ML IAFILT2 - Revision history

Karsai at 08:41, 1 August 2022

2022-08-01T08:41:56Z

← Older revision		Revision as of 08:41, 1 August 2022
Line 12:		Line 12:
	*{{TAG\|ML_IAFILT2}}=2: The angular filtering function{{cite\|boyd:book:2000}} is described as <math>\eta_{l,a_{\mathrm{FILT}}}=\frac{1}{1+a_{\mathrm{FILT}} [l (l+1)]^{2}} </math>. Using this function the parameter <math>a_{\mathrm{FILT}}</math> has to be defined too. It is set in the {{TAG\|INCAR}} file by setting {{TAG\|ML_AFILT2}}. This option is the default.		*{{TAG\|ML_IAFILT2}}=2: The angular filtering function{{cite\|boyd:book:2000}} is described as <math>\eta_{l,a_{\mathrm{FILT}}}=\frac{1}{1+a_{\mathrm{FILT}} [l (l+1)]^{2}} </math>. Using this function the parameter <math>a_{\mathrm{FILT}}</math> has to be defined too. It is set in the {{TAG\|INCAR}} file by setting {{TAG\|ML_AFILT2}}. This option is the default.

	In the case of the angular descriptor two radial basis functions are multiplied by each other (see [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]). Both basis functions use the same filtering function and hence the filtering is done by the square of the filtering function. This is plotted in Fig. 1 for the two different functions used for {{TAG\|ML_IAFILT2}}=1 and 2 (labeled as TYPE1 and TYPE2, respectively). In the case of {{TAG\|ML_IAFILT2}}=2 it can be seen that for the filtering parameter {{TAG\|ML_AFILT2}}=0.02 and <math>l</math>=5 the function has only a contribution of 0.15. Using this filtering parameter the maximum cut-off for the angular quantum number can be reduced to {{TAG\|ML_LMAX2}}=4.		In the case of the angular descriptor two radial basis functions are multiplied by each other (see [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]). Both basis functions use the same filtering function and hence the filtering is done by the square of the filtering function. This is plotted in Fig. 1 for the two different functions used for {{TAG\|ML_IAFILT2}}=1 and 2 (labeled as TYPE1 and TYPE2, respectively). In the case of {{TAG\|ML_IAFILT2}}=2 it can be seen that for the filtering parameter {{TAG\|ML_AFILT2}}=0.002 and <math>l</math>=5 the function has only a contribution of 0.15. Using this filtering parameter the maximum cut-off for the angular quantum number can be reduced to {{TAG\|ML_LMAX2}}=4.

Miranda.henrique at 14:57, 28 July 2022

2022-07-28T14:57:47Z

← Older revision		Revision as of 14:57, 28 July 2022
Line 10:		Line 10:
	Following cases are possible for the angular filtering function <math>\eta</math> (see also [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]):		Following cases are possible for the angular filtering function <math>\eta</math> (see also [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]):
	*{{TAG\|ML_IAFILT2}}=1: The angular filtering function is described as <math>\eta_{l}=\frac{1}{(2l+1)^{1/4}}</math>.		*{{TAG\|ML_IAFILT2}}=1: The angular filtering function is described as <math>\eta_{l}=\frac{1}{(2l+1)^{1/4}}</math>.
	*{{TAG\|ML_IAFILT2}}=2: The angular filtering function{{cite\|boyd:book:2000}} is described as <math>\eta_{l,a_{\mathrm{FILT}}}=\frac{1}{1+a_{\mathrm{FILT}} [l (l+1)]^{2}} </math>. Using this function the parameter <math>a_{\mathrm{FILT}}</math> has to ~~bes~~ defined too. It is set in the {{TAG\|INCAR}} file by setting {{TAG\|ML_AFILT2}}. This option is the default.		*{{TAG\|ML_IAFILT2}}=2: The angular filtering function{{cite\|boyd:book:2000}} is described as <math>\eta_{l,a_{\mathrm{FILT}}}=\frac{1}{1+a_{\mathrm{FILT}} [l (l+1)]^{2}} </math>. Using this function the parameter <math>a_{\mathrm{FILT}}</math> has to be defined too. It is set in the {{TAG\|INCAR}} file by setting {{TAG\|ML_AFILT2}}. This option is the default.

	In the case of the angular descriptor two radial basis functions are multiplied ~~with~~ each other (see [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]). Both basis functions use the same filtering function and hence the filtering is done by the square of the filtering function. This is plotted in Fig. 1 for the two different functions used for {{TAG\|ML_IAFILT2}}=1 and 2 (labeled as TYPE1 and TYPE2, respectively). In the case of {{TAG\|ML_IAFILT2}}=2 it can be seen that for the ~~default~~ filtering parameter {{TAG\|ML_AFILT2}}=0.02 and <math>l</math>=5 the function has only a contribution of 0.15. Using this filtering parameter the maximum cut off for the angular quantum number can be reduced to {{TAG\|ML_LMAX2}}=4.		In the case of the angular descriptor two radial basis functions are multiplied by each other (see [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]). Both basis functions use the same filtering function and hence the filtering is done by the square of the filtering function. This is plotted in Fig. 1 for the two different functions used for {{TAG\|ML_IAFILT2}}=1 and 2 (labeled as TYPE1 and TYPE2, respectively). In the case of {{TAG\|ML_IAFILT2}}=2 it can be seen that for the filtering parameter {{TAG\|ML_AFILT2}}=0.02 and <math>l</math>=5 the function has only a contribution of 0.15. Using this filtering parameter the maximum cut-off for the angular quantum number can be reduced to {{TAG\|ML_LMAX2}}=4.

Miranda.henrique: Reverted edits by Miranda.henrique (talk) to last revision by Karsai

2022-07-28T14:51:58Z

Reverted edits by Miranda.henrique (talk) to last revision by Karsai

← Older revision		Revision as of 14:51, 28 July 2022
Line 12:		Line 12:
	*{{TAG\|ML_IAFILT2}}=2: The angular filtering function{{cite\|boyd:book:2000}} is described as <math>\eta_{l,a_{\mathrm{FILT}}}=\frac{1}{1+a_{\mathrm{FILT}} [l (l+1)]^{2}} </math>. Using this function the parameter <math>a_{\mathrm{FILT}}</math> has to bes defined too. It is set in the {{TAG\|INCAR}} file by setting {{TAG\|ML_AFILT2}}. This option is the default.		*{{TAG\|ML_IAFILT2}}=2: The angular filtering function{{cite\|boyd:book:2000}} is described as <math>\eta_{l,a_{\mathrm{FILT}}}=\frac{1}{1+a_{\mathrm{FILT}} [l (l+1)]^{2}} </math>. Using this function the parameter <math>a_{\mathrm{FILT}}</math> has to bes defined too. It is set in the {{TAG\|INCAR}} file by setting {{TAG\|ML_AFILT2}}. This option is the default.

	In the case of the angular descriptor two radial basis functions are multiplied with each other (see [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]). Both basis functions use the same filtering function and hence the filtering is done by the square of the filtering function. This is plotted in Fig. 1 for the two different functions used for {{TAG\|ML_IAFILT2}}=1 and 2 (labeled as TYPE1 and TYPE2, respectively). In the case of {{TAG\|ML_IAFILT2}}=2 it can be seen that for the default filtering parameter {{TAG\|ML_AFILT2}}=0.~~002~~ and <math>l</math>=5 the function has only a contribution of 0.15. Using this filtering parameter the maximum cut off for the angular quantum number can be reduced to {{TAG\|ML_LMAX2}}=4.		In the case of the angular descriptor two radial basis functions are multiplied with each other (see [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]). Both basis functions use the same filtering function and hence the filtering is done by the square of the filtering function. This is plotted in Fig. 1 for the two different functions used for {{TAG\|ML_IAFILT2}}=1 and 2 (labeled as TYPE1 and TYPE2, respectively). In the case of {{TAG\|ML_IAFILT2}}=2 it can be seen that for the default filtering parameter {{TAG\|ML_AFILT2}}=0.02 and <math>l</math>=5 the function has only a contribution of 0.15. Using this filtering parameter the maximum cut off for the angular quantum number can be reduced to {{TAG\|ML_LMAX2}}=4.

Miranda.henrique at 14:47, 28 July 2022

2022-07-28T14:47:33Z

← Older revision		Revision as of 14:47, 28 July 2022
Line 12:		Line 12:
	*{{TAG\|ML_IAFILT2}}=2: The angular filtering function{{cite\|boyd:book:2000}} is described as <math>\eta_{l,a_{\mathrm{FILT}}}=\frac{1}{1+a_{\mathrm{FILT}} [l (l+1)]^{2}} </math>. Using this function the parameter <math>a_{\mathrm{FILT}}</math> has to bes defined too. It is set in the {{TAG\|INCAR}} file by setting {{TAG\|ML_AFILT2}}. This option is the default.		*{{TAG\|ML_IAFILT2}}=2: The angular filtering function{{cite\|boyd:book:2000}} is described as <math>\eta_{l,a_{\mathrm{FILT}}}=\frac{1}{1+a_{\mathrm{FILT}} [l (l+1)]^{2}} </math>. Using this function the parameter <math>a_{\mathrm{FILT}}</math> has to bes defined too. It is set in the {{TAG\|INCAR}} file by setting {{TAG\|ML_AFILT2}}. This option is the default.

	In the case of the angular descriptor two radial basis functions are multiplied with each other (see [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]). Both basis functions use the same filtering function and hence the filtering is done by the square of the filtering function. This is plotted in Fig. 1 for the two different functions used for {{TAG\|ML_IAFILT2}}=1 and 2 (labeled as TYPE1 and TYPE2, respectively). In the case of {{TAG\|ML_IAFILT2}}=2 it can be seen that for the default filtering parameter {{TAG\|ML_AFILT2}}=0.02 and <math>l</math>=5 the function has only a contribution of 0.15. Using this filtering parameter the maximum cut off for the angular quantum number can be reduced to {{TAG\|ML_LMAX2}}=4.		In the case of the angular descriptor two radial basis functions are multiplied with each other (see [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]). Both basis functions use the same filtering function and hence the filtering is done by the square of the filtering function. This is plotted in Fig. 1 for the two different functions used for {{TAG\|ML_IAFILT2}}=1 and 2 (labeled as TYPE1 and TYPE2, respectively). In the case of {{TAG\|ML_IAFILT2}}=2 it can be seen that for the default filtering parameter {{TAG\|ML_AFILT2}}=0.002 and <math>l</math>=5 the function has only a contribution of 0.15. Using this filtering parameter the maximum cut off for the angular quantum number can be reduced to {{TAG\|ML_LMAX2}}=4.

Karsai: /* Related tags and articles */

2022-04-08T13:23:28Z

Karsai: /* Related Tags and Sections */

2022-04-07T07:26:10Z

Related Tags and Sections

← Older revision		Revision as of 07:26, 7 April 2022
Line 22:		Line 22:
	<noinclude>		<noinclude>

	== Related ~~Tags~~ and ~~Sections~~ ==		== Related tags and articles ==
	{{TAG\|ML_LMLFF}}, {{TAG\|ML_LAFILT2}}, {{TAG\|ML_AFILT2}}		{{TAG\|ML_LMLFF}}, {{TAG\|ML_LAFILT2}}, {{TAG\|ML_AFILT2}}

Karsai at 07:25, 7 April 2022

2022-04-07T07:25:24Z

← Older revision		Revision as of 07:25, 7 April 2022
Line 1:		Line 1:
			{{DISPLAYTITLE:ML_IAFILT2}}
	{{TAGDEF\|ML_IAFILT2\|[integer]\|2}}		{{TAGDEF\|ML_IAFILT2\|[integer]\|2}}

Line 27:		Line 28:
	----		----

	[[Category:INCAR]][[Category:Machine Learning]][[Category:Machine Learned Force Fields~~]][[Category: Alpha~~]]		[[Category:INCAR tag]][[Category:Machine Learning]][[Category:Machine Learned Force Fields]]

Karsai at 11:13, 6 September 2021

2021-09-06T11:13:13Z

← Older revision		Revision as of 11:13, 6 September 2021
Line 11:		Line 11:
	*{{TAG\|ML_IAFILT2}}=2: The angular filtering function{{cite\|boyd:book:2000}} is described as <math>\eta_{l,a_{\mathrm{FILT}}}=\frac{1}{1+a_{\mathrm{FILT}} [l (l+1)]^{2}} </math>. Using this function the parameter <math>a_{\mathrm{FILT}}</math> has to bes defined too. It is set in the {{TAG\|INCAR}} file by setting {{TAG\|ML_AFILT2}}. This option is the default.		*{{TAG\|ML_IAFILT2}}=2: The angular filtering function{{cite\|boyd:book:2000}} is described as <math>\eta_{l,a_{\mathrm{FILT}}}=\frac{1}{1+a_{\mathrm{FILT}} [l (l+1)]^{2}} </math>. Using this function the parameter <math>a_{\mathrm{FILT}}</math> has to bes defined too. It is set in the {{TAG\|INCAR}} file by setting {{TAG\|ML_AFILT2}}. This option is the default.

	In the case of the angular descriptor two radial basis functions are multiplied with each other (see [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]). Both basis functions use the same filtering function and hence the filtering is done by the square of the filtering function. This is plotted in Fig. 1 for the two different functions used for {{TAG\|~~ML_FF_IAFILT2_MB~~}}=1 and 2 (labeled as TYPE1 and TYPE2, respectively). In the case of {{TAG\|ML_IAFILT2}}=2 it can be seen that for the default filtering parameter {{TAG\|ML_AFILT2}}=0.02 and <math>l</math>=5 the function has only a contribution of 0.15. Using this filtering parameter the maximum cut off for the angular quantum number can be reduced to {{TAG\|ML_LMAX2}}=4.		In the case of the angular descriptor two radial basis functions are multiplied with each other (see [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]). Both basis functions use the same filtering function and hence the filtering is done by the square of the filtering function. This is plotted in Fig. 1 for the two different functions used for {{TAG\|ML_IAFILT2}}=1 and 2 (labeled as TYPE1 and TYPE2, respectively). In the case of {{TAG\|ML_IAFILT2}}=2 it can be seen that for the default filtering parameter {{TAG\|ML_AFILT2}}=0.02 and <math>l</math>=5 the function has only a contribution of 0.15. Using this filtering parameter the maximum cut off for the angular quantum number can be reduced to {{TAG\|ML_LMAX2}}=4.

Karsai at 08:27, 23 August 2021

2021-08-23T08:27:40Z

← Older revision		Revision as of 08:27, 23 August 2021
Line 1:		Line 1:
	{{TAGDEF\|~~ML_FF_IAFILT2_MB~~\|[integer]\|2}}		{{TAGDEF\|ML_IAFILT2\|[integer]\|2}}

	Description: This tag specifies the type of angular filtering used in the machine learning force field method.		Description: This tag specifies the type of angular filtering used in the machine learning force field method.
Line 5:		Line 5:
	[[File:Angular filtering MLFF cropped.png\|400px\|thumb\|Fig. 1: Square of filtering function.]]		[[File:Angular filtering MLFF cropped.png\|400px\|thumb\|Fig. 1: Square of filtering function.]]

	This tag is only used if {{TAG\|~~ML_FF_LAFILT2_MB~~}}=''.TRUE.'' is set.		This tag is only used if {{TAG\|ML_LAFILT2}}=''.TRUE.'' is set.

	Following cases are possible for the angular filtering function <math>\eta</math> (see also [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]):		Following cases are possible for the angular filtering function <math>\eta</math> (see also [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]):
	*{{TAG\|~~ML_FF_IAFILT2_MB~~}}=1: The angular filtering function is described as <math>\eta_{l}=\frac{1}{(2l+1)^{1/4}}</math>.		*{{TAG\|ML_IAFILT2}}=1: The angular filtering function is described as <math>\eta_{l}=\frac{1}{(2l+1)^{1/4}}</math>.
	*{{TAG\|~~ML_FF_IAFILT2_MB~~}}=2: The angular filtering function{{cite\|boyd:book:2000}} is described as <math>\eta_{l,a_{\mathrm{FILT}}}=\frac{1}{1+a_{\mathrm{FILT}} [l (l+1)]^{2}} </math>. Using this function the parameter <math>a_{\mathrm{FILT}}</math> has to bes defined too. It is set in the {{TAG\|INCAR}} file by setting {{TAG\|~~ML_FF_AFILT2_MB~~}}. This option is the default.		*{{TAG\|ML_IAFILT2}}=2: The angular filtering function{{cite\|boyd:book:2000}} is described as <math>\eta_{l,a_{\mathrm{FILT}}}=\frac{1}{1+a_{\mathrm{FILT}} [l (l+1)]^{2}} </math>. Using this function the parameter <math>a_{\mathrm{FILT}}</math> has to bes defined too. It is set in the {{TAG\|INCAR}} file by setting {{TAG\|ML_AFILT2}}. This option is the default.

	In the case of the angular descriptor two radial basis functions are multiplied with each other (see [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]). Both basis functions use the same filtering function and hence the filtering is done by the square of the filtering function. This is plotted in Fig. 1 for the two different functions used for {{TAG\|ML_FF_IAFILT2_MB}}=1 and 2 (labeled as TYPE1 and TYPE2, respectively). In the case of {{TAG\|~~ML_FF_IAFILT2_MB~~}}=2 it can be seen that for the default filtering parameter {{TAG\|~~ML_FF_AFILT2_MB~~}}=0.02 and <math>l</math>=5 the function has only a contribution of 0.15. Using this filtering parameter the maximum cut off for the angular quantum number can be reduced to {{TAG\|~~ML_FF_LMAX2_MB~~}}=4.		In the case of the angular descriptor two radial basis functions are multiplied with each other (see [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]). Both basis functions use the same filtering function and hence the filtering is done by the square of the filtering function. This is plotted in Fig. 1 for the two different functions used for {{TAG\|ML_FF_IAFILT2_MB}}=1 and 2 (labeled as TYPE1 and TYPE2, respectively). In the case of {{TAG\|ML_IAFILT2}}=2 it can be seen that for the default filtering parameter {{TAG\|ML_AFILT2}}=0.02 and <math>l</math>=5 the function has only a contribution of 0.15. Using this filtering parameter the maximum cut off for the angular quantum number can be reduced to {{TAG\|ML_LMAX2}}=4.


Line 22:		Line 22:

	== Related Tags and Sections ==		== Related Tags and Sections ==
	{{TAG\|~~ML_FF_LMLFF~~}}, {{TAG\|~~ML_FF_LAFILT2_MB~~}}, {{TAG\|~~ML_FF_AFILT2_MB~~}}		{{TAG\|ML_LMLFF}}, {{TAG\|ML_LAFILT2}}, {{TAG\|ML_AFILT2}}

	{{sc\|~~ML_FF_IAFILT2_MB~~\|Examples\|Examples that use this tag}}		{{sc\|ML_IAFILT2\|Examples\|Examples that use this tag}}
	----		----

	[[Category:INCAR]][[Category:Machine Learning]][[Category:Machine Learned Force Fields]][[Category: Alpha]]		[[Category:INCAR]][[Category:Machine Learning]][[Category:Machine Learned Force Fields]][[Category: Alpha]]

Karsai: Karsai moved page ML FF IAFILT2 MB to ML IAFILT2

2021-08-23T06:37:35Z

Karsai moved page ML FF IAFILT2 MB to ML IAFILT2

← Older revision	Revision as of 06:37, 23 August 2021
(No difference)

← Older revision		Revision as of 14:57, 28 July 2022
Line 10:		Line 10:
	Following cases are possible for the angular filtering function <math>\eta</math> (see also [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]):		Following cases are possible for the angular filtering function <math>\eta</math> (see also [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]):
	*{{TAG\|ML_IAFILT2}}=1: The angular filtering function is described as <math>\eta_{l}=\frac{1}{(2l+1)^{1/4}}</math>.		*{{TAG\|ML_IAFILT2}}=1: The angular filtering function is described as <math>\eta_{l}=\frac{1}{(2l+1)^{1/4}}</math>.
	*{{TAG\|ML_IAFILT2}}=2: The angular filtering function{{cite\|boyd:book:2000}} is described as <math>\eta_{l,a_{\mathrm{FILT}}}=\frac{1}{1+a_{\mathrm{FILT}} [l (l+1)]^{2}} </math>. Using this function the parameter <math>a_{\mathrm{FILT}}</math> has to ~~bes~~ defined too. It is set in the {{TAG\|INCAR}} file by setting {{TAG\|ML_AFILT2}}. This option is the default.		*{{TAG\|ML_IAFILT2}}=2: The angular filtering function{{cite\|boyd:book:2000}} is described as <math>\eta_{l,a_{\mathrm{FILT}}}=\frac{1}{1+a_{\mathrm{FILT}} [l (l+1)]^{2}} </math>. Using this function the parameter <math>a_{\mathrm{FILT}}</math> has to be defined too. It is set in the {{TAG\|INCAR}} file by setting {{TAG\|ML_AFILT2}}. This option is the default.

	In the case of the angular descriptor two radial basis functions are multiplied ~~with~~ each other (see [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]). Both basis functions use the same filtering function and hence the filtering is done by the square of the filtering function. This is plotted in Fig. 1 for the two different functions used for {{TAG\|ML_IAFILT2}}=1 and 2 (labeled as TYPE1 and TYPE2, respectively). In the case of {{TAG\|ML_IAFILT2}}=2 it can be seen that for the ~~default~~ filtering parameter {{TAG\|ML_AFILT2}}=0.02 and <math>l</math>=5 the function has only a contribution of 0.15. Using this filtering parameter the maximum cut off for the angular quantum number can be reduced to {{TAG\|ML_LMAX2}}=4.		In the case of the angular descriptor two radial basis functions are multiplied by each other (see [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]). Both basis functions use the same filtering function and hence the filtering is done by the square of the filtering function. This is plotted in Fig. 1 for the two different functions used for {{TAG\|ML_IAFILT2}}=1 and 2 (labeled as TYPE1 and TYPE2, respectively). In the case of {{TAG\|ML_IAFILT2}}=2 it can be seen that for the filtering parameter {{TAG\|ML_AFILT2}}=0.02 and <math>l</math>=5 the function has only a contribution of 0.15. Using this filtering parameter the maximum cut-off for the angular quantum number can be reduced to {{TAG\|ML_LMAX2}}=4.

← Older revision		Revision as of 08:27, 23 August 2021
Line 1:		Line 1:
	{{TAGDEF\|~~ML_FF_IAFILT2_MB~~\|[integer]\|2}}		{{TAGDEF\|ML_IAFILT2\|[integer]\|2}}

	Description: This tag specifies the type of angular filtering used in the machine learning force field method.		Description: This tag specifies the type of angular filtering used in the machine learning force field method.
Line 5:		Line 5:
	[[File:Angular filtering MLFF cropped.png\|400px\|thumb\|Fig. 1: Square of filtering function.]]		[[File:Angular filtering MLFF cropped.png\|400px\|thumb\|Fig. 1: Square of filtering function.]]

	This tag is only used if {{TAG\|~~ML_FF_LAFILT2_MB~~}}=''.TRUE.'' is set.		This tag is only used if {{TAG\|ML_LAFILT2}}=''.TRUE.'' is set.

	Following cases are possible for the angular filtering function <math>\eta</math> (see also [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]):		Following cases are possible for the angular filtering function <math>\eta</math> (see also [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]):
	*{{TAG\|~~ML_FF_IAFILT2_MB~~}}=1: The angular filtering function is described as <math>\eta_{l}=\frac{1}{(2l+1)^{1/4}}</math>.		*{{TAG\|ML_IAFILT2}}=1: The angular filtering function is described as <math>\eta_{l}=\frac{1}{(2l+1)^{1/4}}</math>.
	*{{TAG\|~~ML_FF_IAFILT2_MB~~}}=2: The angular filtering function{{cite\|boyd:book:2000}} is described as <math>\eta_{l,a_{\mathrm{FILT}}}=\frac{1}{1+a_{\mathrm{FILT}} [l (l+1)]^{2}} </math>. Using this function the parameter <math>a_{\mathrm{FILT}}</math> has to bes defined too. It is set in the {{TAG\|INCAR}} file by setting {{TAG\|~~ML_FF_AFILT2_MB~~}}. This option is the default.		*{{TAG\|ML_IAFILT2}}=2: The angular filtering function{{cite\|boyd:book:2000}} is described as <math>\eta_{l,a_{\mathrm{FILT}}}=\frac{1}{1+a_{\mathrm{FILT}} [l (l+1)]^{2}} </math>. Using this function the parameter <math>a_{\mathrm{FILT}}</math> has to bes defined too. It is set in the {{TAG\|INCAR}} file by setting {{TAG\|ML_AFILT2}}. This option is the default.

	In the case of the angular descriptor two radial basis functions are multiplied with each other (see [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]). Both basis functions use the same filtering function and hence the filtering is done by the square of the filtering function. This is plotted in Fig. 1 for the two different functions used for {{TAG\|ML_FF_IAFILT2_MB}}=1 and 2 (labeled as TYPE1 and TYPE2, respectively). In the case of {{TAG\|~~ML_FF_IAFILT2_MB~~}}=2 it can be seen that for the default filtering parameter {{TAG\|~~ML_FF_AFILT2_MB~~}}=0.02 and <math>l</math>=5 the function has only a contribution of 0.15. Using this filtering parameter the maximum cut off for the angular quantum number can be reduced to {{TAG\|~~ML_FF_LMAX2_MB~~}}=4.		In the case of the angular descriptor two radial basis functions are multiplied with each other (see [[On-the-fly machine learning force field generation using Bayesian linear regression#Basis set expansion\|here]]). Both basis functions use the same filtering function and hence the filtering is done by the square of the filtering function. This is plotted in Fig. 1 for the two different functions used for {{TAG\|ML_FF_IAFILT2_MB}}=1 and 2 (labeled as TYPE1 and TYPE2, respectively). In the case of {{TAG\|ML_IAFILT2}}=2 it can be seen that for the default filtering parameter {{TAG\|ML_AFILT2}}=0.02 and <math>l</math>=5 the function has only a contribution of 0.15. Using this filtering parameter the maximum cut off for the angular quantum number can be reduced to {{TAG\|ML_LMAX2}}=4.


Line 22:		Line 22:

	== Related Tags and Sections ==		== Related Tags and Sections ==
	{{TAG\|~~ML_FF_LMLFF~~}}, {{TAG\|~~ML_FF_LAFILT2_MB~~}}, {{TAG\|~~ML_FF_AFILT2_MB~~}}		{{TAG\|ML_LMLFF}}, {{TAG\|ML_LAFILT2}}, {{TAG\|ML_AFILT2}}

	{{sc\|~~ML_FF_IAFILT2_MB~~\|Examples\|Examples that use this tag}}		{{sc\|ML_IAFILT2\|Examples\|Examples that use this tag}}
	----		----

	[[Category:INCAR]][[Category:Machine Learning]][[Category:Machine Learned Force Fields]][[Category: Alpha]]		[[Category:INCAR]][[Category:Machine Learning]][[Category:Machine Learned Force Fields]][[Category: Alpha]]

← Older revision		Revision as of 13:23, 8 April 2022
Line 28:		Line 28:
	----		----

	[[Category:INCAR tag]][[Category:Machine ~~Learning]][[Category:Machine Learned Force Fields~~]]		[[Category:INCAR tag]][[Category:Machine-learned force fields]]