课件和 Demo 地址

最新的的Keras 版本 keras.optimizers.adam 改为 keras.optimizers.Adam了 train.py中的一处修改

训练好的h5模型能否提供？这样就可以不用等结果，继续后面的内容了。 abu820@qq.com

请问老师，训练模型阶段，这个问题怎么解决: File "/tensorflow_python/workspace/tensorflow-101-master/experts/code_sample/chapter-456/keras-retinanet/keras_retinanet/bin/../../keras_retinanet/models/resnet.py", line 37, in retinanet return resnet_retinanet(*args, backbone=self.backbone, **kwargs) File "/tensorflow_python/workspace/tensorflow-101-master/experts/code_sample/chapter-456/keras-retinanet/keras_retinanet/bin/../../keras_retinanet/models/resnet.py", line 98, in resnet_retinanet resnet = keras_resnet.models.ResNet50(inputs, include_top=False, freeze_bn=True) File "/home/ps/anaconda3/envs/tf2_gpu/lib/python3.8/site-packages/keras_resnet/models/_2d.py", line 233, in __init__ super(ResNet2D50, self).__init__( File "/home/ps/anaconda3/envs/tf2_gpu/lib/python3.8/site-packages/keras_resnet/models/_2d.py", line 76, in __init__ x = keras.layers.Conv2D(64, (7, 7), strides=(2, 2), use_bias=False, name="conv1", padding="same")(inputs) File "/home/ps/anaconda3/envs/tf2_gpu/lib/python3.8/site-packages/keras/backend/tensorflow_backend.py", line 75, in symbolic_fn_wrapper return func(*args, **kwargs) File "/home/ps/anaconda3/envs/tf2_gpu/lib/python3.8/site-packages/keras/engine/base_layer.py", line 446, in __call__ self.assert_input_compatibility(inputs) File "/home/ps/anaconda3/envs/tf2_gpu/lib/python3.8/site-packages/keras/engine/base_layer.py", line 312, in assert_input_compatibility raise ValueError('Layer ' + self.name + ' was called with ' ValueError: Layer conv1 was called with an input that isn't a symbolic tensor. Received type: <class 'tensorflow.python.keras.engine.keras_tensor.KerasTensor'>. Full input: [<KerasTensor: shape=(None, None, None, 3) dtype=float32 (created by layer 'input_1')>]. All inputs to the layer should be tensors.

Traceback (most recent call last): File "keras-retinanet/keras-retinanet/keras_retinanet/bin/train.py", line 34, in <module> from .. import layers # noqa: F401 File "keras-retinanet/keras-retinanet/keras_retinanet/bin\..\..\keras_retinanet\layers\__init__.py", line 1, in <module> from ._misc import RegressBoxes, UpsampleLike, Anchors, ClipBoxes # noqa: F401 File "keras-retinanet/keras-retinanet/keras_retinanet/bin\..\..\keras_retinanet\layers\_misc.py", line 20, in <module> from ..utils import anchors as utils_anchors File "keras-retinanet/keras-retinanet/keras_retinanet/bin\..\..\keras_retinanet\utils\anchors.py", line 20, in <module> from ..utils.compute_overlap import compute_overlap ImportError: DLL load failed: 找不到指定的模块。 老师能帮看看什么问题吗

ModuleNotFoundError: No module named 'keras_retinanet.utils.compute_overlap',执行过程中出现这个错误如何解决

老师，用NVIDIA Quadro P620训练显示内存不足，该怎么处理，GPU内存是2G的

请问老师，怎么讲keras_retinanet 训练出来的h5转化为tflite? model = tf.keras.models.load_model('models/resnet50_csv_02.h5') converter = tf.lite.TFLiteConverter.from_keras_model(model) tflite_model = converter.convert() open("converted_model.tflite", "wb").write(tflite_model) 这段代码会报TypeError: ('Keyword argument not understood:', 'freeze') 错误

在train_info里报了一个这样的错误,ValueError: invalid CSV class file: data/baijiu\class.csv: line 2: format should be 'class_name,class_id' 找了很久没找到怎么解决

最新版本里好像找不到 keras.initializers.normal，官方文档中给到的是initializers.RandomNormal

老师，关于RetinaNet的训练出现一个error，安装github上测试的例子，“ResNet-50-model.keras.h5”已下载并放在当前目录，进入python命令行，执行如下： from keras_retinanet.models import load_model model = load_model('ResNet-50-model.keras.h5',backbone_name='resnet50') 报： /root/.local/lib/python3.7/site-packages/keras_resnet/layers/_batch_normalization.py:17 call * return super(BatchNormalization, self).call(training=(not self.freeze), *args, **kwargs) TypeError: type object got multiple values for keyword argument 'training'