Upload folder using huggingface_hub

README.md

@@ -1,3 +1,140 @@
----
-license: mit
----
+---
+license: mit
+pipeline_tag: tabular-regression
+tags:
+- chemistry
+- microbiology
+- antibiotics
+library_name: duvida
+datasets:
+- scbirlab/thomas-2018-spark-wt
+---
+
+# Predictor of _Staphylococcus aureus_ MICs 
+
+_Updated:_ Fri 28 Mar 14:27:41 GMT 2025
+
+Trained on the _Staphylococcus aureus_, WT accumulator phenotype subset of the [human-curated SPARK dataset](https://doi.org/10.1021/acsinfecdis.8b00193) (2115 rows in total for _Staphylococcus aureus_).
+
+## Model details
+
+This model was trained using [our Duvida framework](https://github.com/scbirlab/duvida), 
+as a result of hyperparameter searches and selecting the model that performs best on unseen test data 
+(from a scaffold split). 
+
+Duvida also saves the training data in this checkpoint to allows the calculation of uncertainty metrics 
+based on that training data.
+
+This model is the best regression model from a hyperparameter search, determined
+by Spearman's $\rho$ on a held-out test set not used in training or early stopping. 
+
+### Model architecture
+
+- **Regression**
+
+```json
+
+{
+    "dropout": 0.2,
+    "ensemble_size": 10,
+    "extra_featurizers": null,
+    "learning_rate": 0.0001,
+    "model_class": "ChempropModelBox",
+    "n_hidden": 1,
+    "n_units": 16,
+    "use_2d": false,
+    "use_fp": true
+}
+```
+
+### Model usage
+
+You can use this model with:
+
+```python
+from duvida.autoclasses import AutoModelBox
+modelbox = AutoModelBox.from_pretrained("hf://scbirlab/spark-dv-2503-saur")
+modelbox.predict(filename=..., inputs=[...], columns=[...])  # make predictions on your own data
+```
+
+## Training details
+
+- **Dataset:** [SPARK, WT accumulator, _Staphylococcus aureus_ subset](https://huggingface.co/datasets/scbirlab/thomas-2018-spark-wt)
+- **Input column:** smiles
+- **Output column:** pmic
+- **Split type:** Murcko scaffold
+- **Split proportions:** 
+    - 70% training (1424 rows)
+    - 15% validation (for early stopping) (309 rows)
+    - 15% test (for selecting hyperparameters) (316 rows)
+
+Here is the training log:
+
+<img src="training-log.png" width=450>
+
+And these are the evaluation scores.
+
+Train (1424 rows):
+
+```json
+
+{
+    "Pearson r": 0.8988537726404331,
+    "RMSE": 0.26244109869003296,
+    "Spearman rho": 0.7846286417647882
+}
+```
+
+Validation (309 rows):
+
+```json
+
+{
+    "Pearson r": 0.9436229039637082,
+    "RMSE": 0.3413218855857849,
+    "Spearman rho": 0.8756560069371535
+}
+```
+
+
+Test (316 rows):
+
+```json
+
+{
+    "Pearson r": 0.7476855201027743,
+    "RMSE": 0.7906074523925781,
+    "Spearman rho": 0.8293761872496432
+}
+```
+
+## Training data details
+
+The training data were collated by the authors of:
+
+> Joe Thomas, Marc Navre, Aileen Rubio, and Allan Coukell
+> Shared Platform for Antibiotic Research and Knowledge: A Collaborative Tool to SPARK Antibiotic Discovery
+> ACS Infectious Diseases 2018 4 (11), 1536-1539
+> DOI: 10.1021/acsinfecdis.8b00193
+
+We cleaned the original SPARK dataset to subset the most relevant columns, remove empty values, 
+give succint column titles, and split by species.
+
+This particular dataset retains only measurements on bacteria with wild-type accumulation phenotypes.
+
+### Dataset Sources
+
+- **Repository:** https://www.collaborativedrug.com/spark-data-downloads
+- **Paper:** https://doi.org/10.1021/acsinfecdis.8b00193
+
+### Data Collection and Processing
+
+Data were processed using [schemist](https://github.com/scbirlab/schemist), a tool for processing chemical datasets.
+
+The SMILES strings have been canonicalized, and split into training (70%), validation (15%), and test (15%) sets 
+by Murcko scaffold for each species with more than 1000 entries. Additional features like molecular weight and 
+topological polar surface area have also been calculated.
+
+### Who are the source data producers?
+
+Joe Thomas, Marc Navre, Aileen Rubio, and Allan Coukell

data-config.json

@@ -0,0 +1,17 @@
+{
+    "_default_cache": "cache/duvida/data",
+    "_in_key": "inputs",
+    "_input_cols": [
+        "smiles"
+    ],
+    "_label_cols": [
+        "pmic"
+    ],
+    "_out_key": "labels",
+    "input_shape": [
+        2048
+    ],
+    "output_shape": [
+        1
+    ]
+}

data-load-args.json

@@ -0,0 +1,10 @@
+{
+    "cache": "/nemo/lab/johnsone/home/users/johnsoe/projects/abx-discovery-strategy/models/spark/Staphylococcus-aureus/40/cache",
+    "features": [
+        "smiles"
+    ],
+    "filename": "/nemo/lab/johnsone/home/users/johnsoe/data/datasets/thomas-2018-spark-wt/Staphylococcus-aureus/scaffold-split-train.csv.gz",
+    "labels": [
+        "pmic"
+    ]
+}

eval-metrics_test.json

@@ -0,0 +1,5 @@
+{
+    "Pearson r": 0.7476855201027743,
+    "RMSE": 0.7906074523925781,
+    "Spearman rho": 0.8293761872496432
+}

eval-metrics_train.json

@@ -0,0 +1,5 @@
+{
+    "Pearson r": 0.8988537726404331,
+    "RMSE": 0.26244109869003296,
+    "Spearman rho": 0.7846286417647882
+}

eval-metrics_validation.json

@@ -0,0 +1,5 @@
+{
+    "Pearson r": 0.9436229039637082,
+    "RMSE": 0.3413218855857849,
+    "Spearman rho": 0.8756560069371535
+}

input-data.hf/data-00000-of-00001.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:900da3bf86034317afa2c12fc68b300aedad60e84d41775622b7d1c5f53e85c2
+size 201832

input-data.hf/dataset_info.json

@@ -0,0 +1,52 @@
+{
+  "builder_name": "csv",
+  "citation": "",
+  "config_name": "default",
+  "dataset_name": "csv",
+  "dataset_size": 547635,
+  "description": "",
+  "download_checksums": {
+    "/nemo/lab/johnsone/home/users/johnsoe/data/datasets/thomas-2018-spark-wt/Staphylococcus-aureus/scaffold-split-train.csv.gz": {
+      "num_bytes": 89427,
+      "checksum": null
+    }
+  },
+  "download_size": 89427,
+  "features": {
+    "smiles": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "inputs": {
+      "feature": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "_type": "Sequence"
+    },
+    "labels": {
+      "feature": {
+        "dtype": "float64",
+        "_type": "Value"
+      },
+      "_type": "Sequence"
+    }
+  },
+  "homepage": "",
+  "license": "",
+  "size_in_bytes": 637062,
+  "splits": {
+    "train": {
+      "name": "train",
+      "num_bytes": 547635,
+      "num_examples": 1424,
+      "dataset_name": "csv"
+    }
+  },
+  "version": {
+    "version_str": "0.0.0",
+    "major": 0,
+    "minor": 0,
+    "patch": 0
+  }
+}

input-data.hf/state.json

@@ -0,0 +1,13 @@
+{
+  "_data_files": [
+    {
+      "filename": "data-00000-of-00001.arrow"
+    }
+  ],
+  "_fingerprint": "c23b3d599a50a0d3",
+  "_format_columns": null,
+  "_format_kwargs": {},
+  "_format_type": null,
+  "_output_all_columns": false,
+  "_split": "train"
+}

logs-csv/lightning_logs/version_0/hparams.yaml

@@ -0,0 +1 @@
+{}

logs-csv/lightning_logs/version_0/metrics.csv

@@ -0,0 +1,85 @@
+epoch,loss,step,val_loss
+0,,88,4.601920127868652
+0,7.492936611175537,88,
+1,,177,4.012445449829102
+1,1.5991687774658203,177,
+2,,266,3.6562659740448
+2,1.41478431224823,266,
+3,,355,2.806457996368408
+3,1.3028432130813599,355,
+4,,444,2.421854257583618
+4,1.1816277503967285,444,
+5,,533,2.188049554824829
+5,1.0972480773925781,533,
+6,,622,1.5965540409088135
+6,1.025787591934204,622,
+7,,711,1.5013176202774048
+7,0.9804941415786743,711,
+8,,800,1.1018073558807373
+8,0.9172889590263367,800,
+9,,889,1.0003383159637451
+9,0.8965839743614197,889,
+10,,978,0.8400489687919617
+10,0.8749857544898987,978,
+11,,1067,0.8077839612960815
+11,0.8283694982528687,1067,
+12,,1156,0.6920057535171509
+12,0.7916938662528992,1156,
+13,,1245,0.7309020161628723
+13,0.7782658934593201,1245,
+14,,1334,0.5796863436698914
+14,0.7399182319641113,1334,
+15,,1423,0.44175010919570923
+15,0.6986851096153259,1423,
+16,,1512,0.47962722182273865
+16,0.6836443543434143,1512,
+17,,1601,0.4024624228477478
+17,0.6603368520736694,1601,
+18,,1690,0.38161563873291016
+18,0.6568854451179504,1690,
+19,,1779,0.351207435131073
+19,0.6209790706634521,1779,
+20,,1868,0.3669344484806061
+20,0.5972228050231934,1868,
+21,,1957,0.3335961699485779
+21,0.598533570766449,1957,
+22,,2046,0.35952678322792053
+22,0.5858601927757263,2046,
+23,,2135,0.33657750487327576
+23,0.5555119514465332,2135,
+24,,2224,0.338556706905365
+24,0.5531841516494751,2224,
+25,,2313,0.3455941379070282
+25,0.5307201147079468,2313,
+26,,2402,0.3448736071586609
+26,0.5229456424713135,2402,
+27,,2491,0.3512353301048279
+27,0.49847835302352905,2491,
+28,,2580,0.35296955704689026
+28,0.4923514425754547,2580,
+29,,2669,0.3565228581428528
+29,0.501798152923584,2669,
+30,,2758,0.3635339140892029
+30,0.4853680729866028,2758,
+31,,2847,0.3745820224285126
+31,0.4735918343067169,2847,
+32,,2936,0.3685125708580017
+32,0.4744315445423126,2936,
+33,,3025,0.37449511885643005
+33,0.4655072093009949,3025,
+34,,3114,0.40137070417404175
+34,0.4816225469112396,3114,
+35,,3203,0.3752176761627197
+35,0.4698217213153839,3203,
+36,,3292,0.38168954849243164
+36,0.45470479130744934,3292,
+37,,3381,0.38085946440696716
+37,0.44848036766052246,3381,
+38,,3470,0.3908793032169342
+38,0.4452821612358093,3470,
+39,,3559,0.3829881548881531
+39,0.4430929124355316,3559,
+40,,3648,0.38418999314308167
+40,0.4391266405582428,3648,
+41,,3737,0.38192063570022583
+41,0.4466865658760071,3737,

logs/lightning_logs/version_0/events.out.tfevents.1743097756.cn020.3140491.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f0a9d09df8a21cc42ef0056f12542d37b8b4ffa3e297477a28c26ee74a4445f
+size 7560

logs/lightning_logs/version_0/hparams.yaml

@@ -0,0 +1 @@
+{}

metrics.csv

@@ -0,0 +1,4 @@
+split,split_filename,config_i,model_class,n_parameters,filename,features,labels,cache,extra_featurizers,use_2d,use_fp,dropout,ensemble_size,learning_rate,n_hidden,n_units,val_filename,epochs,batch_size,RMSE,Pearson r,Spearman rho
+train,/nemo/lab/johnsone/home/users/johnsoe/data/datasets/thomas-2018-spark-wt/Staphylococcus-aureus/scaffold-split-train.csv.gz,40,ChempropModelBox,2653010,/nemo/lab/johnsone/home/users/johnsoe/data/datasets/thomas-2018-spark-wt/Staphylococcus-aureus/scaffold-split-train.csv.gz,['smiles'],['pmic'],/nemo/lab/johnsone/home/users/johnsoe/projects/abx-discovery-strategy/models/spark/Staphylococcus-aureus/40/cache,,False,True,0.2,10,0.0001,1,16,/nemo/lab/johnsone/home/users/johnsoe/data/datasets/thomas-2018-spark-wt/Staphylococcus-aureus/scaffold-split-validation.csv.gz,2000,16,0.26244109869003296,0.8988537726404331,0.7846286417647882
+validation,/nemo/lab/johnsone/home/users/johnsoe/data/datasets/thomas-2018-spark-wt/Staphylococcus-aureus/scaffold-split-validation.csv.gz,40,ChempropModelBox,2653010,/nemo/lab/johnsone/home/users/johnsoe/data/datasets/thomas-2018-spark-wt/Staphylococcus-aureus/scaffold-split-train.csv.gz,['smiles'],['pmic'],/nemo/lab/johnsone/home/users/johnsoe/projects/abx-discovery-strategy/models/spark/Staphylococcus-aureus/40/cache,,False,True,0.2,10,0.0001,1,16,/nemo/lab/johnsone/home/users/johnsoe/data/datasets/thomas-2018-spark-wt/Staphylococcus-aureus/scaffold-split-validation.csv.gz,2000,16,0.3413218855857849,0.9436229039637082,0.8756560069371535
+test,/nemo/lab/johnsone/home/users/johnsoe/data/datasets/thomas-2018-spark-wt/Staphylococcus-aureus/scaffold-split-test.csv.gz,40,ChempropModelBox,2653010,/nemo/lab/johnsone/home/users/johnsoe/data/datasets/thomas-2018-spark-wt/Staphylococcus-aureus/scaffold-split-train.csv.gz,['smiles'],['pmic'],/nemo/lab/johnsone/home/users/johnsoe/projects/abx-discovery-strategy/models/spark/Staphylococcus-aureus/40/cache,,False,True,0.2,10,0.0001,1,16,/nemo/lab/johnsone/home/users/johnsoe/data/datasets/thomas-2018-spark-wt/Staphylococcus-aureus/scaffold-split-validation.csv.gz,2000,16,0.7906074523925781,0.7476855201027743,0.8293761872496432

modelbox-config.json

@@ -0,0 +1,11 @@
+{
+    "dropout": 0.2,
+    "ensemble_size": 10,
+    "extra_featurizers": null,
+    "learning_rate": 0.0001,
+    "model_class": "ChempropModelBox",
+    "n_hidden": 1,
+    "n_units": 16,
+    "use_2d": false,
+    "use_fp": true
+}

params.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:736b9d735e4d4c5e1397edee5768bacbca0f5b0e42893aa34208c4323ca64bf3
+size 10696484

predictions_test.csv.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9acde7d7a0bb5b152ac795d43830589811b908ed646b5fdf2b8466524217b5d
+size 120405

predictions_train.csv.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf2f6fb9ac3ebdae0a7d061bfa9d6b306144b9c8f9aa8fcc2d1a8e721360f119
+size 423363

predictions_validation.csv.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7cb0dfa5badc93dcc9080d96c1367b6a54a77c20ae6618c665f5782ca6f3b409
+size 111443

training-args.json

@@ -0,0 +1,5 @@
+{
+    "batch_size": 16,
+    "epochs": 2000,
+    "val_filename": "/nemo/lab/johnsone/home/users/johnsoe/data/datasets/thomas-2018-spark-wt/Staphylococcus-aureus/scaffold-split-validation.csv.gz"
+}

training-data.hf/cache-5e5f3860b4738024.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c30c7291475d48b6adfa4a3184562593468c7eda33ae3a89ef91f54907695a6a
+size 42076408

training-data.hf/data-00000-of-00001.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b7ae972edc8c8b602b7dd444ead34dabd1be1a54538e8cee54a59f1eb395d18
+size 41882032

training-data.hf/dataset_info.json

@@ -0,0 +1,126 @@
+{
+  "builder_name": "csv",
+  "citation": "",
+  "config_name": "default",
+  "dataset_name": "csv",
+  "dataset_size": 547635,
+  "description": "",
+  "download_checksums": {
+    "/nemo/lab/johnsone/home/users/johnsoe/data/datasets/thomas-2018-spark-wt/Staphylococcus-aureus/scaffold-split-train.csv.gz": {
+      "num_bytes": 89427,
+      "checksum": null
+    }
+  },
+  "download_size": 89427,
+  "features": {
+    "smiles": {
+      "feature": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "_type": "Sequence"
+    },
+    "inputs": {
+      "V_d": {
+        "dtype": "null",
+        "_type": "Value"
+      },
+      "gt_mask": {
+        "dtype": "null",
+        "_type": "Value"
+      },
+      "lt_mask": {
+        "dtype": "null",
+        "_type": "Value"
+      },
+      "mg": {
+        "E": {
+          "feature": {
+            "feature": {
+              "dtype": "float32",
+              "_type": "Value"
+            },
+            "_type": "Sequence"
+          },
+          "_type": "Sequence"
+        },
+        "V": {
+          "feature": {
+            "feature": {
+              "dtype": "float32",
+              "_type": "Value"
+            },
+            "_type": "Sequence"
+          },
+          "_type": "Sequence"
+        },
+        "edge_index": {
+          "feature": {
+            "feature": {
+              "dtype": "float32",
+              "_type": "Value"
+            },
+            "_type": "Sequence"
+          },
+          "_type": "Sequence"
+        },
+        "rev_edge_index": {
+          "feature": {
+            "dtype": "float32",
+            "_type": "Value"
+          },
+          "_type": "Sequence"
+        }
+      },
+      "weight": {
+        "dtype": "float32",
+        "_type": "Value"
+      },
+      "x_d": {
+        "feature": {
+          "dtype": "float32",
+          "_type": "Value"
+        },
+        "_type": "Sequence"
+      },
+      "y": {
+        "feature": {
+          "dtype": "float32",
+          "_type": "Value"
+        },
+        "_type": "Sequence"
+      }
+    },
+    "labels": {
+      "feature": {
+        "dtype": "float64",
+        "_type": "Value"
+      },
+      "_type": "Sequence"
+    },
+    "extra_features": {
+      "feature": {
+        "dtype": "float32",
+        "_type": "Value"
+      },
+      "_type": "Sequence"
+    }
+  },
+  "homepage": "",
+  "license": "",
+  "size_in_bytes": 637062,
+  "splits": {
+    "train": {
+      "name": "train",
+      "num_bytes": 547635,
+      "num_examples": 1424,
+      "dataset_name": "csv"
+    }
+  },
+  "version": {
+    "version_str": "0.0.0",
+    "major": 0,
+    "minor": 0,
+    "patch": 0
+  }
+}

training-data.hf/state.json

@@ -0,0 +1,15 @@
+{
+  "_data_files": [
+    {
+      "filename": "data-00000-of-00001.arrow"
+    }
+  ],
+  "_fingerprint": "5629eb6ca89ad40c",
+  "_format_columns": null,
+  "_format_kwargs": {
+    "dtype": "float"
+  },
+  "_format_type": "numpy",
+  "_output_all_columns": false,
+  "_split": "train"
+}

training-log.csv

@@ -0,0 +1,43 @@
+epoch,step,loss,val_loss
+0,88,7.492936611175537,4.601920127868652
+1,177,1.5991687774658203,4.012445449829102
+2,266,1.41478431224823,3.6562659740448
+3,355,1.30284321308136,2.806457996368408
+4,444,1.1816277503967283,2.421854257583618
+5,533,1.097248077392578,2.188049554824829
+6,622,1.025787591934204,1.5965540409088137
+7,711,0.9804941415786744,1.5013176202774048
+8,800,0.9172889590263368,1.1018073558807373
+9,889,0.8965839743614197,1.0003383159637451
+10,978,0.8749857544898987,0.8400489687919617
+11,1067,0.8283694982528687,0.8077839612960815
+12,1156,0.7916938662528992,0.6920057535171509
+13,1245,0.7782658934593201,0.7309020161628723
+14,1334,0.7399182319641113,0.5796863436698914
+15,1423,0.6986851096153259,0.4417501091957092
+16,1512,0.6836443543434143,0.4796272218227386
+17,1601,0.6603368520736694,0.4024624228477478
+18,1690,0.6568854451179504,0.3816156387329101
+19,1779,0.6209790706634521,0.351207435131073
+20,1868,0.5972228050231934,0.3669344484806061
+21,1957,0.598533570766449,0.3335961699485779
+22,2046,0.5858601927757263,0.3595267832279205
+23,2135,0.5555119514465332,0.3365775048732757
+24,2224,0.5531841516494751,0.338556706905365
+25,2313,0.5307201147079468,0.3455941379070282
+26,2402,0.5229456424713135,0.3448736071586609
+27,2491,0.498478353023529,0.3512353301048279
+28,2580,0.4923514425754547,0.3529695570468902
+29,2669,0.501798152923584,0.3565228581428528
+30,2758,0.4853680729866028,0.3635339140892029
+31,2847,0.4735918343067169,0.3745820224285126
+32,2936,0.4744315445423126,0.3685125708580017
+33,3025,0.4655072093009949,0.37449511885643
+34,3114,0.4816225469112396,0.4013707041740417
+35,3203,0.4698217213153839,0.3752176761627197
+36,3292,0.4547047913074493,0.3816895484924316
+37,3381,0.4484803676605224,0.3808594644069671
+38,3470,0.4452821612358093,0.3908793032169342
+39,3559,0.4430929124355316,0.3829881548881531
+40,3648,0.4391266405582428,0.3841899931430816
+41,3737,0.4466865658760071,0.3819206357002258