From 1f1f4790875001194ace850aa9281fb2b5414d8b Mon Sep 17 00:00:00 2001 From: "Peter G. Mikhael" Date: Mon, 8 Apr 2024 12:36:53 -0400 Subject: [PATCH] update paths --- configs/screening.json | 51 +++++--------------------- configs/train/clip_cgr.json | 4 +- configs/train/clip_cgr_rxn_string.json | 4 +- configs/train/clip_egnn.json | 4 +- configs/train/clip_esm.json | 4 +- configs/train/clip_esm_frozen.json | 4 +- configs/train/clip_msa.json | 4 +- configs/train/clip_wldn.json | 4 +- 8 files changed, 24 insertions(+), 55 deletions(-) diff --git a/configs/screening.json b/configs/screening.json index 34c5822..9da003e 100644 --- a/configs/screening.json +++ b/configs/screening.json @@ -1,50 +1,19 @@ { "available_gpus": ["4,5,6,7"], - "script": "main", + "script": "screen", "cartesian_hyperparams": { - "dataset_file_path": [], - "inference_dir": [], - "save_hiddens": [], - "save_predictions": [], - "use_as_protein_encoder": [], - "use_as_reaction_encoder": [], - "protein_cache_dir": [], - "gpus": [], + "dataset_file_path": ["files/new_data.csv"], + "inference_dir": [""], + "save_hiddens": [false], + "save_predictions": [false], + "use_as_protein_encoder": [false], + "use_as_reaction_encoder": [false], + "gpus": [1], "batch_size": [8], "precision": ["bf16"], - "dataset_name": ["reactions_dataset"], - "max_protein_length": [650], - "eval_on_train_multigpu": [true], - "lightning_name": ["base"], - "model_name": ["enzyme_reaction_clip"], - "from_checkpoint": [true], "checkpoint_path": ["files/clipzyme_model.ckpt"], - "chemprop_node_dim": [9], - "chemprop_edge_dim": [3], - "protein_encoder": ["egnn_sparse_network"], - "use_protein_graphs": [true], - "protein_dim": [1280], - "neighbour_aggr": ["mean"], - "pool_type": ["sum"], - "message_dim": [24], - "norm_feats": [true], - "norm_coors": [true], - "update_feats": [true], - "update_coors": [true], - "egcl_layers": [6], - "use_sinusoidal": [true], - "chemprop_num_layers": [5], - "chemprop_hidden_dim": [1280], - "chemprop_pool": ["none"], - "dropout": [0], - "num_workers": [24], - "train": [false], - "dev": [false], - "test": [false], - "num_sanity_val_steps": [0], - "val_check_interval": [1.0], - "logger_name": ["tensorboard"], - "logger_tags": ["screen"] + "esm_dir": ["/data/esm/checkpoints"], + "num_workers": [24] } } \ No newline at end of file diff --git a/configs/train/clip_cgr.json b/configs/train/clip_cgr.json index ee7c48f..2715775 100644 --- a/configs/train/clip_cgr.json +++ b/configs/train/clip_cgr.json @@ -5,8 +5,8 @@ "precision": ["bf16"], "accumulate_grad_batches": [1], "dataset_name": ["enzymemap_reaction_graph"], - "dataset_file_path": ["/home/datasets/EnzymeMap/version2/enzymemap_v2_brenda2_blip.json"], - "dataset_cache_path": ["/home/cache/cached_egnn_blip_dataset_version2.p"], + "dataset_file_path": ["files/enzymemap.json"], + "dataset_cache_path": ["files/cached_enzymemap.p"], "remove_duplicate_reactions": [true], "use_pesto_scores": [true], "max_protein_length": [650], diff --git a/configs/train/clip_cgr_rxn_string.json b/configs/train/clip_cgr_rxn_string.json index 3a87b4e..ab07308 100644 --- a/configs/train/clip_cgr_rxn_string.json +++ b/configs/train/clip_cgr_rxn_string.json @@ -5,8 +5,8 @@ "precision": ["bf16"], "accumulate_grad_batches": [1], "dataset_name": ["enzymemap_reaction_graph"], - "dataset_file_path": ["/home/datasets/EnzymeMap/version2/enzymemap_v2_brenda2_blip.json"], - "dataset_cache_path": ["/home/cache/cached_egnn_blip_dataset_version2.p"], + "dataset_file_path": ["files/enzymemap.json"], + "dataset_cache_path": ["files/cached_enzymemap.p"], "remove_duplicate_reactions": [true], "use_pesto_scores": [false], "max_protein_length": [650], diff --git a/configs/train/clip_egnn.json b/configs/train/clip_egnn.json index 882fa7b..662bd56 100644 --- a/configs/train/clip_egnn.json +++ b/configs/train/clip_egnn.json @@ -5,8 +5,8 @@ "precision": ["bf16"], "accumulate_grad_batches": [1], "dataset_name": ["enzymemap_reaction_graph"], - "dataset_file_path": ["/home/datasets/EnzymeMap/version2/enzymemap_v2_brenda2_blip.json"], - "dataset_cache_path": ["/home/cache/cached_egnn_blip_dataset_version2.p"], + "dataset_file_path": ["files/enzymemap.json"], + "dataset_cache_path": ["files/cached_enzymemap.p"], "remove_duplicate_reactions": [true], "use_pesto_scores": [false], "max_protein_length": [650], diff --git a/configs/train/clip_esm.json b/configs/train/clip_esm.json index 2df5764..930d839 100644 --- a/configs/train/clip_esm.json +++ b/configs/train/clip_esm.json @@ -5,8 +5,8 @@ "precision": ["bf16"], "accumulate_grad_batches": [2], "dataset_name": ["enzymemap_reaction_graph"], - "dataset_file_path": ["/home/datasets/EnzymeMap/version2/enzymemap_v2_brenda2_blip.json"], - "dataset_cache_path": ["/home/cache/cached_egnn_blip_dataset_version2.p"], + "dataset_file_path": ["files/enzymemap.json"], + "dataset_cache_path": ["files/cached_enzymemap.p"], "from_checkpoint": [true], "checkpoint_path": ["/home/snapshots/metabolomics/6a518ff6eda2b4a4299e9e1523b24d27/6a518ff6eda2b4a4299e9e1523b24d27epoch=15.ckpt"], "remove_duplicate_reactions": [true], diff --git a/configs/train/clip_esm_frozen.json b/configs/train/clip_esm_frozen.json index 1c5476e..030c999 100644 --- a/configs/train/clip_esm_frozen.json +++ b/configs/train/clip_esm_frozen.json @@ -5,8 +5,8 @@ "precision": ["bf16"], "accumulate_grad_batches": [1], "dataset_name": ["enzymemap_reaction_graph"], - "dataset_file_path": ["/home/datasets/Enzymes/EnzymeMap/version2/enzymemap_v2_brenda2_blip.json"], - "dataset_cache_path": ["/home/cache/cached_egnn_blip_dataset_version2.p"], + "dataset_file_path": ["files/enzymemap.json"], + "dataset_cache_path": ["files/cached_enzymemap.p"], "remove_duplicate_reactions": [true], "use_pesto_scores": [false], "max_protein_length": [650], diff --git a/configs/train/clip_msa.json b/configs/train/clip_msa.json index 3c9726d..bd66cc0 100644 --- a/configs/train/clip_msa.json +++ b/configs/train/clip_msa.json @@ -5,8 +5,8 @@ "precision": ["bf16"], "accumulate_grad_batches": [2], "dataset_name": ["enzymemap_reaction_graph"], - "dataset_file_path": ["/home/datasets/EnzymeMap/version2/enzymemap_v2_brenda2_blip.json"], - "dataset_cache_path": ["/home/cache/cached_egnn_blip_dataset_version2.p"], + "dataset_file_path": ["files/enzymemap.json"], + "dataset_cache_path": ["files/cached_enzymemap.p"], "remove_duplicate_reactions": [true], "use_pesto_scores": [false], "max_protein_length": [650], diff --git a/configs/train/clip_wldn.json b/configs/train/clip_wldn.json index d950072..f06e45d 100644 --- a/configs/train/clip_wldn.json +++ b/configs/train/clip_wldn.json @@ -5,8 +5,8 @@ "precision": ["bf16"], "accumulate_grad_batches": [1], "dataset_name": ["enzymemap_reaction_graph"], - "dataset_file_path": ["/home/datasets/EnzymeMap/version2/enzymemap_v2_brenda2_blip.json"], - "dataset_cache_path": ["/home/cache/cached_egnn_blip_dataset_version2.p"], + "dataset_file_path": ["files/enzymemap.json"], + "dataset_cache_path": ["files/cached_enzymemap.p"], "remove_duplicate_reactions": [true], "use_pesto_scores": [false], "max_protein_length": [650],