ML FFN - Revision history

Singraber at 07:57, 24 October 2025

2025-10-24T07:57:35Z

← Older revision		Revision as of 07:57, 24 October 2025
Line 1:		Line 1:
	{{DISPLAYTITLE:ML_FFN}}		{{DISPLAYTITLE:ML_FFN}}
	The {{FILE\|ML_FFN}} file is a binary file that contains a force field generated from [[Machine learning force field calculations: Basics\|machine-learning runs]] with the options {{~~TAGO~~\|ML_MODE\|train, refit or select}}. {{FILE\|ML_FFN}} and {{FILE\|ML_FF}} file structures are identical, just the former file name is used for {{VASP}} output files while the latter is the expected input file name. In order to apply a force field, copy the {{FILE\|ML_FFN}} file to {{FILE\|ML_FF}} and set {{~~TAGO~~\|ML_MODE\|run}} in the {{FILE\|INCAR}} file. See [[Machine_learning_force_field_calculations:_Basics#Step-by-step_instructions\|machine learning force field calculations: Basics]] for the basic workflow.		The {{FILE\|ML_FFN}} file is a binary file that contains a force field generated from [[Machine learning force field calculations: Basics\|machine-learning runs]] with the options {{TAG\|ML_MODE\|train, refit or select}}. {{FILE\|ML_FFN}} and {{FILE\|ML_FF}} file structures are identical, just the former file name is used for {{VASP}} output files while the latter is the expected input file name. In order to apply a force field, copy the {{FILE\|ML_FFN}} file to {{FILE\|ML_FF}} and set {{TAG\|ML_MODE\|run}} in the {{FILE\|INCAR}} file. See [[Machine_learning_force_field_calculations:_Basics#Step-by-step_instructions\|machine learning force field calculations: Basics]] for the basic workflow.
	{{NB\|mind\|Available as of {{VASP}} 6.3.0. File header and fast prediction mode available as of {{VASP}} 6.4.0}}		{{NB\|mind\|Available as of {{VASP}} 6.3.0. File header and fast prediction mode available as of {{VASP}} 6.4.0}}
	== File header ==		== File header ==
Line 47:		Line 47:
	== Fast prediction mode ==		== Fast prediction mode ==

	As of {{VASP}} 6.4.0 a crucial step to achieve optimal execution speeds when a force field is deemed ready for production is to apply the "refit" mode ({{~~TAGO~~\|ML_MODE\|refit}}, this is also outlined in the [[Machine_learning_force_field_calculations:_Basics#Step-by-step_instructions\|step-by-step instructions]]). Similarly to training (continuation) ({{~~TAGO~~\|ML_MODE\|train}}) or selection ({{~~TAGO~~\|ML_MODE\|select}}) runs there will be a new {{FILE\|ML_FFN}} generated. However, in contrast to the other options the force field from refitting is ready for use of fast prediction mode during prediction-only runs ({{~~TAGO~~\|ML_MODE\|run}}). We can query the <code>"ML_LFAST"</code> key-value pair in the JSON part of the file header to check whether fast prediction mode is enabled:		As of {{VASP}} 6.4.0 a crucial step to achieve optimal execution speeds when a force field is deemed ready for production is to apply the "refit" mode ({{TAG\|ML_MODE\|refit}}, this is also outlined in the [[Machine_learning_force_field_calculations:_Basics#Step-by-step_instructions\|step-by-step instructions]]). Similarly to training (continuation) ({{TAG\|ML_MODE\|train}}) or selection ({{TAG\|ML_MODE\|select}}) runs there will be a new {{FILE\|ML_FFN}} generated. However, in contrast to the other options the force field from refitting is ready for use of fast prediction mode during prediction-only runs ({{TAG\|ML_MODE\|run}}). We can query the <code>"ML_LFAST"</code> key-value pair in the JSON part of the file header to check whether fast prediction mode is enabled:
	* <code>"ML_LFAST" : true</code>: The {{FILE\|ML_FFN}} file is ready for the fast prediction mode, no Bayesian error estimates are possible (no <code>BEEF</code> lines in {{FILE\|ML_LOGFILE}}).		* <code>"ML_LFAST" : true</code>: The {{FILE\|ML_FFN}} file is ready for the fast prediction mode, no Bayesian error estimates are possible (no <code>BEEF</code> lines in {{FILE\|ML_LOGFILE}}).
	* <code>"ML_LFAST" : false</code>: Fast prediction mode is not possible, {{~~TAGO~~\|ML_MODE\|run}} runs much slower but provides Bayesian error estimate output in {{FILE\|ML_LOGFILE}}.		* <code>"ML_LFAST" : false</code>: Fast prediction mode is not possible, {{TAG\|ML_MODE\|run}} runs much slower but provides Bayesian error estimate output in {{FILE\|ML_LOGFILE}}.
	If the fast prediction mode is available, {{VASP}} will automatically use it because this is highly recommended for production runs. However, if refitting is performed for other reasons (e.g. [[Best_practices_for_machine-learned_force_fields#Retraining_with_re-selection_of_local_reference_configurations\|hyper-parameter optimization]]) it may be useful to create a "slow" force field which allows to compute Bayesian error estimates. This can be achieved with the alternative refitting mode {{~~TAGO~~\|ML_MODE\|refitbayesian}}.		If the fast prediction mode is available, {{VASP}} will automatically use it because this is highly recommended for production runs. However, if refitting is performed for other reasons (e.g. [[Best_practices_for_machine-learned_force_fields#Retraining_with_re-selection_of_local_reference_configurations\|hyper-parameter optimization]]) it may be useful to create a "slow" force field which allows to compute Bayesian error estimates. This can be achieved with the alternative refitting mode {{TAG\|ML_MODE\|refitbayesian}}.

	== Related tags and articles ==		== Related tags and articles ==

Huebsch at 08:55, 23 October 2025

2025-10-23T08:55:24Z

← Older revision		Revision as of 08:55, 23 October 2025
Line 1:		Line 1:
	{{DISPLAYTITLE:ML_FFN}} {{NB\|~~mind\|Available as of {{VASP~~}} ~~6.3.0. File header and fast prediction mode available as of {{VASP}} 6.4.0}} This~~ binary file contains a force field generated from machine learning runs with the options {{TAGO\|ML_MODE\|train, refit or select}}. {{FILE\|ML_FFN}} and {{FILE\|ML_FF}} file structures are identical, just the former file name is used for {{VASP}} output files while the latter is the expected input file name. In order to apply a force field ~~(in~~ {{FILE\|ML_FFN}}~~) created by {{VASP}} the~~ file ~~needs to be copied~~ to {{FILE\|ML_FF}} and ~~the~~ {{~~FILE~~\|~~INCAR~~}} ~~tag~~ {{~~TAGO~~\|~~ML_MODE\|run~~}} ~~has to be set (see~~ [[Machine_learning_force_field_calculations:_Basics#Step-by-step_instructions\|~~here~~]] for the basic workflow).		{{DISPLAYTITLE:ML_FFN}}
			The {{FILE\|ML_FFN}} file is a binary file that contains a force field generated from [[Machine learning force field calculations: Basics\|machine-learning runs]] with the options {{TAGO\|ML_MODE\|train, refit or select}}. {{FILE\|ML_FFN}} and {{FILE\|ML_FF}} file structures are identical, just the former file name is used for {{VASP}} output files while the latter is the expected input file name. In order to apply a force field, copy the {{FILE\|ML_FFN}} file to {{FILE\|ML_FF}} and set {{TAGO\|ML_MODE\|run}} in the {{FILE\|INCAR}} file. See [[Machine_learning_force_field_calculations:_Basics#Step-by-step_instructions\|machine learning force field calculations: Basics]] for the basic workflow.
			{{NB\|mind\|Available as of {{VASP}} 6.3.0. File header and fast prediction mode available as of {{VASP}} 6.4.0}}
	== File header ==		== File header ==

Singraber at 09:41, 16 October 2025

2025-10-16T09:41:23Z

← Older revision		Revision as of 09:41, 16 October 2025
Line 16:		Line 16:
	\| 6.3.X		\| 6.3.X
	\| 0.1.0		\| 0.1.0
	\| Initial version (no file header).		\| Initial version (no file header)
	\|-		\|-
	\| 6.4.0		\| 6.4.0
	\| 0.2.0		\| 0.2.0
	\| Fast prediction mode, ASCII header.		\| Fast prediction mode, ASCII header
	\|-		\|-
	\| 6.4.1		\| 6.4.1

Singraber at 09:37, 16 October 2025

2025-10-16T09:37:11Z

@@ Line 7: / Line 7: @@
 The output may look like this:
   ML_FF 0.2.1 binary { "date" : "2023-03-16T13:49:44.829", "ML_LFAST" : false, "ML_DESC_TYPE" :   0, "types" : [ "Si" ], "training_structures" : 984, "local_reference_cfgs" : [ 110 ], "descriptors" : [ 142 ], "ML_IALGO_LINREG" : 3, "ML_RCUT1" :  6.0000E+00, "ML_RCUT2" :  6.0000E+00, "ML_W1" :  1.0000E-01, "ML_SION1" :  5.0000E-01, "ML_SION2" :  5.0000E-01, "ML_LMAX2" : 4, "ML_MRB1" : 8, "ML_MRB2" : 8, "ML_IWEIGHT" : 3, "ML_WTOTEN" :  1.0000E+00, "ML_WTIFOR" :  1.0000E+00, "ML_WTSIF" :  1.0000E-10 }
-followed by some extra spaces (because the header is always 4096 characters long). After the string <code>ML_FF</code> the header contains the file version number. {{VASP}} machine-learned force field files are versioned since {{VASP}} 6.4.0. This is necessary because whenever a new feature related to machine-learned force fields is available the contents of the generated files could change. Hence, when reading an {{FILE|ML_FF}} file a {{VASP}} executable deduces the possible features from the file's version number, i.e., performs a compatibility check. Typically, {{FILE|ML_FFN}} files are backward compatible, i.e., newer versions of {{VASP}} will be able to read files generated by older {{VASP}} versions. The opposite is generally not true. However, the {{FILE|ML_FFN}} version number does not automatically increase with the {{VASP}} version, so two or more consecutive {{VASP}} versions could create {{FILE|ML_FFN}} files with identical version number. The version number in the header is followed by the word <code>binary</code>. The remainder of the header contains a [https://www.json.org JSON] string with general information about the force field and values of selected {{FILE|INCAR}} tags. {{NB|tip|The timestamp following the <code>"date"</code> key is also written to the <code>FFOUT</code> lines in {{FILE|ML_LOGFILE}} which helps to identify later on which force field files belong to which log files.}}
+followed by some extra spaces (because the header is always 4096 characters long). After the string <code>ML_FF</code> the header contains the file version number. {{VASP}} machine-learned force field files are versioned since {{VASP}} 6.4.0. This is necessary because whenever a new feature related to machine-learned force fields is available the contents of the generated files could change. Hence, when reading an {{FILE|ML_FF}} file a {{VASP}} executable deduces the possible features with the help of the file's version number, i.e., performs a compatibility check. Typically, {{FILE|ML_FFN}} files are backward compatible, i.e., newer versions of {{VASP}} will be able to read files generated by older {{VASP}} versions. The opposite is generally not true: if you provide an {{FILE|ML_FF}} file to an older {{VASP}} version you may receive an error message stating that the file is too new and not supported. However, the {{FILE|ML_FFN}} version number does not automatically increase with the {{VASP}} version, so two or more consecutive {{VASP}} versions could create {{FILE|ML_FFN}} files with identical version number. The following table gives an overview of existing {{FILE|ML_FFN}} file versions:
 == Fast prediction mode ==

Singraber: /* Fast prediction mode */

2025-10-16T08:49:46Z

Fast prediction mode

← Older revision		Revision as of 08:49, 16 October 2025
Line 11:		Line 11:
	== Fast prediction mode ==		== Fast prediction mode ==

	A crucial step to achieve optimal execution speeds when a force field is deemed ready for production is to apply the "refit" mode ({{TAGO\|ML_MODE\|refit}}, this is also outlined in the [[Machine_learning_force_field_calculations:_Basics#Step-by-step_instructions\|step-by-step instructions]]). Similarly to training (continuation) ({{TAGO\|ML_MODE\|train}}) or selection ({{TAGO\|ML_MODE\|select}}) runs there will be a new {{FILE\|ML_FFN}} generated. However, in contrast to the other options the force field from refitting is ready for use of fast prediction mode during prediction-only runs ({{TAGO\|ML_MODE\|run}}). We can query the <code>"ML_LFAST"</code> key-value pair in the JSON part of the file header to check whether fast prediction mode is enabled:		As of {{VASP}} 6.4.0 a crucial step to achieve optimal execution speeds when a force field is deemed ready for production is to apply the "refit" mode ({{TAGO\|ML_MODE\|refit}}, this is also outlined in the [[Machine_learning_force_field_calculations:_Basics#Step-by-step_instructions\|step-by-step instructions]]). Similarly to training (continuation) ({{TAGO\|ML_MODE\|train}}) or selection ({{TAGO\|ML_MODE\|select}}) runs there will be a new {{FILE\|ML_FFN}} generated. However, in contrast to the other options the force field from refitting is ready for use of fast prediction mode during prediction-only runs ({{TAGO\|ML_MODE\|run}}). We can query the <code>"ML_LFAST"</code> key-value pair in the JSON part of the file header to check whether fast prediction mode is enabled:
	* <code>"ML_LFAST" : true</code>: The {{FILE\|ML_FFN}} file is ready for the fast prediction mode, no Bayesian error estimates are possible (no <code>BEEF</code> lines in {{FILE\|ML_LOGFILE}}).		* <code>"ML_LFAST" : true</code>: The {{FILE\|ML_FFN}} file is ready for the fast prediction mode, no Bayesian error estimates are possible (no <code>BEEF</code> lines in {{FILE\|ML_LOGFILE}}).
	* <code>"ML_LFAST" : false</code>: Fast prediction mode is not possible, {{TAGO\|ML_MODE\|run}} runs much slower but provides Bayesian error estimate output in {{FILE\|ML_LOGFILE}}.		* <code>"ML_LFAST" : false</code>: Fast prediction mode is not possible, {{TAGO\|ML_MODE\|run}} runs much slower but provides Bayesian error estimate output in {{FILE\|ML_LOGFILE}}.

Singraber at 08:48, 16 October 2025

2025-10-16T08:48:38Z

← Older revision		Revision as of 08:48, 16 October 2025
Line 1:		Line 1:
	{{DISPLAYTITLE:ML_FFN}}		{{DISPLAYTITLE:ML_FFN}} {{NB\|mind\|Available as of {{VASP}} 6.3.0. File header and fast prediction mode available as of {{VASP}} 6.4.0}} This binary file contains a force field generated from machine learning runs with the options {{TAGO\|ML_MODE\|train, refit or select}}. {{FILE\|ML_FFN}} and {{FILE\|ML_FF}} file structures are identical, just the former file name is used for {{VASP}} output files while the latter is the expected input file name. In order to apply a force field (in {{FILE\|ML_FFN}}) created by {{VASP}} the file needs to be copied to {{FILE\|ML_FF}} and the {{FILE\|INCAR}} tag {{TAGO\|ML_MODE\|run}} has to be set (see [[Machine_learning_force_field_calculations:_Basics#Step-by-step_instructions\|here]] for the basic workflow).
	This binary file contains a ~~newly created~~ force field from machine learning runs with the options {{~~TAG~~\|ML_MODE~~}}=<code>~~train~~</code>~~, ~~<code>~~refit~~</code>~~ or ~~<code>~~select~~</code>~~. ~~It's structure is~~ identical to the {{~~TAG\|ML_FF~~}} file. ~~To be able~~ to ~~use the new~~ force field in ~~the~~ {{FILE\|ML_FFN}} file ~~it has~~ to be ~~simply~~ copied to {{FILE\|ML_FF}} and the {{FILE\|INCAR}} tag {{~~TAG~~\|ML_MODE}}~~=<code>run</code>~~ has to be set (see [[Machine_learning_force_field_calculations:_Basics#Step-by-step_instructions\|here]] for the basic workflow).

	== ~~{{FILE\|ML_FFN}} file~~ header ==		== File header ==

	Since VASP 6.4.0 the {{FILE\|ML_FFN}} file starts with an ASCII header containing the most important INCAR tags in effect during generation of this force field. In Linux shells this can be easily extracted issuing the following command:		Since VASP 6.4.0 the {{FILE\|ML_FFN}} file starts with an ASCII header containing the most important INCAR tags in effect during generation of this force field. In Linux shells this can be easily extracted issuing the following command:
	head -n 1 ML_FFN		head -n 1 ML_FFN
	The output may look like this:		The output may look like this:
	ML_FF 0.2.1 binary { "date" : "2023-03-16T13:49:44.829", "ML_LFAST" : ~~False~~, "ML_DESC_TYPE" : 0, "types" : [ "Si" ], "training_structures" : 984, "local_reference_cfgs" : [ 110 ], "descriptors" : [ 142 ], "ML_IALGO_LINREG" : 3, "ML_RCUT1" : 6.0000E+00, "ML_RCUT2" : 6.0000E+00, "ML_W1" : 1.0000E-01, "ML_SION1" : 5.0000E-01, "ML_SION2" : 5.0000E-01, "ML_LMAX2" : 4, "ML_MRB1" : 8, "ML_MRB2" : 8, "ML_IWEIGHT" : 3, "ML_WTOTEN" : 1.0000E+00, "ML_WTIFOR" : 1.0000E+00, "ML_WTSIF" : 1.0000E-10 }		ML_FF 0.2.1 binary { "date" : "2023-03-16T13:49:44.829", "ML_LFAST" : false, "ML_DESC_TYPE" : 0, "types" : [ "Si" ], "training_structures" : 984, "local_reference_cfgs" : [ 110 ], "descriptors" : [ 142 ], "ML_IALGO_LINREG" : 3, "ML_RCUT1" : 6.0000E+00, "ML_RCUT2" : 6.0000E+00, "ML_W1" : 1.0000E-01, "ML_SION1" : 5.0000E-01, "ML_SION2" : 5.0000E-01, "ML_LMAX2" : 4, "ML_MRB1" : 8, "ML_MRB2" : 8, "ML_IWEIGHT" : 3, "ML_WTOTEN" : 1.0000E+00, "ML_WTIFOR" : 1.0000E+00, "ML_WTSIF" : 1.0000E-10 }
	followed by some extra spaces (because the header is always 4096 characters long). After the string <code>ML_FF</code> the header contains the file version number. {{VASP}} machine-learned force field files are versioned since {{VASP}} 6.4.0. This is necessary because whenever a new feature related to machine-learned force fields is available the contents of the generated files could change. Hence, when reading an {{FILE\|ML_FF}} file a {{VASP}} executable deduces the possible features from the file's version number, i.e., performs a compatibility check. Typically, {{FILE\|ML_FFN}} files are backward compatible, i.e., newer versions of {{VASP}} will be able to read files generated by older {{VASP}} versions. The opposite is generally not true. However, the {{FILE\|ML_FFN}} version number does not automatically increase with the {{VASP}} version, so two or more consecutive {{VASP}} versions could create {{FILE\|ML_FFN}} files with identical version number. The version number in the header is followed by the word <code>binary</code>. The remainder of the header contains a [https://www.json.org JSON] string with general information about the force field and values of selected {{FILE\|INCAR}} tags. ~~For example, the~~ timestamp following the <code>"date"</code> key is also written to the <code>FFOUT</code> lines in {{FILE\|ML_LOGFILE}} which helps to identify later on which force field files belong to which log files. ~~Another important piece of information is <code>"ML_LFAST"</code> key which allows you to check whether this force field is ready for fast prediction mode.~~		followed by some extra spaces (because the header is always 4096 characters long). After the string <code>ML_FF</code> the header contains the file version number. {{VASP}} machine-learned force field files are versioned since {{VASP}} 6.4.0. This is necessary because whenever a new feature related to machine-learned force fields is available the contents of the generated files could change. Hence, when reading an {{FILE\|ML_FF}} file a {{VASP}} executable deduces the possible features from the file's version number, i.e., performs a compatibility check. Typically, {{FILE\|ML_FFN}} files are backward compatible, i.e., newer versions of {{VASP}} will be able to read files generated by older {{VASP}} versions. The opposite is generally not true. However, the {{FILE\|ML_FFN}} version number does not automatically increase with the {{VASP}} version, so two or more consecutive {{VASP}} versions could create {{FILE\|ML_FFN}} files with identical version number. The version number in the header is followed by the word <code>binary</code>. The remainder of the header contains a [https://www.json.org JSON] string with general information about the force field and values of selected {{FILE\|INCAR}} tags. {{NB\|tip\|The timestamp following the <code>"date"</code> key is also written to the <code>FFOUT</code> lines in {{FILE\|ML_LOGFILE}} which helps to identify later on which force field files belong to which log files.}}

	~~<!--~~
	== Fast prediction mode ==		== Fast prediction mode ==

	As outlined in the [[Machine_learning_force_field_calculations:_Basics#Step-by-step_instructions\|step-by-step instructions]] -->		A crucial step to achieve optimal execution speeds when a force field is deemed ready for production is to apply the "refit" mode ({{TAGO\|ML_MODE\|refit}}, this is also outlined in the [[Machine_learning_force_field_calculations:_Basics#Step-by-step_instructions\|step-by-step instructions]]). Similarly to training (continuation) ({{TAGO\|ML_MODE\|train}}) or selection ({{TAGO\|ML_MODE\|select}}) runs there will be a new {{FILE\|ML_FFN}} generated. However, in contrast to the other options the force field from refitting is ready for use of fast prediction mode during prediction-only runs ({{TAGO\|ML_MODE\|run}}). We can query the <code>"ML_LFAST"</code> key-value pair in the JSON part of the file header to check whether fast prediction mode is enabled:
			* <code>"ML_LFAST" : true</code>: The {{FILE\|ML_FFN}} file is ready for the fast prediction mode, no Bayesian error estimates are possible (no <code>BEEF</code> lines in {{FILE\|ML_LOGFILE}}).
			* <code>"ML_LFAST" : false</code>: Fast prediction mode is not possible, {{TAGO\|ML_MODE\|run}} runs much slower but provides Bayesian error estimate output in {{FILE\|ML_LOGFILE}}.
			If the fast prediction mode is available, {{VASP}} will automatically use it because this is highly recommended for production runs. However, if refitting is performed for other reasons (e.g. [[Best_practices_for_machine-learned_force_fields#Retraining_with_re-selection_of_local_reference_configurations\|hyper-parameter optimization]]) it may be useful to create a "slow" force field which allows to compute Bayesian error estimates. This can be achieved with the alternative refitting mode {{TAGO\|ML_MODE\|refitbayesian}}.

			== Related tags and articles ==
			{{TAG\|ML_LMLFF}}, {{TAG\|ML_MODE}}, {{TAG\|ML_LFAST}}, {{TAG\|ML_ESTBLOCK}}, {{FILE\|ML_LOGFILE}}
	----		----

	[[Category:Files]][[Category:Machine-learned force fields]][[Category:Output files]]		[[Category:Files]][[Category:Machine-learned force fields]][[Category:Output files]]

Singraber at 13:56, 15 October 2025

2025-10-15T13:56:21Z

← Older revision		Revision as of 13:56, 15 October 2025
Line 1:		Line 1:
	{{DISPLAYTITLE:ML_FFN}}		{{DISPLAYTITLE:ML_FFN}}
	This binary file contains a newly created force field from machine learning runs with the options {{TAG\|ML_MODE}}=<code>train</code>, <code>refit</code> or <code>select</code>. It's structure is identical to the {{TAG\|ML_FF}} file. To be able to use the new force field in the {{FILE\|ML_FFN}} file it has to be simply copied to {{FILE\|ML_FF}} and the {{FILE\|INCAR}} tag {{TAG\|ML_MODE}}=<code>run</code> has to be set (see [[Machine_learning_force_field_calculations:_Basics#Step-by-step_instructions\|here]] for the basic workflow).		This binary file contains a newly created force field from machine learning runs with the options {{TAG\|ML_MODE}}=<code>train</code>, <code>refit</code> or <code>select</code>. It's structure is identical to the {{TAG\|ML_FF}} file. To be able to use the new force field in the {{FILE\|ML_FFN}} file it has to be simply copied to {{FILE\|ML_FF}} and the {{FILE\|INCAR}} tag {{TAG\|ML_MODE}}=<code>run</code> has to be set (see [[Machine_learning_force_field_calculations:_Basics#Step-by-step_instructions\|here]] for the basic workflow).

			== {{FILE\|ML_FFN}} file header ==

	Since VASP 6.4.0 the {{FILE\|ML_FFN}} file starts with an ASCII header containing the most important INCAR tags in effect during generation of this force field. In Linux shells this can be easily extracted issuing the following command:		Since VASP 6.4.0 the {{FILE\|ML_FFN}} file starts with an ASCII header containing the most important INCAR tags in effect during generation of this force field. In Linux shells this can be easily extracted issuing the following command:
Line 6:		Line 8:
	The output may look like this:		The output may look like this:
	ML_FF 0.2.1 binary { "date" : "2023-03-16T13:49:44.829", "ML_LFAST" : False, "ML_DESC_TYPE" : 0, "types" : [ "Si" ], "training_structures" : 984, "local_reference_cfgs" : [ 110 ], "descriptors" : [ 142 ], "ML_IALGO_LINREG" : 3, "ML_RCUT1" : 6.0000E+00, "ML_RCUT2" : 6.0000E+00, "ML_W1" : 1.0000E-01, "ML_SION1" : 5.0000E-01, "ML_SION2" : 5.0000E-01, "ML_LMAX2" : 4, "ML_MRB1" : 8, "ML_MRB2" : 8, "ML_IWEIGHT" : 3, "ML_WTOTEN" : 1.0000E+00, "ML_WTIFOR" : 1.0000E+00, "ML_WTSIF" : 1.0000E-10 }		ML_FF 0.2.1 binary { "date" : "2023-03-16T13:49:44.829", "ML_LFAST" : False, "ML_DESC_TYPE" : 0, "types" : [ "Si" ], "training_structures" : 984, "local_reference_cfgs" : [ 110 ], "descriptors" : [ 142 ], "ML_IALGO_LINREG" : 3, "ML_RCUT1" : 6.0000E+00, "ML_RCUT2" : 6.0000E+00, "ML_W1" : 1.0000E-01, "ML_SION1" : 5.0000E-01, "ML_SION2" : 5.0000E-01, "ML_LMAX2" : 4, "ML_MRB1" : 8, "ML_MRB2" : 8, "ML_IWEIGHT" : 3, "ML_WTOTEN" : 1.0000E+00, "ML_WTIFOR" : 1.0000E+00, "ML_WTSIF" : 1.0000E-10 }
	followed by some extra spaces (because the header is always 4096 characters long). After the string <code>ML_FF</code> the header contains the file version number. {{VASP}} machine-learned force field files are versioned since {{VASP}} 6.4.0 to ~~allow for~~ compatibility ~~checks~~. Typically, {{FILE\|ML_FFN}} files are backward compatible, i.e., newer versions of {{VASP}} will be able to read files generated by older {{VASP}} versions. The opposite is generally not true. However, the {{FILE\|ML_FFN}} version number does not automatically increase with the {{VASP}} version, so two or more consecutive {{VASP}} versions could create {{FILE\|ML_FFN}} files with identical version number. The version number in the header is followed by the word <code>binary</code>. The remainder of the header contains a [https://www.json.org JSON] string with general information about the force field and values of selected {{FILE\|INCAR}} tags. For example, the timestamp following the <code>"date"</code> key is also written to the <code>FFOUT</code> lines in {{FILE\|ML_LOGFILE}} which helps to identify later on which force field files belong to which log files. Another important piece of information is <code>"ML_LFAST"</code> key which allows you to check whether this force field is ready for fast prediction mode.		followed by some extra spaces (because the header is always 4096 characters long). After the string <code>ML_FF</code> the header contains the file version number. {{VASP}} machine-learned force field files are versioned since {{VASP}} 6.4.0. This is necessary because whenever a new feature related to machine-learned force fields is available the contents of the generated files could change. Hence, when reading an {{FILE\|ML_FF}} file a {{VASP}} executable deduces the possible features from the file's version number, i.e., performs a compatibility check. Typically, {{FILE\|ML_FFN}} files are backward compatible, i.e., newer versions of {{VASP}} will be able to read files generated by older {{VASP}} versions. The opposite is generally not true. However, the {{FILE\|ML_FFN}} version number does not automatically increase with the {{VASP}} version, so two or more consecutive {{VASP}} versions could create {{FILE\|ML_FFN}} files with identical version number. The version number in the header is followed by the word <code>binary</code>. The remainder of the header contains a [https://www.json.org JSON] string with general information about the force field and values of selected {{FILE\|INCAR}} tags. For example, the timestamp following the <code>"date"</code> key is also written to the <code>FFOUT</code> lines in {{FILE\|ML_LOGFILE}} which helps to identify later on which force field files belong to which log files. Another important piece of information is <code>"ML_LFAST"</code> key which allows you to check whether this force field is ready for fast prediction mode.

			<!--
			== Fast prediction mode ==

			As outlined in the [[Machine_learning_force_field_calculations:_Basics#Step-by-step_instructions\|step-by-step instructions]] -->


	----		----

	[[Category:Files]][[Category:Machine-learned force fields]][[Category:Output files]]		[[Category:Files]][[Category:Machine-learned force fields]][[Category:Output files]]

Singraber at 09:18, 15 October 2025

2025-10-15T09:18:18Z

← Older revision		Revision as of 09:18, 15 October 2025
Line 1:		Line 1:
	{{DISPLAYTITLE:ML_FFN}}		{{DISPLAYTITLE:ML_FFN}}
	This binary file contains a ~~knewly~~ created force field from machine learning runs with the options {{TAG\|ML_MODE}}=<code>train</code>, <code>refit</code> or <code>select</code>. It's structure is identical to the {{TAG\|ML_FF}} file. To be able to use the new force field ~~from~~ {{FILE\|ML_FFN}} file it has to be simply copied to {{FILE\|ML_FF}} and the {{FILE\|INCAR}} tag {{TAG\|ML_MODE}}=<code>run</code> has to be set.		This binary file contains a newly created force field from machine learning runs with the options {{TAG\|ML_MODE}}=<code>train</code>, <code>refit</code> or <code>select</code>. It's structure is identical to the {{TAG\|ML_FF}} file. To be able to use the new force field in the {{FILE\|ML_FFN}} file it has to be simply copied to {{FILE\|ML_FF}} and the {{FILE\|INCAR}} tag {{TAG\|ML_MODE}}=<code>run</code> has to be set (see [[Machine_learning_force_field_calculations:_Basics#Step-by-step_instructions\|here]] for the basic workflow).

	Since VASP 6.4.0 the {{FILE\|ML_FFN}} file starts with an ASCII header containing the most important INCAR tags in effect during generation of this force field. In Linux shells this can be easily extracted issuing the following command:		Since VASP 6.4.0 the {{FILE\|ML_FFN}} file starts with an ASCII header containing the most important INCAR tags in effect during generation of this force field. In Linux shells this can be easily extracted issuing the following command:
Line 6:		Line 6:
	The output may look like this:		The output may look like this:
	ML_FF 0.2.1 binary { "date" : "2023-03-16T13:49:44.829", "ML_LFAST" : False, "ML_DESC_TYPE" : 0, "types" : [ "Si" ], "training_structures" : 984, "local_reference_cfgs" : [ 110 ], "descriptors" : [ 142 ], "ML_IALGO_LINREG" : 3, "ML_RCUT1" : 6.0000E+00, "ML_RCUT2" : 6.0000E+00, "ML_W1" : 1.0000E-01, "ML_SION1" : 5.0000E-01, "ML_SION2" : 5.0000E-01, "ML_LMAX2" : 4, "ML_MRB1" : 8, "ML_MRB2" : 8, "ML_IWEIGHT" : 3, "ML_WTOTEN" : 1.0000E+00, "ML_WTIFOR" : 1.0000E+00, "ML_WTSIF" : 1.0000E-10 }		ML_FF 0.2.1 binary { "date" : "2023-03-16T13:49:44.829", "ML_LFAST" : False, "ML_DESC_TYPE" : 0, "types" : [ "Si" ], "training_structures" : 984, "local_reference_cfgs" : [ 110 ], "descriptors" : [ 142 ], "ML_IALGO_LINREG" : 3, "ML_RCUT1" : 6.0000E+00, "ML_RCUT2" : 6.0000E+00, "ML_W1" : 1.0000E-01, "ML_SION1" : 5.0000E-01, "ML_SION2" : 5.0000E-01, "ML_LMAX2" : 4, "ML_MRB1" : 8, "ML_MRB2" : 8, "ML_IWEIGHT" : 3, "ML_WTOTEN" : 1.0000E+00, "ML_WTIFOR" : 1.0000E+00, "ML_WTSIF" : 1.0000E-10 }
	followed by some extra spaces (because the header is always 4096 characters long). The timestamp following <code>"date"</code> is also written to the <code>FFOUT</code> lines in {{FILE\|ML_LOGFILE}}. ~~The~~ <code>"ML_LFAST"</code> ~~item~~ allows you to check whether this force field is ready for fast prediction mode.		followed by some extra spaces (because the header is always 4096 characters long). After the string <code>ML_FF</code> the header contains the file version number. {{VASP}} machine-learned force field files are versioned since {{VASP}} 6.4.0 to allow for compatibility checks. Typically, {{FILE\|ML_FFN}} files are backward compatible, i.e., newer versions of {{VASP}} will be able to read files generated by older {{VASP}} versions. The opposite is generally not true. However, the {{FILE\|ML_FFN}} version number does not automatically increase with the {{VASP}} version, so two or more consecutive {{VASP}} versions could create {{FILE\|ML_FFN}} files with identical version number. The version number in the header is followed by the word <code>binary</code>. The remainder of the header contains a [https://www.json.org JSON] string with general information about the force field and values of selected {{FILE\|INCAR}} tags. For example, the timestamp following the <code>"date"</code> key is also written to the <code>FFOUT</code> lines in {{FILE\|ML_LOGFILE}} which helps to identify later on which force field files belong to which log files. Another important piece of information is <code>"ML_LFAST"</code> key which allows you to check whether this force field is ready for fast prediction mode.
	----		----

	[[Category:Files]][[Category:Machine-learned force fields]][[Category:Output files]]		[[Category:Files]][[Category:Machine-learned force fields]][[Category:Output files]]

Schlipf at 13:36, 7 March 2025

2025-03-07T13:36:28Z

← Older revision		Revision as of 13:36, 7 March 2025
Line 9:		Line 9:
	----		----

	[[Category:Files]][[Category:Machine-learned force fields]][[Category:Output ~~Files~~]]		[[Category:Files]][[Category:Machine-learned force fields]][[Category:Output files]]

Singraber at 13:38, 5 May 2023

2023-05-05T13:38:58Z

← Older revision		Revision as of 13:38, 5 May 2023
Line 1:		Line 1:
	{{DISPLAYTITLE:ML_FFN}}		{{DISPLAYTITLE:ML_FFN}}
	This file contains a knewly created force field from machine learning runs with the options {{TAG\|~~ML_ISTART~~}}=0 or ~~{{TAG\|ML_ISTART}}=1~~. It's structure is identical to the {{TAG\|ML_FF}} file. To be able to use the new force field from {{~~TAG~~\|ML_FFN}} file it has to be simply copied to {{~~TAG~~\|ML_FF}} and the {{~~TAG~~\|INCAR}} tag {{TAG\|~~ML_ISTART~~}}=2 has to be set.		This binary file contains a knewly created force field from machine learning runs with the options {{TAG\|ML_MODE}}=<code>train</code>, <code>refit</code> or <code>select</code>. It's structure is identical to the {{TAG\|ML_FF}} file. To be able to use the new force field from {{FILE\|ML_FFN}} file it has to be simply copied to {{FILE\|ML_FF}} and the {{FILE\|INCAR}} tag {{TAG\|ML_MODE}}=<code>run</code> has to be set.

			Since VASP 6.4.0 the {{FILE\|ML_FFN}} file starts with an ASCII header containing the most important INCAR tags in effect during generation of this force field. In Linux shells this can be easily extracted issuing the following command:
			head -n 1 ML_FFN
			The output may look like this:
			ML_FF 0.2.1 binary { "date" : "2023-03-16T13:49:44.829", "ML_LFAST" : False, "ML_DESC_TYPE" : 0, "types" : [ "Si" ], "training_structures" : 984, "local_reference_cfgs" : [ 110 ], "descriptors" : [ 142 ], "ML_IALGO_LINREG" : 3, "ML_RCUT1" : 6.0000E+00, "ML_RCUT2" : 6.0000E+00, "ML_W1" : 1.0000E-01, "ML_SION1" : 5.0000E-01, "ML_SION2" : 5.0000E-01, "ML_LMAX2" : 4, "ML_MRB1" : 8, "ML_MRB2" : 8, "ML_IWEIGHT" : 3, "ML_WTOTEN" : 1.0000E+00, "ML_WTIFOR" : 1.0000E+00, "ML_WTSIF" : 1.0000E-10 }
			followed by some extra spaces (because the header is always 4096 characters long). The timestamp following <code>"date"</code> is also written to the <code>FFOUT</code> lines in {{FILE\|ML_LOGFILE}}. The <code>"ML_LFAST"</code> item allows you to check whether this force field is ready for fast prediction mode.
	----		----

	[[Category:Files]][[Category:Machine-learned force fields]][[Category:Output Files]]		[[Category:Files]][[Category:Machine-learned force fields]][[Category:Output Files]]