zheng-yuwei / multi-label-classification

基于tf.keras的多标签多分类模型

Hi there 👋

• 🔭 I’m currently working on computer vision (image-to-image translation, inpainting, etc.)
• 📫 Welcome to visit my blog: zhihu

multi-label-classification's Issues

直接运行报错

File "run.py", line 106, in run
classifier = MultiLabelClassifier()
File "/data_2/everyday/0323/face/multi-label-classification-master/multi_label/trainer.py", line 95, in init
optimizer = RAdam(lr=1e-3)
File "/data_2/everyday/0323/face/multi-label-classification-master/utils/radam.py", line 42, in init
super(RAdam, self).init(**kwargs)
TypeError: init() missing 1 required positional argument: 'name'

不知道是不是哪个库版本问题

我是torch选手，咨询一下GHMLoss的valid mask是怎么产生的

改了个torch版本的GHMloss，没太理解valid mask是如何定义的
因此自己写的torch版本吧mask去掉了，但是模型欠拟合了
btw，我是做的多标签文本分类，GHMLoss不知道是不是不适合文本问题

BACKBONE_MIXNET_18

可以训练，但是跑预测的时候报错。。。不知道哪里形状对不上了，跑这个也是 ResNeXt18
下面的log是BACKBONE_MIXNET_18

Deprecated in favor of operator or tf.math.divide.
2020-03-25 10:22:54,336 file_util.py:102 INFO 4. 完成tf.data (image, label, path) 测试数据集构造；
2020-03-25 10:22:54,353 file_util.py:111 INFO 5. 构造tf.data多epoch训练模式；
2020-03-25 10:22:54.692572: W tensorflow/core/framework/cpu_allocator_impl.cc:81] Allocation of 153600000 exceeds 10% of system memory.
2020-03-25 10:22:55.040592: W tensorflow/core/framework/cpu_allocator_impl.cc:81] Allocation of 153600000 exceeds 10% of system memory.
Traceback (most recent call last):
File "run.py", line 214, in
run()
File "run.py", line 122, in run
total_test, wrong_count, great_total_count, great_wrong_count, great_wrong_records = test_model(classifier)
File "run.py", line 176, in test_model
results = classifier.predict(np.array(images))
File "/data_2/everyday/0323/face/multi-label-2/multi-label-classification-master/multi_label/trainer.py", line 203, in predict
predictions = self.model.predict(test_images)
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/training.py", line 908, in predict
use_multiprocessing=use_multiprocessing)
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/training_v2.py", line 477, in predict
workers=workers, use_multiprocessing=use_multiprocessing, **kwargs)
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/training_v2.py", line 454, in _model_iteration
total_epochs=1)
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/training_v2.py", line 123, in run_one_epoch
batch_outs = execution_function(iterator)
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/training_v2_utils.py", line 86, in execution_function
distributed_function(input_fn))
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/eager/def_function.py", line 449, in call
self._initialize(args, kwds, add_initializers_to=initializer_map)
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/eager/def_function.py", line 392, in _initialize
*args, **kwds))
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/eager/function.py", line 1847, in _get_concrete_function_internal_garbage_collected
graph_function, _, _ = self._maybe_define_function(args, kwargs)
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/eager/function.py", line 2147, in _maybe_define_function
graph_function = self._create_graph_function(args, kwargs)
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/eager/function.py", line 2038, in _create_graph_function
capture_by_value=self._capture_by_value),
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/framework/func_graph.py", line 915, in func_graph_from_py_func
func_outputs = python_func(*func_args, **func_kwargs)
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/eager/def_function.py", line 335, in wrapped_fn
return weak_wrapped_fn().wrapped(*args, **kwds)
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/training_v2_utils.py", line 73, in distributed_function
per_replica_function, args=(x, y, sample_weights))
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/distribute/distribute_lib.py", line 764, in experimental_run_v2
return self._extended.call_for_each_replica(fn, args=args, kwargs=kwargs)
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/distribute/distribute_lib.py", line 1810, in call_for_each_replica
return self._call_for_each_replica(fn, args, kwargs)
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/distribute/distribute_lib.py", line 2155, in _call_for_each_replica
return fn(*args, **kwargs)
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/autograph/impl/api.py", line 292, in wrapper
return func(*args, **kwargs)
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/training_v2_utils.py", line 162, in _predict_on_batch
return predict_on_batch(model, x)
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/training_v2_utils.py", line 370, in predict_on_batch
return model(inputs) # pylint: disable=not-callable
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/base_layer.py", line 854, in call
outputs = call_fn(cast_inputs, *args, **kwargs)
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/network.py", line 695, in call
return self._run_internal_graph(inputs, training=training, mask=mask)
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/network.py", line 844, in _run_internal_graph
output_tensors = layer(computed_tensors, **kwargs)
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/base_layer.py", line 819, in call
self.name)
File "/data_1/Yang/software_install/Anaconda1105/envs/tf_1.15/lib/python3.6/site-packages/tensorflow_core/python/keras/engine/input_spec.py", line 213, in assert_input_compatibility
' but received input with shape ' + str(shape))
ValueError: Input 0 of layer depthwise_conv2d_1 is incompatible with the layer: expected axis 3 of input shape to have value 16 but received input with shape [None, 40, 40, 8]

求原论文

求原论文

valid_mask设置

作者你好 我在测试GHM 使用的时候 valid_mask 参数怎么设置啊