Merge pull request WayScience#36 from MikeLippincott/31-output-model-…

…architecture-to-json-for-increased-readability 31 output model architecture to json for increased readability
MikeLippincott · Jul 11, 2023 · 8c11899 · 8c11899
2 parents 0dc143a + 24eb581
commit 8c11899
Show file tree

Hide file tree

Showing 20 changed files with 1,933 additions and 1,544 deletions.
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
@@ -1,53 +1,34 @@
 ---
 repos:
-# isort for jupyter notebooks
-  - repo: https://github.com/nbQA-dev/nbQA
-    rev: 1.3.1
+  # Formats import order
+  - repo: https://github.com/pycqa/isort
+    rev: 5.12.0
     hooks:
-      - id: nbqa-isort
+      - id: isort
+        name: isort (python)
+        args: ["--profile", "black", "--filter-files"]
 
-  # Code formatter for both python files and jupyter notebooks
+  #Code formatter for both python files and jupyter notebooks
   - repo: https://github.com/psf/black
     rev: 22.10.0
     hooks:
+      - id: black-jupyter
       - id: black
         language_version: python3.10
 
+  # code for converting jupyter notebooks to python scripts
   - repo: https://github.com/mwouts/jupytext
     rev: v1.14.0  # CURRENT_TAG/COMMIT_HASH
     hooks:
     - id: jupytext
       args: [--from, notebooks///ipynb, --to, "scripts///py:percent", --sync, --pipe, black]
 
-  # Formats import order
-  - repo: https://github.com/pycqa/isort
-    rev: 5.12.0
-    hooks:
-      - id: isort
-        name: isort (python)
-        args: ["--profile", "black", "--filter-files"]
-
   # remove unused imports
   - repo: https://github.com/hadialqattan/pycln.git
     rev: v2.1.3
     hooks:
       - id: pycln
 
-
-  # Code formatter for both python files and jupyter notebooks
-  - repo: https://github.com/psf/black
-    rev: 22.10.0
-    hooks:
-      - id: black-jupyter
-      - id: black
-        language_version: python3.10
-
-  # isort for jupyter notebooks
-  - repo: https://github.com/nbQA-dev/nbQA
-    rev: 1.3.1
-    hooks:
-      - id: nbqa-isort
-
   # additional hooks found with in the pre-commit lib
   - repo: https://github.com/pre-commit/pre-commit-hooks
     rev: v3.4.0
@@ -61,5 +42,3 @@ repos:
           - --autofix
           - --indent=4
           - --no-sort-keys
-
-
diff --git a/MLP_Model/MLP_utils/config.toml b/MLP_Model/MLP_utils/config.toml
@@ -22,4 +22,4 @@ LEARNING_RATE_MAX = 0.1
 OPTIMIZER_LIST = [ "Adam", "RMSprop", "SGD" ]
 METRIC = "loss"
 DIRECTION = "minimize"
-MODEL_NAME = "LPS_10_vs_DMSO"
+MODEL_NAME = "LPS_10_vs_LPS_100"
diff --git a/MLP_Model/MLP_utils/utils.py b/MLP_Model/MLP_utils/utils.py
@@ -3,7 +3,7 @@
 These are helper functions meant to be called in a separate notebook or script
 """
 
-import ast
+import json
 import pathlib
 from pathlib import Path
 from typing import Tuple
@@ -21,7 +21,7 @@
     ModelTypeError,
     OptimizationMetricError,
     TrainingValidationTestingSplitError,
-    yDataTypeError,
+    YDataTypeError,
 )
 from MLP_utils.parameters import Parameters
 from sklearn.metrics import (
@@ -370,7 +370,7 @@ def train_n_validate(
         elif params.MODEL_TYPE == "Regression":
             pass
         else:
-            raise yDataTypeError
+            raise YDataTypeError
 
         X_train_batch, y_train_batch = X_train_batch.to(
             params.DEVICE
@@ -426,7 +426,7 @@ def train_n_validate(
             elif params.MODEL_TYPE == "Regression":
                 pass
             else:
-                raise yDataTypeError
+                raise YDataTypeError
 
             X_valid_batch, y_valid_batch = X_valid_batch.to(
                 params.DEVICE
@@ -673,26 +673,26 @@ def extract_best_trial_params(
 
     if MLP_params.MODEL_TYPE == "Multi_Class":
         with open(
-            f"../../trained_models/architectures/Multi_Class/Multi_Class_{model_name}.txt",
+            f"../../trained_models/architectures/Multi_Class/Multi_Class_{model_name}.json",
             "w",
         ) as f:
-            f.write(str(param_dict))
+            json.dump(param_dict, f, indent=4)
         f.close()
 
     elif MLP_params.MODEL_TYPE == "Binary_Classification":
         with open(
-            f"../../trained_models/architectures/Binary_Classification/Binary_Classification_{model_name}.txt",
+            f"../../trained_models/architectures/Binary_Classification/Binary_Classification_{model_name}.json",
             "w",
         ) as f:
-            f.write(str(param_dict))
+            json.dump(param_dict, f, indent=4)
         f.close()
 
     elif MLP_params.MODEL_TYPE == "Regression":
         with open(
-            f"../../trained_models/architectures/Regression/Regression_{model_name}.txt",
+            f"../../trained_models/architectures/Regression/Regression_{model_name}.json",
             "w",
         ) as f:
-            f.write(str(param_dict))
+            json.dump(param_dict, f, indent=4)
         f.close()
 
     else:
@@ -727,28 +727,25 @@ def optimized_model_create(
     # load in model architecture from saved model architecture
     if params.MODEL_TYPE == "Multi_Class":
         with open(
-            f"../../trained_models/architectures/Multi_Class/Multi_Class_{model_name}.txt",
+            f"../../trained_models/architectures/Multi_Class/Multi_Class_{model_name}.json",
             "r",
         ) as f:
-            parameter_dict = ast.literal_eval(f.read())
+            parameter_dict = json.load(f)
         f.close()
-
     elif params.MODEL_TYPE == "Binary_Classification":
         with open(
-            f"../../trained_models/architectures/Binary_Classification/Binary_Classification_{model_name}.txt",
+            f"../../trained_models/architectures/Binary_Classification/Binary_Classification_{model_name}.json",
             "r",
         ) as f:
-            parameter_dict = ast.literal_eval(f.read())
+            parameter_dict = json.load(f)
         f.close()
-
     elif params.MODEL_TYPE == "Regression":
         with open(
-            f"../../trained_models/architectures/Regression/Regression_{model_name}.txt",
+            f"../../trained_models/architectures/Regression/Regression_{model_name}.json",
             "r",
         ) as f:
-            parameter_dict = ast.literal_eval(f.read())
+            parameter_dict = json.load(f)
         f.close()
-
     else:
         raise ModelTypeError
 

diff --git a/MLP_Model/model_testing/notebooks/binary_classification_testing.ipynb b/MLP_Model/model_testing/notebooks/binary_classification_testing.ipynb
diff --git a/MLP_Model/model_testing/scripts/binary_classification_testing.py b/MLP_Model/model_testing/scripts/binary_classification_testing.py
@@ -39,7 +39,7 @@
 )
 from sklearn.model_selection import train_test_split
 
-sys.path.append("..")
+sys.path.append("../..")
 from MLP_utils.parameters import Parameters
 from MLP_utils.utils import (
     Dataset_formatter,
@@ -55,20 +55,20 @@
     un_nest,
 )
 
-sys.path.append("../..")
+sys.path.append("../../..")
 from utils.utils import df_stats
 
 # %%
 # Import Data
 # set data file path under pathlib path for multi-system use
 file_path = Path(
-    "../../../Extracted_Features_(CSV_files)/SHSY5Y_preprocessed_df_sc_norm.parquet"
+    "../../../../Extracted_Features_(CSV_files)/SHSY5Y_preprocessed_df_sc_norm.parquet"
 )
 
 df = pq.read_table(file_path).to_pandas()
 
 # %%
-data = Path("../MLP_utils/config.toml")
+data = Path("../../MLP_utils/config.toml")
 config = toml.load(data)
 params = Parameters()
 params = parameter_set(params, config)