dpgen-simplify

Name: dpgen-simplify
Author: jinzhezenggroup

113stars23forks

How to install dpgen-simplify?

jinzhezenggroup/computational-chemistry-agent-skills/dpgen-simplify

$npx -y skills add jinzhezenggroup/computational-chemistry-agent-skills --skill dpgen-simplify

Installs into the current project.

Summary

Prepare, explain, validate, and run DP-GEN simplify workflows for reducing repeated or redundant DeepMD datasets. Use when the user wants to generate or modify param.json and machine.json, run dpgen simplify param.json machine.json, organize repeated simplify experiments, o

Files · 1

View on GitHub

SKILL.md

1	# DP-GEN Simplify
2
3	Use this skill when the user wants to prepare, explain, validate, or execute the `dpgen simplify` workflow.
4
5	This skill is for dataset simplification workflows where the user already has candidate data in DeepMD-compatible format and wants to reduce repeated or redundant structures through iterative selection.
6
7	## Core Rule (Critical)
8
9	DP-GEN simplify always uses two parameter classes and therefore two JSON files:
10
11	- Workflow parameters -> `param.json`
12	- Execution / machine parameters -> `machine.json`
13
14	Run exactly:
15
16	```bash
17	dpgen simplify param.json machine.json
18	```
19
20	Environment boundary rule:
21
22	- Outer layer: run `dpgen simplify param.json machine.json` in an activated environment where `dpgen --version` works.
23	- Inner layer: for scheduler stages, explicitly activate runtime in `resources.source_list` on the server side.
24
25	## Agent responsibilities
26
27	When using this skill, the agent should:
28
29	1. confirm that the task is a simplify workflow
30	1. check whether existing configs or templates are already available
31	1. collect only the missing dataset, training, FP, and machine inputs
32	1. generate or patch `param.json`
33	1. generate or patch `machine.json`
34	1. explain important simplify parameters in plain language when asked
35	1. validate the workflow before execution
36	1. provide the exact command for running simplify
37	1. if requested, help structure repeated experiments
38	1. after execution, summarize outputs and next inspection targets
39
40	## Working policy
41
42	### 1. Ask only for missing inputs
43
44	Do not ask the user for everything if part of the configuration is already available.
45
46	If the user already provides:
47
48	- a partial `param.json`
49	- a partial `machine.json`
50	- a known training template
51	- a known cluster template
52
53	then patch those files instead of rebuilding everything from scratch.
54
55	### 2. Preserve the user's scientific choices
56
57	Do not silently change:
58
59	- descriptor family
60	- fitting net structure
61	- fp backend
62	- trust thresholds
63	- `type_map` ordering
64
65	If a value looks scientifically questionable, explain the concern instead of silently replacing it.
66
67	### 3. Keep local and scheduler execution explicit
68
69	If the user wants local execution, produce local-friendly commands.
70
71	If the user wants scheduler execution, produce scheduler-friendly commands and keep queue, partition, and resource requests explicit.
72
73	Do not invent scheduler module names or executable paths.
74
75	### 4. Do not invent environment activation commands
76
77	If the user already has a working activation command such as:
78
79	- `conda activate ...`
80	- `module load ...`
81	- `source ...`
82
83	reuse it exactly.
84
85	If execution is requested and the activation method is unknown, ask the user for the precise activation command.
86
87	Do not guess conda environment names, module names, or site-specific paths.
88
89	### 4.1 Outer launcher policy
90
91	Use an activated DP-GEN environment and verify with:
92
93	```bash
94	dpgen --version
95	```
96
97	Do not start simplify from a shell where `dpgen` is unavailable.
98
99	### 4.2 Outer vs inner runtime boundaries (critical)
100
101	Treat simplify execution as two separate environment layers:
102
103	1. Outer layer: the shell that launches `dpgen simplify param.json machine.json` (must have `dpgen` in PATH)
104	1. Inner layer: stage tasks dispatched by DP-GEN (`train` / `model_devi` / `fp`) on server/runtime side
105
106	Even if the outer layer is correct, inner stage tasks still need explicit runtime setup in `machine.json`.
107	Do not assume the outer shell environment will be inherited by dispatched stage jobs.
108	For scheduler-style execution, `resources.source_list` must explicitly activate the required runtime environment.
109
110	### 5. Prefer reproducible output layout
111
112	When generating a simplify workflow, keep files organized and predictable.
113
114	Recommended structure:
115
116	```text
117	project/
118	├── param.json
119	├── machine.json
120	├── run.sh
121	├── logs/
122	└── summary/
123	```
124
125	For repeated experiments:
126
127	```text
128	project/
129	├── base/
130	├── exp_01/
131	├── exp_02/
132	├── exp_03/
133	└── summary/
134	```
135
136	## Minimum required inputs
137
138	Collect the following information before generating files.
139
140	### Dataset information
141
142	- `pick_data`
143	- `sys_configs`
144	- `init_data_prefix`
145	- `init_data_sys`
146	- `sys_bat

People who liked this also viewed

6 picks

Type

dpgen-simplify

113stars23forks

View on GitHub

How to install dpgen-simplify?

jinzhezenggroup/computational-chemistry-agent-skills/dpgen-simplify

$npx -y skills add jinzhezenggroup/computational-chemistry-agent-skills --skill dpgen-simplify

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# DP-GEN Simplify
2
3	Use this skill when the user wants to prepare, explain, validate, or execute the `dpgen simplify` workflow.
4
5	This skill is for dataset simplification workflows where the user already has candidate data in DeepMD-compatible format and wants to reduce repeated or redundant structures through iterative selection.
6
7	## Core Rule (Critical)
8
9	DP-GEN simplify always uses two parameter classes and therefore two JSON files:
10
11	- Workflow parameters -> `param.json`
12	- Execution / machine parameters -> `machine.json`
13
14	Run exactly:
15
16	```bash
17	dpgen simplify param.json machine.json
18	```
19
20	Environment boundary rule:
21
22	- Outer layer: run `dpgen simplify param.json machine.json` in an activated environment where `dpgen --version` works.
23	- Inner layer: for scheduler stages, explicitly activate runtime in `resources.source_list` on the server side.
24
25	## Agent responsibilities
26
27	When using this skill, the agent should:
28
29	1. confirm that the task is a simplify workflow
30	1. check whether existing configs or templates are already available
31	1. collect only the missing dataset, training, FP, and machine inputs
32	1. generate or patch `param.json`
33	1. generate or patch `machine.json`
34	1. explain important simplify parameters in plain language when asked
35	1. validate the workflow before execution
36	1. provide the exact command for running simplify
37	1. if requested, help structure repeated experiments
38	1. after execution, summarize outputs and next inspection targets
39
40	## Working policy
41
42	### 1. Ask only for missing inputs
43
44	Do not ask the user for everything if part of the configuration is already available.
45
46	If the user already provides:
47
48	- a partial `param.json`
49	- a partial `machine.json`
50	- a known training template
51	- a known cluster template
52
53	then patch those files instead of rebuilding everything from scratch.
54
55	### 2. Preserve the user's scientific choices
56
57	Do not silently change:
58
59	- descriptor family
60	- fitting net structure
61	- fp backend
62	- trust thresholds
63	- `type_map` ordering
64
65	If a value looks scientifically questionable, explain the concern instead of silently replacing it.
66
67	### 3. Keep local and scheduler execution explicit
68
69	If the user wants local execution, produce local-friendly commands.
70
71	If the user wants scheduler execution, produce scheduler-friendly commands and keep queue, partition, and resource requests explicit.
72
73	Do not invent scheduler module names or executable paths.
74
75	### 4. Do not invent environment activation commands
76
77	If the user already has a working activation command such as:
78
79	- `conda activate ...`
80	- `module load ...`
81	- `source ...`
82
83	reuse it exactly.
84
85	If execution is requested and the activation method is unknown, ask the user for the precise activation command.
86
87	Do not guess conda environment names, module names, or site-specific paths.
88
89	### 4.1 Outer launcher policy
90
91	Use an activated DP-GEN environment and verify with:
92
93	```bash
94	dpgen --version
95	```
96
97	Do not start simplify from a shell where `dpgen` is unavailable.
98
99	### 4.2 Outer vs inner runtime boundaries (critical)
100
101	Treat simplify execution as two separate environment layers:
102
103	1. Outer layer: the shell that launches `dpgen simplify param.json machine.json` (must have `dpgen` in PATH)
104	1. Inner layer: stage tasks dispatched by DP-GEN (`train` / `model_devi` / `fp`) on server/runtime side
105
106	Even if the outer layer is correct, inner stage tasks still need explicit runtime setup in `machine.json`.
107	Do not assume the outer shell environment will be inherited by dispatched stage jobs.
108	For scheduler-style execution, `resources.source_list` must explicitly activate the required runtime environment.
109
110	### 5. Prefer reproducible output layout
111
112	When generating a simplify workflow, keep files organized and predictable.
113
114	Recommended structure:
115
116	```text
117	project/
118	├── param.json
119	├── machine.json
120	├── run.sh
121	├── logs/
122	└── summary/
123	```
124
125	For repeated experiments:
126
127	```text
128	project/
129	├── base/
130	├── exp_01/
131	├── exp_02/
132	├── exp_03/
133	└── summary/
134	```
135
136	## Minimum required inputs
137
138	Collect the following information before generating files.
139
140	### Dataset information
141
142	- `pick_data`
143	- `sys_configs`
144	- `init_data_prefix`
145	- `init_data_sys`
146	- `sys_bat

People who liked this also viewed

6 picks

Type