Merge pull request #44 from Acellera/simplify_molscore_conf

Simplify molscore conf

scripts/a2c/a2c.py

@@ -1,9 +1,7 @@
 #! /usr/bin/python3
 import datetime
-import json
 import os
 import random
-import shutil
 from pathlib import Path
 
 import hydra
@@ -88,16 +86,12 @@ def main(cfg: "DictConfig"):
                 ) from MOLSCORE_ERR
 
             if cfg.molscore_mode == "single":
-                # Save molscore output. Also redirect output to save_dir
-                cfg.molscore_task = shutil.copy(cfg.molscore_task, save_dir)
-                data = json.load(open(cfg.molscore_task, "r"))
-                json.dump(data, open(cfg.molscore_task, "w"), indent=4)
                 task = MolScore(
                     model_name=cfg.agent_name,
                     task_config=cfg.molscore_task,
                     budget=cfg.total_smiles,
                     output_dir=os.path.abspath(save_dir),
-                    add_run_dir=False,
+                    add_run_dir=True,
                     **cfg.get("molscore_kwargs", {}),
                 )
                 run_a2c(cfg, task)

scripts/a2c/config_denovo.yaml

@@ -10,11 +10,9 @@ num_envs: 16 # Number of smiles to generate in parallel
 total_smiles: 10_000  # Total number of smiles to generate
 
 # Scoring function
-molscore_mode: benchmark # single, benchmark, or curriculum
-molscore_task: MolOpt # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
-molscore_kwargs:
-  include: ["Albuterol_similarity"]
-custom_task: null # Requires molscore to be set to null
+molscore_mode: single # single, benchmark, or curriculum
+molscore_task: MolOpt:Albuterol_similarity # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
+custom_task: null # Requires custom_task mode to be set to null
 
 # Promptsmiles configuration
 prompt: null  # e.g. c1ccccc  # Fix the beginning of the generated molecules

scripts/a2c/config_fragment.yaml

@@ -10,11 +10,9 @@ num_envs: 16 # Number of smiles to generate in parallel
 total_smiles: 10_000  # Total number of smiles to generate
 
 # Scoring function
-molscore_mode: benchmark # single, benchmark, or curriculum
-molscore_task: MolOpt # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
-molscore_kwargs:
-  include: ["Celecoxxib_rediscovery"]
-custom_task: null # Requires molscore to be set to null
+molscore_mode: single # single, benchmark, or curriculum
+molscore_task: MolOpt:Celecoxxib_rediscovery # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
+custom_task: null # Requires molscore_task to be set to null
 
 # Promptsmiles configuration
 promptsmiles: c1(C)ccc(*)cc1.NS(=O)(=O)(*)

scripts/a2c/config_scaffold.yaml

@@ -10,11 +10,9 @@ num_envs: 16 # Number of smiles to generate in parallel
 total_smiles: 10_000  # Total number of smiles to generate
 
 # Scoring function
-molscore_mode: benchmark # single, benchmark, or curriculum
-molscore_task: LibINVENT_Exp1 # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
-molscore_kwargs:
-  include: ["DRD2_SelRF_SubFilt_DF"]
-custom_task: null # Requires molscore to be set to null
+molscore_mode: single # single, benchmark, or curriculum
+molscore_task: LibINVENT_Exp1:DRD2_SelRF_SubFilt_DF # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
+custom_task: null # Requires molscore_task to be set to null
 
 # Promptsmiles configuration
 promptsmiles: N1(*)CCN(CC1)CCCCN(*)

scripts/ahc/ahc.py

@@ -1,9 +1,7 @@
 #! /usr/bin/python3
 import datetime
-import json
 import os
 import random
-import shutil
 from copy import deepcopy
 from pathlib import Path
 
@@ -90,16 +88,12 @@ def main(cfg: "DictConfig"):
                 ) from MOLSCORE_ERR
 
             if cfg.molscore_mode == "single":
-                # Save molscore output. Also redirect output to save_dir
-                cfg.molscore_task = shutil.copy(cfg.molscore_task, save_dir)
-                data = json.load(open(cfg.molscore_task, "r"))
-                json.dump(data, open(cfg.molscore_task, "w"), indent=4)
                 task = MolScore(
                     model_name=cfg.agent_name,
                     task_config=cfg.molscore_task,
                     budget=cfg.total_smiles,
                     output_dir=os.path.abspath(save_dir),
-                    add_run_dir=False,
+                    add_run_dir=True,
                     **cfg.get("molscore_kwargs", {}),
                 )
                 run_ahc(cfg, task)

scripts/ahc/config_denovo.yaml

@@ -10,11 +10,9 @@ num_envs: 128 # Number of smiles to generate in parallel
 total_smiles: 10_000  # Total number of smiles to generate
 
 # Scoring function
-molscore_mode: benchmark # single, benchmark, or curriculum
-molscore_task: MolOpt # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
-molscore_kwargs:
-  include: ["Albuterol_similarity"]
-custom_task: null # Requires molscore to be set to null
+molscore_mode: single # single, benchmark, or curriculum
+molscore_task: MolOpt:Albuterol_similarity # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
+custom_task: null # Requires custom_task mode to be set to null
 
 # Promptsmiles configuration
 prompt: null  # e.g. c1ccccc  # Fix the beginning of the generated molecules

scripts/ahc/config_fragment.yaml

@@ -10,11 +10,9 @@ num_envs: 128 # Number of smiles to generate in parallel
 total_smiles: 10_000  # Total number of smiles to generate
 
 # Scoring function
-molscore_mode: benchmark # single, benchmark, or curriculum
-molscore_task: MolOpt # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
-molscore_kwargs:
-  include: ["Celecoxxib_rediscovery"]
-custom_task: null # Requires molscore to be set to null
+molscore_mode: single # single, benchmark, or curriculum
+molscore_task: MolOpt:Celecoxxib_rediscovery # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
+custom_task: null # Requires molscore_task to be set to null
 
 # Promptsmiles configuration
 promptsmiles: c1(C)ccc(*)cc1.NS(=O)(=O)(*)

scripts/ahc/config_scaffold.yaml

@@ -10,11 +10,9 @@ num_envs: 128 # Number of smiles to generate in parallel
 total_smiles: 10_000  # Total number of smiles to generate
 
 # Scoring function
-molscore_mode: benchmark # single, benchmark, or curriculum
-molscore_task: LibINVENT_Exp1 # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
-molscore_kwargs:
-  include: ["DRD2_SelRF_SubFilt_DF"]
-custom_task: null # Requires molscore to be set to null
+molscore_mode: single # single, benchmark, or curriculum
+molscore_task: LibINVENT_Exp1:DRD2_SelRF_SubFilt_DF # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
+custom_task: null # Requires molscore_task to be set to null
 
 # Promptsmiles configuration
 promptsmiles: N1(*)CCN(CC1)CCCCN(*)

scripts/dpo/config_denovo.yaml

@@ -10,11 +10,9 @@ num_envs: 128 # Number of smiles to generate in parallel
 total_smiles: 10_000 # Total number of smiles to generate
 
 # Scoring function
-molscore_mode: benchmark # single, benchmark, or curriculum
-molscore_task: MolOpt # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
-molscore_kwargs:
-  include: ["Albuterol_similarity"]
-custom_task: null # Requires molscore to be set to null
+molscore_mode: single # single, benchmark, or curriculum
+molscore_task: MolOpt:Albuterol_similarity # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
+custom_task: null # Requires custom_task mode to be set to null
 
 # Promptsmiles configuration
 prompt: null  # e.g. c1ccccc  # Fix the beginning of the generated molecules

scripts/dpo/config_fragment.yaml

@@ -10,11 +10,9 @@ num_envs: 128 # Number of smiles to generate in parallel
 total_smiles: 10_000  # Total number of smiles to generate
 
 # Scoring function
-molscore_mode: benchmark # single, benchmark, or curriculum
-molscore_task: MolOpt # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
-molscore_kwargs:
-  include: ["Celecoxxib_rediscovery"]
-custom_task: null # Requires molscore to be set to null
+molscore_mode: single # single, benchmark, or curriculum
+molscore_task: MolOpt:Celecoxxib_rediscovery # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
+custom_task: null # Requires molscore_task to be set to null
 
 # Promptsmiles configuration
 promptsmiles: c1(C)ccc(*)cc1.NS(=O)(=O)(*)

scripts/dpo/config_scaffold.yaml

@@ -10,11 +10,9 @@ num_envs: 128 # Number of smiles to generate in parallel
 total_smiles: 10_000  # Total number of smiles to generate
 
 # Scoring function
-molscore_mode: benchmark # single, benchmark, or curriculum
-molscore_task: LibINVENT_Exp1 # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
-molscore_kwargs:
-  include: ["DRD2_SelRF_SubFilt_DF"]
-custom_task: null # Requires molscore to be set to null
+molscore_mode: single # single, benchmark, or curriculum
+molscore_task: LibINVENT_Exp1:DRD2_SelRF_SubFilt_DF # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
+custom_task: null # Requires molscore_task to be set to null
 
 # Promptsmiles configuration
 promptsmiles: N1(*)CCN(CC1)CCCCN(*)

scripts/dpo/dpo.py

@@ -1,9 +1,7 @@
 #! /usr/bin/python3
 import datetime
-import json
 import os
 import random
-import shutil
 from copy import deepcopy
 from pathlib import Path
 
@@ -83,16 +81,12 @@ def main(cfg: "DictConfig"):
                 ) from MOLSCORE_ERR
 
             if cfg.molscore_mode == "single":
-                # Save molscore output. Also redirect output to save_dir
-                cfg.molscore_task = shutil.copy(cfg.molscore_task, save_dir)
-                data = json.load(open(cfg.molscore_task, "r"))
-                json.dump(data, open(cfg.molscore_task, "w"), indent=4)
                 task = MolScore(
                     model_name=cfg.agent_name,
                     task_config=cfg.molscore_task,
                     budget=cfg.total_smiles,
                     output_dir=os.path.abspath(save_dir),
-                    add_run_dir=False,
+                    add_run_dir=True,
                     **cfg.get("molscore_kwargs", {}),
                 )
                 run_dpo(cfg, task)

scripts/hill_climb/config_denovo.yaml

@@ -10,11 +10,9 @@ num_envs: 1024 # Number of smiles to generate in parallel
 total_smiles: 10_000 # Total number of smiles to generate
 
 # Scoring function
-molscore_mode: benchmark # single, benchmark, or curriculum
-molscore_task: MolOpt # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
-molscore_kwargs:
-  include: ["Albuterol_similarity"]
-custom_task: null # Requires molscore to be set to null
+molscore_mode: single # single, benchmark, or curriculum
+molscore_task: MolOpt:Albuterol_similarity # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
+custom_task: null # Requires custom_task mode to be set to null
 
 # Promptsmiles configuration
 prompt: null  # e.g. c1ccccc  # Fix the beginning of the generated molecules

scripts/hill_climb/config_fragment.yaml

@@ -10,11 +10,9 @@ num_envs: 128 # Number of smiles to generate in parallel
 total_smiles: 10_000  # Total number of smiles to generate
 
 # Scoring function
-molscore_mode: benchmark # single, benchmark, or curriculum
-molscore_task: MolOpt # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
-molscore_kwargs:
-  include: ["Celecoxxib_rediscovery"]
-custom_task: null # Requires molscore to be set to null
+molscore_mode: single # single, benchmark, or curriculum
+molscore_task: MolOpt:Celecoxxib_rediscovery # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
+custom_task: null # Requires molscore_task to be set to null
 
 # Promptsmiles configuration
 promptsmiles: c1(C)ccc(*)cc1.NS(=O)(=O)(*)

scripts/hill_climb/config_scaffold.yaml

@@ -10,11 +10,9 @@ num_envs: 128 # Number of smiles to generate in parallel
 total_smiles: 10_000  # Total number of smiles to generate
 
 # Scoring function
-molscore_mode: benchmark # single, benchmark, or curriculum
-molscore_task: LibINVENT_Exp1 # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
-molscore_kwargs:
-  include: ["DRD2_SelRF_SubFilt_DF"]
-custom_task: null # Requires molscore to be set to null
+molscore_mode: single # single, benchmark, or curriculum
+molscore_task: LibINVENT_Exp1:DRD2_SelRF_SubFilt_DF # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
+custom_task: null # Requires molscore_task to be set to null
 
 # Promptsmiles configuration
 promptsmiles: N1(*)CCN(CC1)CCCCN(*)

scripts/hill_climb/hill_climb.py

@@ -1,9 +1,7 @@
 #! /usr/bin/python3
 import datetime
-import json
 import os
 import random
-import shutil
 from copy import deepcopy
 from pathlib import Path
 
@@ -91,16 +89,12 @@ def main(cfg: "DictConfig"):
                 ) from MOLSCORE_ERR
 
             if cfg.molscore_mode == "single":
-                # Save molscore output. Also redirect output to save_dir
-                cfg.molscore_task = shutil.copy(cfg.molscore_task, save_dir)
-                data = json.load(open(cfg.molscore_task, "r"))
-                json.dump(data, open(cfg.molscore_task, "w"), indent=4)
                 task = MolScore(
                     model_name=cfg.agent_name,
                     task_config=cfg.molscore_task,
                     budget=cfg.total_smiles,
                     output_dir=os.path.abspath(save_dir),
-                    add_run_dir=False,
+                    add_run_dir=True,
                     **cfg.get("molscore_kwargs", {}),
                 )
                 run_hill_climb(cfg, task)

scripts/ppo/config_denovo.yaml

@@ -10,11 +10,9 @@ num_envs: 64 # Number of smiles to generate in parallel
 total_smiles: 10_000  # Total number of smiles to generate
 
 # Scoring function
-molscore_mode: benchmark # single, benchmark, or curriculum
-molscore_task: MolOpt # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
-molscore_kwargs:
-  include: ["Albuterol_similarity"]
-custom_task: null # Requires molscore to be set to null
+molscore_mode: single # single, benchmark, or curriculum
+molscore_task: MolOpt:Albuterol_similarity # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
+custom_task: null # Requires custom_task mode to be set to null
 
 # Promptsmiles configuration
 prompt: null  # e.g. c1ccccc  # Fix the beginning of the generated molecules

scripts/ppo/config_fragment.yaml

@@ -10,11 +10,9 @@ num_envs: 64 # Number of smiles to generate in parallel
 total_smiles: 10_000  # Total number of smiles to generate
 
 # Scoring function
-molscore_mode: benchmark # single, benchmark, or curriculum
-molscore_task: MolOpt # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
-molscore_kwargs:
-  include: ["Celecoxxib_rediscovery"]
-custom_task: null # Requires molscore to be set to null
+molscore_mode: single # single, benchmark, or curriculum
+molscore_task: MolOpt:Celecoxxib_rediscovery # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
+custom_task: null # Requires molscore_task to be set to null
 
 # Promptsmiles configuration
 promptsmiles: c1(C)ccc(*)cc1.NS(=O)(=O)(*)

scripts/ppo/config_scaffold.yaml

@@ -10,11 +10,9 @@ num_envs: 64 # Number of smiles to generate in parallel
 total_smiles: 10_000  # Total number of smiles to generate
 
 # Scoring function
-molscore_mode: benchmark # single, benchmark, or curriculum
-molscore_task: LibINVENT_Exp1 # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
-molscore_kwargs:
-  include: ["DRD2_SelRF_SubFilt_DF"]
-custom_task: null # Requires molscore to be set to null
+molscore_mode: single # single, benchmark, or curriculum
+molscore_task: LibINVENT_Exp1:DRD2_SelRF_SubFilt_DF # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
+custom_task: null # Requires molscore_task to be set to null
 
 # Promptsmiles configuration
 promptsmiles: N1(*)CCN(CC1)CCCCN(*)

scripts/ppo/ppo.py

@@ -1,9 +1,7 @@
 #! /usr/bin/python3
 import datetime
-import json
 import os
 import random
-import shutil
 from pathlib import Path
 
 import hydra
@@ -93,16 +91,12 @@ def main(cfg: "DictConfig"):
                 ) from MOLSCORE_ERR
 
             if cfg.molscore_mode == "single":
-                # Save molscore output. Also redirect output to save_dir
-                cfg.molscore_task = shutil.copy(cfg.molscore_task, save_dir)
-                data = json.load(open(cfg.molscore_task, "r"))
-                json.dump(data, open(cfg.molscore_task, "w"), indent=4)
                 task = MolScore(
                     model_name=cfg.agent_name,
                     task_config=cfg.molscore_task,
                     budget=cfg.total_smiles,
                     output_dir=os.path.abspath(save_dir),
-                    add_run_dir=False,
+                    add_run_dir=True,
                     **cfg.get("molscore_kwargs", {}),
                 )
                 run_ppo(cfg, task)

scripts/reinforce/config_denovo.yaml

@@ -10,11 +10,9 @@ num_envs: 128 # Number of smiles to generate in parallel
 total_smiles: 10_000 # Total number of smiles to generate
 
 # Scoring function
-molscore_mode: benchmark # single, benchmark, or curriculum
-molscore_task: MolOpt # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
-molscore_kwargs:
-  include: ["Albuterol_similarity"]
-custom_task: null # Requires molscore to be set to null
+molscore_mode: single # single, benchmark, or curriculum
+molscore_task: MolOpt:Albuterol_similarity # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
+custom_task: null # Requires custom_task mode to be set to null
 
 # Promptsmiles configuration
 prompt: null  # e.g. c1ccccc  # Fix the beginning of the generated molecules

scripts/reinforce/config_fragment.yaml

@@ -10,11 +10,9 @@ num_envs: 128 # Number of smiles to generate in parallel
 total_smiles: 10_000  # Total number of smiles to generate
 
 # Scoring function
-molscore_mode: benchmark # single, benchmark, or curriculum
-molscore_task: MolOpt # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
-molscore_kwargs:
-  include: ["Celecoxxib_rediscovery"]
-custom_task: null # Requires molscore to be set to null
+molscore_mode: single # single, benchmark, or curriculum
+molscore_task: MolOpt:Celecoxxib_rediscovery # task configuration (JSON), benchmark (preset only), or curriculum task (preset only)
+custom_task: null # Requires molscore_task to be set to null
 
 # Promptsmiles configuration
 promptsmiles: c1(C)ccc(*)cc1.NS(=O)(=O)(*)