Deep learning on galaxy morphology profile - 版本历史

2022年10月1日 (六) 08:15 Cluster

2022-10-01T08:15:58Z

←上一版本		2022年10月1日 (六) 08:15的版本
第19行：		第19行：
	==Result==		==Result==
	===Generate data First Method===		===Generate data First Method===
	*The above network fits the galaxy data to one parameter using a ~~two-dimensional~~ convolutional neural network. The following are the training results for training a single parameter - star magnitude.		*The above network fits the galaxy data to one parameter using a 2D convolutional neural network. The following are the training results for training a single parameter - star magnitude.
	[[File:result_111.png\|500px\|right\|jumengting]]		[[File:result_111.png\|500px\|right\|jumengting]]
	===Generate data Second Method===		===Generate data Second Method===
			*Galsim result.

	===Ground-Truth data Second Method===		===Ground-Truth data Second Method===
			*CANDELS


	==Else==		==Else==

2022年10月1日 (六) 08:10 Cluster

2022-10-01T08:10:21Z

←上一版本		2022年10月1日 (六) 08:10的版本
第18行：		第18行：

	==Result==		==Result==
			===Generate data First Method===
	*The above network fits the galaxy data to one parameter using a two-dimensional convolutional neural network. The following are the training results for training a single parameter - star magnitude.		*The above network fits the galaxy data to one parameter using a two-dimensional convolutional neural network. The following are the training results for training a single parameter - star magnitude.
	[[File:result_111.png\|500px\|right\|jumengting]]		[[File:result_111.png\|500px\|right\|jumengting]]
			===Generate data Second Method===

			===Ground-Truth data Second Method===


	==Else==		==Else==
	Waiting...		Waiting...

2022年10月1日 (六) 08:08 Cluster

2022-10-01T08:08:24Z

←上一版本		2022年10月1日 (六) 08:08的版本
第15行：		第15行：
	===Second NN===		===Second NN===
	*The second task fits the data to multiple parameters.		*The second task fits the data to multiple parameters.
	[[File:~~nn_111~~.png\|500px\|right\|jumengting]]		[[File:Nn_111.png\|500px\|right\|jumengting]]

	==Result==		==Result==

2022年10月1日 (六) 08:07 Cluster

2022-10-01T08:07:27Z

←上一版本		2022年10月1日 (六) 08:07的版本
第7行：		第7行：
	*The cited paper presents a new method for modeling 2D photometric galaxy profiles based on 2D convolutional neural networks: DeepLeGATo for automated and fast analysis of galaxy morphology.		*The cited paper presents a new method for modeling 2D photometric galaxy profiles based on 2D convolutional neural networks: DeepLeGATo for automated and fast analysis of galaxy morphology.

	==Generate ~~First~~ Method==		==Generate Method==
	===~~Generate~~ First Method===		===First Method===
	[[File:111.png\|500px\|right\|jumengting]]		[[File:111.png\|500px\|right\|jumengting]]
	*The neural network structure used in the experiments to construct a neural network structure similar to the original text is shown in the following figure.		*The neural network structure used in the experiments to construct a neural network structure similar to the original text is shown in the following figure.
	*Firstly, GalSim is used to generate information files of relevant galaxies based on random parameters. Because deep learning requires a large number of images and parameter samples, the cost of manual annotation of images is too high; and the expertise of the person conducting the annotation is required. Generating datasets with parameters is a more convenient experimental method.		*Firstly, GalSim is used to generate information files of relevant galaxies based on random parameters. Because deep learning requires a large number of images and parameter samples, the cost of manual annotation of images is too high; and the expertise of the person conducting the annotation is required. Generating datasets with parameters is a more convenient experimental method.
	When generating galaxy data with GalSim software, the dimension size of the generated data is also set to random because the dimension size of the real galaxy data varies. The python library is used to generate 50,000 galaxy information, which is partitioned into a training set, a test set, and a validation set in the ratio of 8:1:1. In order to handle galaxy images of different sizes, this paper uses two-dimensional data with a standard size of 128128 size, which will be different in shape from 128128 size, cut the center of galaxy images larger than 128128 pixels, and fill the edges of images smaller than 128*128 pixels.		When generating galaxy data with GalSim software, the dimension size of the generated data is also set to random because the dimension size of the real galaxy data varies. The python library is used to generate 50,000 galaxy information, which is partitioned into a training set, a test set, and a validation set in the ratio of 8:1:1. In order to handle galaxy images of different sizes, this paper uses two-dimensional data with a standard size of 128128 size, which will be different in shape from 128128 size, cut the center of galaxy images larger than 128128 pixels, and fill the edges of images smaller than 128*128 pixels.
	==~~First~~ ~~Method~~==		===Second NN===
		⚫	*The second task fits the data to multiple parameters.
			[[File:nn_111.png\|500px\|right\|jumengting]]

			==Result==
	*The above network fits the galaxy data to one parameter using a two-dimensional convolutional neural network. The following are the training results for training a single parameter - star magnitude.		*The above network fits the galaxy data to one parameter using a two-dimensional convolutional neural network. The following are the training results for training a single parameter - star magnitude.
	[[File:result_111.png\|500px\|right\|jumengting]]		[[File:result_111.png\|500px\|right\|jumengting]]

	==Else==
⚫	The second task fits the data to multiple parameters.

	==Else==		==Else==
	Waiting...		Waiting...

2022年10月1日 (六) 08:01 Cluster

2022-10-01T08:01:55Z

←上一版本		2022年10月1日 (六) 08:01的版本
第8行：		第8行：

	==Generate First Method==		==Generate First Method==
			===Generate First Method===
	[[File:111.png\|500px\|right\|jumengting]]		[[File:111.png\|500px\|right\|jumengting]]
	*The neural network structure used in the experiments to construct a neural network structure similar to the original text is shown in the following figure.		*The neural network structure used in the experiments to construct a neural network structure similar to the original text is shown in the following figure.

2022年10月1日 (六) 08:01 Cluster

2022-10-01T08:01:36Z

←上一版本		2022年10月1日 (六) 08:01的版本
第7行：		第7行：
	*The cited paper presents a new method for modeling 2D photometric galaxy profiles based on 2D convolutional neural networks: DeepLeGATo for automated and fast analysis of galaxy morphology.		*The cited paper presents a new method for modeling 2D photometric galaxy profiles based on 2D convolutional neural networks: DeepLeGATo for automated and fast analysis of galaxy morphology.

	==First Method==		==Generate First Method==
	[[File:111.png\|500px\|right\|jumengting]]		[[File:111.png\|500px\|right\|jumengting]]
		⚫	*The neural network structure used in the experiments to construct a neural network structure similar to the original text is shown in the following figure.
	*Firstly, GalSim is used to generate information files of relevant galaxies based on random parameters. Because deep learning requires a large number of images and parameter samples, the cost of manual annotation of images is too high; and the expertise of the person conducting the annotation is required. Generating datasets with parameters is a more convenient experimental method.		*Firstly, GalSim is used to generate information files of relevant galaxies based on random parameters. Because deep learning requires a large number of images and parameter samples, the cost of manual annotation of images is too high; and the expertise of the person conducting the annotation is required. Generating datasets with parameters is a more convenient experimental method.
	When generating galaxy data with GalSim software, the dimension size of the generated data is also set to random because the dimension size of the real galaxy data varies. The python library is used to generate 50,000 galaxy information, which is partitioned into a training set, a test set, and a validation set in the ratio of 8:1:1. In order to handle galaxy images of different sizes, this paper uses two-dimensional data with a standard size of 128128 size, which will be different in shape from 128128 size, cut the center of galaxy images larger than 128128 pixels, and fill the edges of images smaller than 128*128 pixels.		When generating galaxy data with GalSim software, the dimension size of the generated data is also set to random because the dimension size of the real galaxy data varies. The python library is used to generate 50,000 galaxy information, which is partitioned into a training set, a test set, and a validation set in the ratio of 8:1:1. In order to handle galaxy images of different sizes, this paper uses two-dimensional data with a standard size of 128128 size, which will be different in shape from 128128 size, cut the center of galaxy images larger than 128128 pixels, and fill the edges of images smaller than 128*128 pixels.
			==First Method==
⚫	*The neural network structure used in the experiments to construct a neural network structure similar to the original text is shown in the following figure.
	*The above network fits the galaxy data to one parameter using a two-dimensional convolutional neural network. The following are the training results for training a single parameter - star magnitude.		*The above network fits the galaxy data to one parameter using a two-dimensional convolutional neural network. The following are the training results for training a single parameter - star magnitude.
	[[File:result_111.png\|500px\|right\|jumengting]]		[[File:result_111.png\|500px\|right\|jumengting]]

2022年10月1日 (六) 07:59 Cluster

2022-10-01T07:59:14Z

←上一版本		2022年10月1日 (六) 07:59的版本
第13行：		第13行：
	*The neural network structure used in the experiments to construct a neural network structure similar to the original text is shown in the following figure.		*The neural network structure used in the experiments to construct a neural network structure similar to the original text is shown in the following figure.
	*The above network fits the galaxy data to one parameter using a two-dimensional convolutional neural network. The following are the training results for training a single parameter - star magnitude.		*The above network fits the galaxy data to one parameter using a two-dimensional convolutional neural network. The following are the training results for training a single parameter - star magnitude.
	[[File:result_111.png\|500px\|~~center~~\|jumengting]]		[[File:result_111.png\|500px\|right\|jumengting]]

	==Else==		==Else==

2022年10月1日 (六) 07:59 Cluster

2022-10-01T07:59:03Z

←上一版本		2022年10月1日 (六) 07:59的版本
第13行：		第13行：
	*The neural network structure used in the experiments to construct a neural network structure similar to the original text is shown in the following figure.		*The neural network structure used in the experiments to construct a neural network structure similar to the original text is shown in the following figure.
	*The above network fits the galaxy data to one parameter using a two-dimensional convolutional neural network. The following are the training results for training a single parameter - star magnitude.		*The above network fits the galaxy data to one parameter using a two-dimensional convolutional neural network. The following are the training results for training a single parameter - star magnitude.
	[[File:result_111.png\|500px\|center~~\|jumengting]] [[File:trian_r2.png\|500px\|right~~\|jumengting]]		[[File:result_111.png\|500px\|center\|jumengting]]
	[[File:test_loss.png\|500px\|center\|jumengting]] [[File:test_r2.png\|500px\|right\|jumengting]]

	==Else==		==Else==

2022年10月1日 (六) 07:57 Cluster

2022-10-01T07:57:48Z

←上一版本		2022年10月1日 (六) 07:57的版本
第13行：		第13行：
	*The neural network structure used in the experiments to construct a neural network structure similar to the original text is shown in the following figure.		*The neural network structure used in the experiments to construct a neural network structure similar to the original text is shown in the following figure.
	*The above network fits the galaxy data to one parameter using a two-dimensional convolutional neural network. The following are the training results for training a single parameter - star magnitude.		*The above network fits the galaxy data to one parameter using a two-dimensional convolutional neural network. The following are the training results for training a single parameter - star magnitude.
	[[File:~~train_loss~~.png\|500px\|center\|jumengting]] [[File:trian_r2.png\|500px\|right\|jumengting]]		[[File:result_111.png\|500px\|center\|jumengting]] [[File:trian_r2.png\|500px\|right\|jumengting]]
	[[File:test_loss.png\|500px\|center\|jumengting]] [[File:test_r2.png\|500px\|right\|jumengting]]		[[File:test_loss.png\|500px\|center\|jumengting]] [[File:test_r2.png\|500px\|right\|jumengting]]

2022年10月1日 (六) 07:46 Cluster

2022-10-01T07:46:59Z

←上一版本		2022年10月1日 (六) 07:46的版本
第13行：		第13行：
	*The neural network structure used in the experiments to construct a neural network structure similar to the original text is shown in the following figure.		*The neural network structure used in the experiments to construct a neural network structure similar to the original text is shown in the following figure.
	*The above network fits the galaxy data to one parameter using a two-dimensional convolutional neural network. The following are the training results for training a single parameter - star magnitude.		*The above network fits the galaxy data to one parameter using a two-dimensional convolutional neural network. The following are the training results for training a single parameter - star magnitude.
	[[File:train_loss.png\|~~400px~~\|center\|jumengting]] [[File:trian_r2.png\|~~400px~~\|right\|jumengting]]		[[File:train_loss.png\|500px\|center\|jumengting]] [[File:trian_r2.png\|500px\|right\|jumengting]]
	[[File:test_loss.png\|~~400px~~\|center\|jumengting]] [[File:test_r2.png\|~~400px~~\|right\|jumengting]]		[[File:test_loss.png\|500px\|center\|jumengting]] [[File:test_r2.png\|500px\|right\|jumengting]]

	==Else==		==Else==