LSTM.py:177: UserWarning: `model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators. steps_per_epoch=t_len, callbacks=callbacks, validation_data=valid_generator, validation_steps=v_len) 2023-05-16 00:50:29.234124: W tensorflow/c/c_api.cc:291] Operation '{name:'batch_normalization_1/moving_variance/Assign' id:578 op device:{requested: '', assigned: ''} def:{{{node batch_normalization_1/moving_variance/Assign}} = AssignVariableOp[_has_manual_control_dependencies=true, dtype=DT_FLOAT, validate_shape=false](batch_normalization_1/moving_variance, batch_normalization_1/moving_variance/Initializer/ones)}}' was changed by setting attribute after it was run by a session. This mutation will have no effect, and will trigger an error in the future. Either don't modify nodes after running them or create a new session. 我在训练LSTM模型，但是运行文件后终端一直卡在前面这句话，这是运行失败了吗？怎么知道我的模型有没有在训练？
05-16
这个警告不是训练失败的原因，只是TensorFlow给出的一个提示。如果你想看到训练过程中的输出，你可以在代码中添加打印语句或者使用TensorBoard来可视化训练过程。在训练过程中，你应该能够看到每个epoch的损失值和...
2025-06-22 18:56:53.192710: I tensorflow/core/util/port.cc:153] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`. 2025-06-22 18:56:54.136866: I tensorflow/core/util/port.cc:153] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`. 2025-06-22 18:56:56,467 - INFO - 加载并增强数据集: augmented_data 2025-06-22 18:56:56,560 - INFO - 原始数据集: 150 张图片, 5 个类别 2025-06-22 18:56:56,565 - INFO - 类别 book: 30 张原始图像 2025-06-22 18:56:56,989 - INFO - 类别 cup: 30 张原始图像 2025-06-22 18:56:57,403 - INFO - 类别 glasses: 30 张原始图像 2025-06-22 18:56:57,820 - INFO - 类别 phone: 30 张原始图像 2025-06-22 18:56:58,248 - INFO - 类别 shoe: 30 张原始图像 2025-06-22 18:56:58,859 - INFO - 增强后数据集: 450 张图片 2025-06-22 18:56:58,954 - INFO - 构建优化的迁移学习模型... 2025-06-22 18:56:58.959007: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations. To enable the following instructions: SSE3 SSE4.1 SSE4.2 AVX AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags. Model: "functional" ┌─────────────────────┬───────────────────┬────────────┬───────────────────┐ │ Layer (type) │ Output Shape │ Param # │ Connected to │ ├─────────────────────┼───────────────────┼────────────┼───────────────────┤ │ input_layer_1 │ (None, 224, 224, │ 0 │ - │ │ (InputLayer) │ 3) │ │ │ ├─────────────────────┼───────────────────┼────────────┼───────────────────┤ │ efficientnetb0 │ (None, 7, 7, │ 4,049,571 │ input_layer_1[0]… │ │ (Functional) │ 1280) │ │ │ ├─────────────────────┼───────────────────┼────────────┼───────────────────┤ │ global_average_poo… │ (None, 1280) │ 0 │ efficientnetb0[0… │ │ (GlobalAveragePool… │ │ │ │ ├─────────────────────┼───────────────────┼────────────┼───────────────────┤ │ dense (Dense) │ (None, 512) │ 655,872 │ global_average_p… │ ├─────────────────────┼───────────────────┼────────────┼───────────────────┤ │ dense_1 (Dense) │ (None, 1280) │ 656,640 │ dense[0][0] │ ├─────────────────────┼───────────────────┼────────────┼───────────────────┤ │ multiply (Multiply) │ (None, 1280) │ 0 │ global_average_p… │ │ │ │ │ dense_1[0][0] │ ├─────────────────────┼───────────────────┼────────────┼───────────────────┤ │ dense_2 (Dense) │ (None, 512) │ 655,872 │ multiply[0][0] │ ├─────────────────────┼───────────────────┼────────────┼───────────────────┤ │ batch_normalization │ (None, 512) │ 2,048 │ dense_2[0][0] │ │ (BatchNormalizatio… │ │ │ │ ├─────────────────────┼───────────────────┼────────────┼───────────────────┤ │ dropout (Dropout) │ (None, 512) │ 0 │ batch_normalizat… │ ├─────────────────────┼───────────────────┼────────────┼───────────────────┤ │ dense_3 (Dense) │ (None, 5) │ 2,565 │ dropout[0][0] │ └─────────────────────┴───────────────────┴────────────┴───────────────────┘ Total params: 6,022,568 (22.97 MB) Trainable params: 1,971,973 (7.52 MB) Non-trainable params: 4,050,595 (15.45 MB) 2025-06-22 18:56:59,882 - INFO - 开始高级训练策略... 2025-06-22 18:56:59,882 - INFO - 阶段1: 冻结基础模型训练 Epoch 1/50 23/23 ━━━━━━━━━━━━━━━━━━━━ 16s 442ms/step - accuracy: 0.2001 - loss: 1.7270 - val_accuracy: 0.2000 - val_loss: 1.6104 - learning_rate: 1.0000e-04 Epoch 2/50 23/23 ━━━━━━━━━━━━━━━━━━━━ 8s 335ms/step - accuracy: 0.1486 - loss: 1.7114 - val_accuracy: 0.2000 - val_loss: 1.6100 - learning_rate: 1.0000e-04 Epoch 3/50 23/23 ━━━━━━━━━━━━━━━━━━━━ 8s 332ms/step - accuracy: 0.2337 - loss: 1.7239 - val_accuracy: 0.2000 - val_loss: 1.6117 - learning_rate: 1.0000e-04 Epoch 4/50 23/23 ━━━━━━━━━━━━━━━━━━━━ 8s 335ms/step - accuracy: 0.2558 - loss: 1.6466 - val_accuracy: 0.2000 - val_loss: 1.6104 - learning_rate: 1.0000e-04 Epoch 5/50 23/23 ━━━━━━━━━━━━━━━━━━━━ 0s 270ms/step - accuracy: 0.2281 - loss: 1.6503 Epoch 5: ReduceLROnPlateau reducing learning rate to 4.999999873689376e-05. 23/23 ━━━━━━━━━━━━━━━━━━━━ 8s 367ms/step - accuracy: 0.2271 - loss: 1.6513 - val_accuracy: 0.2111 - val_loss: 1.6118 - learning_rate: 1.0000e-04 Epoch 6/50 23/23 ━━━━━━━━━━━━━━━━━━━━ 8s 333ms/step - accuracy: 0.1899 - loss: 1.6756 - val_accuracy: 0.2000 - val_loss: 1.6112 - learning_rate: 5.0000e-05 Epoch 7/50 23/23 ━━━━━━━━━━━━━━━━━━━━ 8s 333ms/step - accuracy: 0.2394 - loss: 1.6269 - val_accuracy: 0.2000 - val_loss: 1.6128 - learning_rate: 5.0000e-05 Epoch 8/50 23/23 ━━━━━━━━━━━━━━━━━━━━ 0s 266ms/step - accuracy: 0.2041 - loss: 1.7332 Epoch 8: ReduceLROnPlateau reducing learning rate to 2.499999936844688e-05. 23/23 ━━━━━━━━━━━━━━━━━━━━ 8s 333ms/step - accuracy: 0.2042 - loss: 1.7319 - val_accuracy: 0.2000 - val_loss: 1.6103 - learning_rate: 5.0000e-05 Epoch 9/50 23/23 ━━━━━━━━━━━━━━━━━━━━ 8s 355ms/step - accuracy: 0.1765 - loss: 1.6814 - val_accuracy: 0.3333 - val_loss: 1.6107 - learning_rate: 2.5000e-05 2025-06-22 18:58:18,867 - INFO - 阶段2: 微调部分层 Epoch 1/25 23/23 ━━━━━━━━━━━━━━━━━━━━ 18s 460ms/step - accuracy: 0.2374 - loss: 2.4082 - val_accuracy: 0.2000 - val_loss: 1.6107 - learning_rate: 1.0000e-05 Epoch 2/25 23/23 ━━━━━━━━━━━━━━━━━━━━ 9s 390ms/step - accuracy: 0.2021 - loss: 2.2585 - val_accuracy: 0.2000 - val_loss: 1.6112 - learning_rate: 1.0000e-05 Epoch 3/25 23/23 ━━━━━━━━━━━━━━━━━━━━ 9s 375ms/step - accuracy: 0.2259 - loss: 2.3548 - val_accuracy: 0.2111 - val_loss: 1.6121 - learning_rate: 1.0000e-05 Epoch 4/25 23/23 ━━━━━━━━━━━━━━━━━━━━ 0s 307ms/step - accuracy: 0.2416 - loss: 2.0942 Epoch 4: ReduceLROnPlateau reducing learning rate to 4.999999873689376e-06. 23/23 ━━━━━━━━━━━━━━━━━━━━ 9s 374ms/step - accuracy: 0.2405 - loss: 2.1006 - val_accuracy: 0.2000 - val_loss: 1.6127 - learning_rate: 1.0000e-05 Epoch 5/25 23/23 ━━━━━━━━━━━━━━━━━━━━ 9s 377ms/step - accuracy: 0.2053 - loss: 2.1248 - val_accuracy: 0.2000 - val_loss: 1.6136 - learning_rate: 5.0000e-06 Epoch 6/25 23/23 ━━━━━━━━━━━━━━━━━━━━ 9s 382ms/step - accuracy: 0.1995 - loss: 2.2549 - val_accuracy: 0.2000 - val_loss: 1.6150 - learning_rate: 5.0000e-06 Epoch 7/25 23/23 ━━━━━━━━━━━━━━━━━━━━ 0s 305ms/step - accuracy: 0.1949 - loss: 2.1615 Epoch 7: ReduceLROnPlateau reducing learning rate to 2.499999936844688e-06. 23/23 ━━━━━━━━━━━━━━━━━━━━ 9s 373ms/step - accuracy: 0.1962 - loss: 2.1615 - val_accuracy: 0.2000 - val_loss: 1.6165 - learning_rate: 5.0000e-06 Epoch 8/25 23/23 ━━━━━━━━━━━━━━━━━━━━ 9s 375ms/step - accuracy: 0.2320 - loss: 2.1199 - val_accuracy: 0.2000 - val_loss: 1.6186 - learning_rate: 2.5000e-06 Epoch 9/25 23/23 ━━━━━━━━━━━━━━━━━━━━ 9s 378ms/step - accuracy: 0.2379 - loss: 2.1694 - val_accuracy: 0.2000 - val_loss: 1.6204 - learning_rate: 2.5000e-06 2025-06-22 18:59:46,920 - INFO - 训练完成 2025-06-22 18:59:46,920 - INFO - 评估模型... 2025-06-22 18:59:48,600 - INFO - 测试准确率: 20.00% E:\pycharm\study\计算机视觉\物品识别系统.py:313: UserWarning: Glyph 20934 (\N{CJK UNIFIED IDEOGRAPH-51C6}) missing from font(s) DejaVu Sans. plt.tight_layout() E:\pycharm\study\计算机视觉\物品识别系统.py:313: UserWarning: Glyph 30830 (\N{CJK UNIFIED IDEOGRAPH-786E}) missing from font(s) DejaVu Sans. plt.tight_layout() E:\pycharm\study\计算机视觉\物品识别系统.py:313: UserWarning: Glyph 29575 (\N{CJK UNIFIED IDEOGRAPH-7387}) missing from font(s) DejaVu Sans. plt.tight_layout() E:\pycharm\study\计算机视觉\物品识别系统.py:313: UserWarning: Glyph 35757 (\N{CJK UNIFIED IDEOGRAPH-8BAD}) missing from font(s) DejaVu Sans. plt.tight_layout() E:\pycharm\study\计算机视觉\物品识别系统.py:313: UserWarning: Glyph 32451 (\N{CJK UNIFIED IDEOGRAPH-7EC3}) missing from font(s) DejaVu Sans. plt.tight_layout() E:\pycharm\study\计算机视觉\物品识别系统.py:313: UserWarning: Glyph 21644 (\N{CJK UNIFIED IDEOGRAPH-548C}) missing from font(s) DejaVu Sans. plt.tight_layout() E:\pycharm\study\计算机视觉\物品识别系统.py:313: UserWarning: Glyph 39564 (\N{CJK UNIFIED IDEOGRAPH-9A8C}) missing from font(s) DejaVu Sans. plt.tight_layout() E:\pycharm\study\计算机视觉\物品识别系统.py:313: UserWarning: Glyph 35777 (\N{CJK UNIFIED IDEOGRAPH-8BC1}) missing from font(s) DejaVu Sans. plt.tight_layout() E:\pycharm\study\计算机视觉\物品识别系统.py:313: UserWarning: Glyph 25439 (\N{CJK UNIFIED IDEOGRAPH-635F}) missing from font(s) DejaVu Sans. plt.tight_layout() E:\pycharm\study\计算机视觉\物品识别系统.py:313: UserWarning: Glyph 22833 (\N{CJK UNIFIED IDEOGRAPH-5931}) missing from font(s) DejaVu Sans. plt.tight_layout() E:\pycharm\study\计算机视觉\物品识别系统.py:314: UserWarning: Glyph 20934 (\N{CJK UNIFIED IDEOGRAPH-51C6}) missing from font(s) DejaVu Sans. plt.savefig('training_history.png') E:\pycharm\study\计算机视觉\物品识别系统.py:314: UserWarning: Glyph 30830 (\N{CJK UNIFIED IDEOGRAPH-786E}) missing from font(s) DejaVu Sans. plt.savefig('training_history.png') E:\pycharm\study\计算机视觉\物品识别系统.py:314: UserWarning: Glyph 29575 (\N{CJK UNIFIED IDEOGRAPH-7387}) missing from font(s) DejaVu Sans. plt.savefig('training_history.png') E:\pycharm\study\计算机视觉\物品识别系统.py:314: UserWarning: Glyph 35757 (\N{CJK UNIFIED IDEOGRAPH-8BAD}) missing from font(s) DejaVu Sans. plt.savefig('training_history.png') E:\pycharm\study\计算机视觉\物品识别系统.py:314: UserWarning: Glyph 32451 (\N{CJK UNIFIED IDEOGRAPH-7EC3}) missing from font(s) DejaVu Sans. plt.savefig('training_history.png') E:\pycharm\study\计算机视觉\物品识别系统.py:314: UserWarning: Glyph 21644 (\N{CJK UNIFIED IDEOGRAPH-548C}) missing from font(s) DejaVu Sans. plt.savefig('training_history.png') E:\pycharm\study\计算机视觉\物品识别系统.py:314: UserWarning: Glyph 39564 (\N{CJK UNIFIED IDEOGRAPH-9A8C}) missing from font(s) DejaVu Sans. plt.savefig('training_history.png') E:\pycharm\study\计算机视觉\物品识别系统.py:314: UserWarning: Glyph 35777 (\N{CJK UNIFIED IDEOGRAPH-8BC1}) missing from font(s) DejaVu Sans. plt.savefig('training_history.png') E:\pycharm\study\计算机视觉\物品识别系统.py:314: UserWarning: Glyph 25439 (\N{CJK UNIFIED IDEOGRAPH-635F}) missing from font(s) DejaVu Sans. plt.savefig('training_history.png') E:\pycharm\study\计算机视觉\物品识别系统.py:314: UserWarning: Glyph 22833 (\N{CJK UNIFIED IDEOGRAPH-5931}) missing from font(s) DejaVu Sans. plt.savefig('training_history.png') 2025-06-22 18:59:48,905 - INFO - 训练历史图表已保存到 training_history.png 2025-06-22 18:59:49,390 - INFO - 模型已保存到: optimized_model.keras 2025-06-22 18:59:49,390 - INFO - 执行内存清理... WARNING:tensorflow:From E:\python3.9.13\lib\site-packages\keras\src\backend\common\global_state.py:82: The name tf.reset_default_graph is deprecated. Please use tf.compat.v1.reset_default_graph instead. 2025-06-22 18:59:50,195 - WARNING - From E:\python3.9.13\lib\site-packages\keras\src\backend\common\global_state.py:82: The name tf.reset_default_graph is deprecated. Please use tf.compat.v1.reset_default_graph instead. 2025-06-22 18:59:50,743 - INFO - 内存清理完成 E:\pycharm\study\计算机视觉\物品识别系统.py:355: UserWarning: set_ticklabels() should only be used with a fixed number of ticks, i.e. after set_ticks() or using a FixedLocator. ax2.set_xticklabels(self.class_labels, rotation=45) E:\pycharm\study\计算机视觉\物品识别系统.py:364: UserWarning: Glyph 39044 (\N{CJK UNIFIED IDEOGRAPH-9884}) missing from font(s) DejaVu Sans. plt.tight_layout() E:\pycharm\study\计算机视觉\物品识别系统.py:364: UserWarning: Glyph 27979 (\N{CJK UNIFIED IDEOGRAPH-6D4B}) missing from font(s) DejaVu Sans. plt.tight_layout() E:\pycharm\study\计算机视觉\物品识别系统.py:364: UserWarning: Glyph 27010 (\N{CJK UNIFIED IDEOGRAPH-6982}) missing from font(s) DejaVu Sans. plt.tight_layout() E:\pycharm\study\计算机视觉\物品识别系统.py:364: UserWarning: Glyph 29575 (\N{CJK UNIFIED IDEOGRAPH-7387}) missing from font(s) DejaVu Sans. plt.tight_layout() E:\pycharm\study\计算机视觉\物品识别系统.py:364: UserWarning: Glyph 31867 (\N{CJK UNIFIED IDEOGRAPH-7C7B}) missing from font(s) DejaVu Sans. plt.tight_layout() E:\pycharm\study\计算机视觉\物品识别系统.py:364: UserWarning: Glyph 21035 (\N{CJK UNIFIED IDEOGRAPH-522B}) missing from font(s) DejaVu Sans. plt.tight_layout() E:\pycharm\study\计算机视觉\物品识别系统.py:364: UserWarning: Glyph 20998 (\N{CJK UNIFIED IDEOGRAPH-5206}) missing from font(s) DejaVu Sans. plt.tight_layout() E:\pycharm\study\计算机视觉\物品识别系统.py:364: UserWarning: Glyph 24067 (\N{CJK UNIFIED IDEOGRAPH-5E03}) missing from font(s) DejaVu Sans. plt.tight_layout() E:\pycharm\study\计算机视觉\物品识别系统.py:365: UserWarning: Glyph 39044 (\N{CJK UNIFIED IDEOGRAPH-9884}) missing from font(s) DejaVu Sans. plt.savefig(output_path) E:\pycharm\study\计算机视觉\物品识别系统.py:365: UserWarning: Glyph 27979 (\N{CJK UNIFIED IDEOGRAPH-6D4B}) missing from font(s) DejaVu Sans. plt.savefig(output_path) E:\pycharm\study\计算机视觉\物品识别系统.py:365: UserWarning: Glyph 27010 (\N{CJK UNIFIED IDEOGRAPH-6982}) missing from font(s) DejaVu Sans. plt.savefig(output_path) E:\pycharm\study\计算机视觉\物品识别系统.py:365: UserWarning: Glyph 29575 (\N{CJK UNIFIED IDEOGRAPH-7387}) missing from font(s) DejaVu Sans. plt.savefig(output_path) E:\pycharm\study\计算机视觉\物品识别系统.py:365: UserWarning: Glyph 31867 (\N{CJK UNIFIED IDEOGRAPH-7C7B}) missing from font(s) DejaVu Sans. plt.savefig(output_path) E:\pycharm\study\计算机视觉\物品识别系统.py:365: UserWarning: Glyph 21035 (\N{CJK UNIFIED IDEOGRAPH-522B}) missing from font(s) DejaVu Sans. plt.savefig(output_path) E:\pycharm\study\计算机视觉\物品识别系统.py:365: UserWarning: Glyph 20998 (\N{CJK UNIFIED IDEOGRAPH-5206}) missing from font(s) DejaVu Sans. plt.savefig(output_path) E:\pycharm\study\计算机视觉\物品识别系统.py:365: UserWarning: Glyph 24067 (\N{CJK UNIFIED IDEOGRAPH-5E03}) missing from font(s) DejaVu Sans. plt.savefig(output_path) 2025-06-22 18:59:52,251 - INFO - 预测结果已保存到 prediction_result.png 2025-06-22 18:59:52,252 - INFO - 真实类别: cup
最新发布
06-23
from tensorflow.keras.callbacks import EarlyStopping, ModelCheckpoint, ReduceLROnPlateau from sklearn.model_selection import train_test_split import matplotlib matplotlib.use('Agg') import matplotlib....
UserWarning: The parameter ‘pretrained‘ is deprecated since 0.13 and will be removed in 0.15.
热门推荐
weixin_55593481的博客
11-01 1万+
UserWarning: The parameter 'pretrained' is deprecated since 0.13 and will be removed in 0.15.
深度学习模型部署之模型优化
Fantastic
03-24 4408
文章目录前言模型剪枝模型量化 前言 前段时间看了知乎有一个问题：训练好的深度学习模型是怎样部署的？，当中有一个高赞的回答，大致意思如下： 深度学习部署方式取决于你的需求： 需求一：简单的demo演示，看看效果。只需要把训练模型切换到inference模式或者用CPython包装供C++工程进行调用。 需求二：放到服务器上开启服务，不要求吞吐不要求时延。基于训练框架（tensorflow，pytorch，caffe）等用C++部署一套。这种尚未脱离框架，导致占用存储空间。 需求三：放到服务器上要求吞吐和时延
关于Keras的格式化输出Loss实现
鹊踏枝-码农的专栏
07-17 9297
在win7 64位，Anaconda安装的Python3.6.1下安装的TensorFlow与Keras，Keras的backend为TensorFlow。在运行Mask R-CNN时，在进行调试时想知道PyCharm (Python IDE)底部窗口输出的Loss格式是在哪里定义的，如下图红框中所示：                                             ...
Tensorflow2.0学习（4）：深度神经网络的构建以及批归一化、dropout的使用
一枚小白的博客
02-25 1793
深度神经网络 合理地增加神经网络的深度和广度，可以提升模型的表达能力，得到性能更加优秀的模型。 批归一化 是深度神经网络训练的一个技巧 # tf.keras.models.Sequential() 构建模型 # 构建深度神经网络 model = keras.models.Sequential() model.add(keras.layers.Flatten(input_shape=[2...
Tensorflow2.0学习（7）：wide&deep模型的多输入与多输出
一枚小白的博客
02-26 3862
修改前篇代码中的输入与输出 确定wide和deep端分别的输入 # 将输入切分为deep和wide输入 # 取前5个特征作为wide端的输入 # 取3->n列的特征作为deep端的输入 x_train_scaled_wide = x_train_scaled[:, :5] x_train_scaled_deep = x_train_scaled[:, 2:] x_valid_scaled...
成功解决 keras\callbacks.py:999: UserWarning: `epsilon` argument is deprecated and will be removed, use
头部AI社区如有邀博主AI主题演讲请私信—心比天高，仗剑走天涯，保持热爱，奔赴向梦想！低调，专注，谦虚，自律，反思，成长，还算比较正能量的博主，公益免费传播…内心特别想在AI界做出一些可以推进历史进程影响力的技术(兴趣使然，有点小情怀，也有点使命感呀
05-17 8583
成功解决 keras\callbacks.py:999: UserWarning: `epsilon` argument is deprecated and will be removed, use 目录 解决问题 解决思路 解决方法 解决问题 F:\Program Files\Python\Python36\lib\site-pa...
成功解决WARNING:tensorflow:From ：read_data_sets (from tensorflow.contrib.learn.python.learn.
头部AI社区如有邀博主AI主题演讲请私信—心比天高，仗剑走天涯，保持热爱，奔赴向梦想！低调，专注，谦虚，自律，反思，成长，还算比较正能量的博主，公益免费传播…内心特别想在AI界做出一些可以推进历史进程影响力的技术(兴趣使然，有点小情怀，也有点使命感呀
07-30 1万+
成功解决WARNING:tensorflow:From ：read_data_sets (from tensorflow.contrib.learn.python.learn. 目录 解决问题 解决思路 解决方法 解决问题 WARNING:tensorflow:From 190403.py:6: read_data_sets ...
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值