done

API_flask_server.py

@@ -0,0 +1,29 @@
+## An python API for our Model
+## Design by edander32 (Edmilson Alexandre) and jjambo(Joaquim Jambo)
+
+from flask import Flask, request, jsonify
+import pandas as pd
+import joblib
+
+app = Flask(__name__)
+
+model = joblib.load("exoplanet_model.pkl")
+
+@app.route("/")
+def home():
+    return {"message": "API do classificador de exoplanetas online."}
+
+@app.route("/predict", methods=["POST"])
+def predict():
+    try:
+        data = request.get_json()
+        df = pd.DataFrame([data])
+        pred = model.predict(df)[0]
+        prob = model.predict_proba(df)[0][1]
+        return jsonify({"prediction": int(pred), "probability": float(prob)})
+    except Exception as e:
+        return jsonify({"error": str(e)}), 400
+
+if __name__ == "__main__":
+    app.run(host="0.0.0.0", port=7860)
+

NASA.ipynb

@@ -0,0 +1 @@
+{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"provenance":[],"authorship_tag":"ABX9TyO5J/RoFrpl8dZBuNzBEwXn"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"}},"cells":[{"cell_type":"code","source":["!pip install -q lightgbm tsfresh optuna scikit-learn pandas numpy matplotlib seaborn joblib"],"metadata":{"id":"wHEiah4us0eh","executionInfo":{"status":"ok","timestamp":1759609790439,"user_tz":-60,"elapsed":9712,"user":{"displayName":"Edmilson Alexandre","userId":"05182212798533402489"}},"colab":{"base_uri":"https://localhost:8080/"},"outputId":"b24f02b2-de03-4178-aa06-6dfdae9e2094"},"execution_count":2,"outputs":[{"output_type":"stream","name":"stdout","text":["\u001b[?25l   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m0.0/400.9 kB\u001b[0m \u001b[31m?\u001b[0m eta \u001b[36m-:--:--\u001b[0m\r\u001b[2K   \u001b[91m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[91m╸\u001b[0m\u001b[90m━━━━━\u001b[0m \u001b[32m348.2/400.9 kB\u001b[0m \u001b[31m10.4 MB/s\u001b[0m eta \u001b[36m0:00:01\u001b[0m\r\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m400.9/400.9 kB\u001b[0m \u001b[31m8.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n","\u001b[?25h"]}]},{"cell_type":"code","source":["import os\n","import numpy\n","import pandas\n","import joblib\n","import optuna\n","import tsfresh\n","import sklearn\n","import seaborn\n","import lightgbm\n","import matplotlib\n","from google.colab import drive\n","\n","\n","def EDA(file):\n","  pandas.set_option('display.max_columns', 200)\n","  target_col = 'target'\n","  print(\"Shape from file: \", file.shape)\n","  # display(file.head())\n","  # print(file.dtypes)\n","\n","def PipelineCreation(file, target_col):\n","  from sklearn.model_selection import train_test_split\n","  from sklearn.pipeline import Pipeline\n","  from sklearn.impute import SimpleImputer\n","  from sklearn.preprocessing import OneHotEncoder\n","  from sklearn.compose import ColumnTransformer\n","\n","  id_cols = [c for c in ['id', 'time', 'index'] if c in file.columns]\n","  x = file.drop(columns=[target_col] + id_cols, errors='ignore')\n","  y = file[target_col]\n","\n","  numeric_features = x.select_dtypes(include=['number']).columns.tolist()\n","  categorical_features = x.select_dtypes(include=['object', 'category', 'bool']).columns.tolist()\n","  # print(\"Numeric: \", len(numeric_features))\n","  # print(\"Categorical: \", len(categorical_features))\n","  # print(\"X: \", x)\n","  # print(\"Y: \", y)\n","\n","  numeric_transformer = Pipeline([('imputer', SimpleImputer(strategy='median')),])\n","\n","  categorical_transformer = Pipeline([('imputer', SimpleImputer(strategy='most_frequent')),\n","                                      ('ohe', OneHotEncoder(handle_unknown='ignore')),])\n","  print(f\"Num:{len(numeric_features)}\")\n","  preprocessor = ColumnTransformer([\n","      ('num', numeric_transformer, numeric_features),\n","      ('cat', categorical_transformer, categorical_features),\n","  ], remainder='drop')\n","  return x, y, preprocessor\n","\n","def Training(x_train_tr, y_train):\n","  from lightgbm import LGBMClassifier\n","  scale_pos_weight = (y_train==0).sum() / max(1, (y_train==1).sum())\n","  model = LGBMClassifier(\n","      objective='binary',\n","      n_estimators=10000,\n","      learning_rate=0.05,\n","      num_leaves=31,\n","      random_state=42,\n","      scale_pos_weight=scale_pos_weight\n","  )\n","  model.fit(\n","      x_train_tr, y_train,\n","      eval_set=[(x_val_tr, y_val)],\n","      eval_metric='auc',\n","      #my version does not support that method. I need to use callbacks\n","      #early_stopping_rounds=100,\n","      #callbacks=[early_stopping(100), log_evaluation(100)]\n","  )\n","  print(\"Best iteration:\", model.best_iteration_)\n","  print(\"Train AUC:\", model.best_score_['training']['auc'])\n","  print(\"Valid AUC:\", model.best_score_['valid_0']['auc'])\n","\n","  return model\n","\n","\n","\n","from sklearn.model_selection import train_test_split\n","if os.path.exists('/content/drive') == 0:\n","  drive.mount('/content/drive')\n","\n","labels = pandas.read_csv('/content/drive/MyDrive/AI_assets/labels.csv')\n","light_curves  = pandas.read_csv('/content/drive/MyDrive/AI_assets/light_curves.csv')\n","metadata = pandas.read_csv('/content/drive/MyDrive/AI_assets/metadata.csv')\n","data = pandas.read_csv('/content/drive/MyDrive/data2.csv')\n","\n","# im gonna change this under cause i nedd more classes. Using binary interpretation insted of 'CONFIRMED' will help a lot\n","#x, y, preprocessor = PipelineCreation(data, target_col='kepoi_name')\n","\n","data['target'] = data['koi_disposition'].map(\n","    lambda v: 1 if v == \"CONFIRMED\" else 0\n",")\n","\n","x, y, preprocessor = PipelineCreation(data, target_col='target')\n","print(\"First step done. 1 -> PIPELINE CREATION\")\n","# print(\"X: \", x)\n","# print(\"Y: \", y)\n","x_train, x_val, y_train, y_val = train_test_split(\n","    x, y, test_size=0.20, stratify=y, random_state=42\n"," )\n","preprocessor.fit(x_train)\n","x_train_tr = preprocessor.transform(x_train)\n","x_val_tr = preprocessor.transform(x_val)\n","EDA(labels)\n","\n","#debug purposes\n","print(\"Second step done. 2 -> EDA DONE\")\n","print(\"X_train shape:\", x_train_tr.shape)\n","print(\"X_val shape:\", x_val_tr.shape)\n","print(\"y_train distribution:\\n\", y_train.value_counts())\n","print(\"y_val distribution:\\n\", y_val.value_counts())\n","print(\"Check for NaNs:\", x_train_tr.isna().sum().sum(), \"in train,\", x_val_tr.isna().sum().sum(), \"in val\")\n","\n","\n","#model = Training(x_train_tr, y_train)\n","#print(\"Finished Training. 3!!\")\n","\n","\n","# A litle degub made by GPT\n","# print(\"Best iteration:\", model.best_iteration_)\n","# print(\"Train AUC:\", model.best_score_['training']['auc'])\n","# print(\"Valid AUC:\", model.best_score_['valid_0']['auc'])\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":873},"id":"-kgzOBe6QXbJ","executionInfo":{"status":"error","timestamp":1759611069253,"user_tz":-60,"elapsed":20,"user":{"displayName":"Edmilson Alexandre","userId":"05182212798533402489"}},"outputId":"98dc3b31-acee-4a6d-c30e-9f15901a1b25"},"execution_count":8,"outputs":[{"output_type":"stream","name":"stdout","text":["Num:133\n","First step done. 1 -> PIPELINE CREATION\n","Shape from file:  (4, 2)\n","Second step done. 2 -> EDA DONE\n","X_train shape: (8, 172)\n","X_val shape: (2, 172)\n","y_train distribution:\n"," target\n","1    6\n","0    2\n","Name: count, dtype: int64\n","y_val distribution:\n"," target\n","0    1\n","1    1\n","Name: count, dtype: int64\n"]},{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.12/dist-packages/sklearn/impute/_base.py:635: UserWarning: Skipping features without any observed values: ['koi_gmag_err' 'koi_rmag_err' 'koi_imag_err' 'koi_zmag_err'\n"," 'koi_kepmag_err' 'koi_model_dof' 'koi_model_chisq' 'koi_eccen_err1'\n"," 'koi_eccen_err2' 'koi_longp' 'koi_longp_err1' 'koi_longp_err2'\n"," 'koi_sma_err1' 'koi_sma_err2' 'koi_ingress' 'koi_ingress_err1'\n"," 'koi_ingress_err2' 'koi_incl_err1' 'koi_incl_err2' 'koi_teq_err1'\n"," 'koi_teq_err2' 'koi_sage' 'koi_sage_err1' 'koi_sage_err2']. At least one non-missing value is needed for imputation with strategy='median'.\n","  warnings.warn(\n","/usr/local/lib/python3.12/dist-packages/sklearn/impute/_base.py:635: UserWarning: Skipping features without any observed values: ['koi_gmag_err' 'koi_rmag_err' 'koi_imag_err' 'koi_zmag_err'\n"," 'koi_kepmag_err' 'koi_model_dof' 'koi_model_chisq' 'koi_eccen_err1'\n"," 'koi_eccen_err2' 'koi_longp' 'koi_longp_err1' 'koi_longp_err2'\n"," 'koi_sma_err1' 'koi_sma_err2' 'koi_ingress' 'koi_ingress_err1'\n"," 'koi_ingress_err2' 'koi_incl_err1' 'koi_incl_err2' 'koi_teq_err1'\n"," 'koi_teq_err2' 'koi_sage' 'koi_sage_err1' 'koi_sage_err2']. At least one non-missing value is needed for imputation with strategy='median'.\n","  warnings.warn(\n","/usr/local/lib/python3.12/dist-packages/sklearn/impute/_base.py:635: UserWarning: Skipping features without any observed values: ['koi_gmag_err' 'koi_rmag_err' 'koi_imag_err' 'koi_zmag_err'\n"," 'koi_kepmag_err' 'koi_model_dof' 'koi_model_chisq' 'koi_eccen_err1'\n"," 'koi_eccen_err2' 'koi_longp' 'koi_longp_err1' 'koi_longp_err2'\n"," 'koi_sma_err1' 'koi_sma_err2' 'koi_ingress' 'koi_ingress_err1'\n"," 'koi_ingress_err2' 'koi_incl_err1' 'koi_incl_err2' 'koi_teq_err1'\n"," 'koi_teq_err2' 'koi_sage' 'koi_sage_err1' 'koi_sage_err2']. At least one non-missing value is needed for imputation with strategy='median'.\n","  warnings.warn(\n"]},{"output_type":"error","ename":"AttributeError","evalue":"'numpy.ndarray' object has no attribute 'isna'","traceback":["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m","\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)","\u001b[0;32m/tmp/ipython-input-2064020500.py\u001b[0m in \u001b[0;36m<cell line: 0>\u001b[0;34m()\u001b[0m\n\u001b[1;32m    109\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"y_train distribution:\\n\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_train\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalue_counts\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    110\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"y_val distribution:\\n\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_val\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalue_counts\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 111\u001b[0;31m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Check for NaNs:\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx_train_tr\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0misna\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"in train,\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx_val_tr\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0misna\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"in val\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    112\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    113\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;31mAttributeError\u001b[0m: 'numpy.ndarray' object has no attribute 'isna'"]}]},{"cell_type":"code","source":[],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"CAErcbk0yL-d","executionInfo":{"status":"ok","timestamp":1759540476879,"user_tz":-60,"elapsed":10596,"user":{"displayName":"Edmilson Alexandre","userId":"05182212798533402489"}},"outputId":"571ef5ef-7d8f-462c-8449-2d4444e39ed6"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Collecting tabgan==1.3.3\n","  Downloading tabgan-1.3.3-py2.py3-none-any.whl.metadata (10.0 kB)\n","Requirement already satisfied: pandas in /usr/local/lib/python3.12/dist-packages (from tabgan==1.3.3) (2.2.2)\n","Requirement already satisfied: numpy in /usr/local/lib/python3.12/dist-packages (from tabgan==1.3.3) (2.0.2)\n","Collecting category-encoders (from tabgan==1.3.3)\n","  Downloading category_encoders-2.8.1-py3-none-any.whl.metadata (7.9 kB)\n","Requirement already satisfied: torch>=1.0 in /usr/local/lib/python3.12/dist-packages (from tabgan==1.3.3) (2.8.0+cu126)\n","Requirement already satisfied: lightgbm>=2.2.3 in /usr/local/lib/python3.12/dist-packages (from tabgan==1.3.3) (4.6.0)\n","Requirement already satisfied: scikit-learn>=1.0.2 in /usr/local/lib/python3.12/dist-packages (from tabgan==1.3.3) (1.6.1)\n","Requirement already satisfied: torchvision in /usr/local/lib/python3.12/dist-packages (from tabgan==1.3.3) (0.23.0+cu126)\n","Requirement already satisfied: python-dateutil in /usr/local/lib/python3.12/dist-packages (from tabgan==1.3.3) (2.9.0.post0)\n","Requirement already satisfied: tqdm in /usr/local/lib/python3.12/dist-packages (from tabgan==1.3.3) (4.67.1)\n","Requirement already satisfied: scipy in /usr/local/lib/python3.12/dist-packages (from lightgbm>=2.2.3->tabgan==1.3.3) (1.16.2)\n","Requirement already satisfied: joblib>=1.2.0 in /usr/local/lib/python3.12/dist-packages (from scikit-learn>=1.0.2->tabgan==1.3.3) (1.5.2)\n","Requirement already satisfied: threadpoolctl>=3.1.0 in /usr/local/lib/python3.12/dist-packages (from scikit-learn>=1.0.2->tabgan==1.3.3) (3.6.0)\n","Requirement already satisfied: filelock in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (3.19.1)\n","Requirement already satisfied: typing-extensions>=4.10.0 in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (4.15.0)\n","Requirement already satisfied: setuptools in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (75.2.0)\n","Requirement already satisfied: sympy>=1.13.3 in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (1.13.3)\n","Requirement already satisfied: networkx in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (3.5)\n","Requirement already satisfied: jinja2 in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (3.1.6)\n","Requirement already satisfied: fsspec in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (2025.3.0)\n","Requirement already satisfied: nvidia-cuda-nvrtc-cu12==12.6.77 in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (12.6.77)\n","Requirement already satisfied: nvidia-cuda-runtime-cu12==12.6.77 in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (12.6.77)\n","Requirement already satisfied: nvidia-cuda-cupti-cu12==12.6.80 in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (12.6.80)\n","Requirement already satisfied: nvidia-cudnn-cu12==9.10.2.21 in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (9.10.2.21)\n","Requirement already satisfied: nvidia-cublas-cu12==12.6.4.1 in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (12.6.4.1)\n","Requirement already satisfied: nvidia-cufft-cu12==11.3.0.4 in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (11.3.0.4)\n","Requirement already satisfied: nvidia-curand-cu12==10.3.7.77 in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (10.3.7.77)\n","Requirement already satisfied: nvidia-cusolver-cu12==11.7.1.2 in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (11.7.1.2)\n","Requirement already satisfied: nvidia-cusparse-cu12==12.5.4.2 in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (12.5.4.2)\n","Requirement already satisfied: nvidia-cusparselt-cu12==0.7.1 in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (0.7.1)\n","Requirement already satisfied: nvidia-nccl-cu12==2.27.3 in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (2.27.3)\n","Requirement already satisfied: nvidia-nvtx-cu12==12.6.77 in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (12.6.77)\n","Requirement already satisfied: nvidia-nvjitlink-cu12==12.6.85 in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (12.6.85)\n","Requirement already satisfied: nvidia-cufile-cu12==1.11.1.6 in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (1.11.1.6)\n","Requirement already satisfied: triton==3.4.0 in /usr/local/lib/python3.12/dist-packages (from torch>=1.0->tabgan==1.3.3) (3.4.0)\n","Requirement already satisfied: patsy>=0.5.1 in /usr/local/lib/python3.12/dist-packages (from category-encoders->tabgan==1.3.3) (1.0.1)\n","Requirement already satisfied: statsmodels>=0.9.0 in /usr/local/lib/python3.12/dist-packages (from category-encoders->tabgan==1.3.3) (0.14.5)\n","Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.12/dist-packages (from pandas->tabgan==1.3.3) (2025.2)\n","Requirement already satisfied: tzdata>=2022.7 in /usr/local/lib/python3.12/dist-packages (from pandas->tabgan==1.3.3) (2025.2)\n","Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.12/dist-packages (from python-dateutil->tabgan==1.3.3) (1.17.0)\n","Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in /usr/local/lib/python3.12/dist-packages (from torchvision->tabgan==1.3.3) (11.3.0)\n","Requirement already satisfied: packaging>=21.3 in /usr/local/lib/python3.12/dist-packages (from statsmodels>=0.9.0->category-encoders->tabgan==1.3.3) (25.0)\n","Requirement already satisfied: mpmath<1.4,>=1.1.0 in /usr/local/lib/python3.12/dist-packages (from sympy>=1.13.3->torch>=1.0->tabgan==1.3.3) (1.3.0)\n","Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.12/dist-packages (from jinja2->torch>=1.0->tabgan==1.3.3) (3.0.3)\n","Downloading tabgan-1.3.3-py2.py3-none-any.whl (28 kB)\n","Downloading category_encoders-2.8.1-py3-none-any.whl (85 kB)\n","\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m85.7/85.7 kB\u001b[0m \u001b[31m2.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n","\u001b[?25hInstalling collected packages: category-encoders, tabgan\n","Successfully installed category-encoders-2.8.1 tabgan-1.3.3\n"]}]}]}

README.md

@@ -1,10 +1,2 @@
----
-title: Exo
-emoji: 😻
-colorFrom: gray
-colorTo: indigo
-sdk: docker
-pinned: false
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# AI-Exo-Hunter
+Hunting exoplanets using AI

model.py

@@ -0,0 +1,279 @@
+import os
+import numpy
+import pandas
+import joblib
+import optuna
+import tsfresh
+import sklearn
+import seaborn
+import lightgbm
+import matplotlib
+from google.colab import drive
+
+
+def EDA(file):
+  pandas.set_option('display.max_columns', 200)
+  target_col = 'target'
+  print("Shape from file: ", file.shape)
+  # display(file.head())
+  # print(file.dtypes)
+
+def PipelineCreation(file, target_col):
+  from sklearn.model_selection import train_test_split
+  from sklearn.pipeline import Pipeline
+  from sklearn.impute import SimpleImputer
+  from sklearn.preprocessing import OneHotEncoder
+  from sklearn.compose import ColumnTransformer
+
+  id_cols = [c for c in ['id', 'time', 'index'] if c in file.columns]
+  x = file.drop(columns=[target_col] + id_cols, errors='ignore')
+  y = file[target_col]
+
+  numeric_features = x.select_dtypes(include=['number']).columns.tolist()
+  categorical_features = x.select_dtypes(include=['object', 'category', 'bool']).columns.tolist()
+  # print("Numeric: ", len(numeric_features))
+  # print("Categorical: ", len(categorical_features))
+  # print("X: ", x)
+  # print("Y: ", y)
+
+  numeric_transformer = Pipeline([('imputer', SimpleImputer(strategy='median')),])
+
+  #categorical_transformer = Pipeline([('imputer', SimpleImputer(strategy='most_frequent')),
+  #                                    ('ohe', OneHotEncoder(handle_unknown='ignore')),])
+  categorical_transformer = Pipeline([
+    ('imputer', SimpleImputer(strategy='most_frequent')),
+    ('ohe', OneHotEncoder(handle_unknown='ignore', sparse_output=False)),
+])
+
+  print(f"Num:{len(numeric_features)}")
+  preprocessor = ColumnTransformer([
+      ('num', numeric_transformer, numeric_features),
+      ('cat', categorical_transformer, categorical_features),
+  ], remainder='drop')
+  return x, y, preprocessor
+
+def Training(x_train_clean, y_train, x_val_clean, y_val):
+  # from lightgbm import LGBMClassifier
+  # scale_pos_weight = (y_train==0).sum() / max(1, (y_train==1).sum())
+  # model = LGBMClassifier(
+  #     objective='binary',
+  #     n_estimators=10000,
+  #     learning_rate=0.05,
+  #     num_leaves=31,
+  #     random_state=42,
+  #     scale_pos_weight=scale_pos_weight
+  # )
+  # model.fit(
+  #     x_train_tr, y_train,
+  #     eval_set=[(x_val_tr, y_val)],
+  #     eval_metric='auc',
+  #     #my version does not support that method. I need to use callbacks
+  #     #early_stopping_rounds=100,
+  #     #callbacks=[early_stopping(100), log_evaluation(100)]
+  # )
+  # print("Best iteration:", model.best_iteration_)
+  # print("Train AUC:", model.best_score_['training']['auc'])
+  # print("Valid AUC:", model.best_score_['valid_0']['auc'])
+  from lightgbm import LGBMClassifier
+
+  # model = LGBMClassifier(
+  #     objective='binary',
+  #     n_estimators=2000,
+  #     learning_rate=0.05,
+  #     num_leaves=31,
+  #     min_child_samples=1,
+  #     random_state=42
+  # )
+
+  model = LGBMClassifier(
+    objective='binary',
+    n_estimators=2000,
+    learning_rate=0.05,
+    num_leaves=31,
+    min_child_samples=1,
+    min_split_gain=0.0,  # permite splits mesmo sem ganho positivo aparente
+    min_data_in_leaf=1,  # remove restrição mínima
+    random_state=42
+)
+
+  model.fit(
+      x_train_clean, y_train,
+      eval_set=[(x_val_clean, y_val)],
+      eval_metric='auc',
+  )
+
+  print("Best iteration:", model.best_iteration_)
+
+
+  return model
+
+
+
+from sklearn.model_selection import train_test_split
+if os.path.exists('/content/drive') == 0:
+  drive.mount('/content/drive')
+
+labels = pandas.read_csv('/content/drive/MyDrive/AI_assets/labels.csv')
+light_curves  = pandas.read_csv('/content/drive/MyDrive/AI_assets/light_curves.csv')
+metadata = pandas.read_csv('/content/drive/MyDrive/AI_assets/metadata.csv')
+#data = pandas.read_csv('/content/drive/MyDrive/exoplanets_normalized_data.csv')
+data = pandas.read_csv('/content/drive/MyDrive/data.csv')
+
+# im gonna change this under cause i nedd more classes. Using binary interpretation insted of 'CONFIRMED' will help a lot
+#x, y, preprocessor = PipelineCreation(data, target_col='kepoi_name')
+
+data['target'] = data['koi_disposition'].map(
+    lambda v: 1 if v == "CONFIRMED" else 0
+)
+
+x, y, preprocessor = PipelineCreation(data, target_col='target')
+print("First step done. 1 -> PIPELINE CREATION")
+# print("X: ", x)
+# print("Y: ", y)
+x_train, x_val, y_train, y_val = train_test_split(
+    x, y, test_size=0.20, stratify=y, random_state=42
+ )
+preprocessor.fit(x_train)
+x_train_tr = preprocessor.transform(x_train)
+x_val_tr = preprocessor.transform(x_val)
+EDA(labels)
+
+#debug purposes
+#print("Second step done. 2 -> EDA DONE")
+#print("X_train shape:", x_train_tr.shape)
+#print("X_val shape:", x_val_tr.shape)
+#print("y_train distribution:\n", y_train.value_counts())
+#print("y_val distribution:\n", y_val.value_counts())
+#print("Check for NaNs:", x_train_tr.isna().sum().sum(), "in train,", x_val_tr.isna().sum().sum(), "in val")
+import pandas as pd
+
+x_train_df = pd.DataFrame(x_train_tr)
+x_val_df = pd.DataFrame(x_val_tr)
+
+# Remover colunas sem variância
+valid_cols = x_train_df.columns[x_train_df.var() > 0]
+x_train_clean = x_train_df[valid_cols]
+x_val_clean = x_val_df[valid_cols]
+
+import numpy as np
+import pandas as pd
+
+# X_train é seu conjunto de treino (DataFrame)
+variancias = x_train_df.var()
+sem_variancia = (variancias == 0).sum()
+
+# print(f"Número de colunas com variância zero: {sem_variancia}/{len(variancias)}")
+# print("Exemplo de variâncias não nulas:")
+# print(variancias[variancias > 0].head())
+
+# print("Número de colunas após filtragem:", len(valid_cols))
+# print("Shape final de treino:", x_train_clean.shape)
+# print("Distribuição de classes:")
+# print(y_train.value_counts(normalize=True))
+# print("Exemplo de valores:")
+# print(x_train_clean.head())
+
+
+# print("Tipo de dados:", x_train_clean.dtypes.unique())
+# print("Faixa de valores nas primeiras colunas:")
+# print(x_train_clean.iloc[:, :5].describe())
+# import numpy as np
+
+# print("Min:", np.min(x_train_clean.values))
+# print("Max:", np.max(x_train_clean.values))
+# print("Mean:", np.mean(x_train_clean.values))
+# print("Std:", np.std(x_train_clean.values))
+from sklearn.preprocessing import StandardScaler
+from sklearn.feature_selection import VarianceThreshold
+from lightgbm import LGBMClassifier
+
+# === 1. Remover colunas quase constantes ===
+var_sel = VarianceThreshold(threshold=1e-4)
+x_train_filtered = var_sel.fit_transform(x_train_clean)
+x_val_filtered = var_sel.transform(x_val_clean)
+
+#print("Shape após VarianceThreshold:", x_train_filtered.shape)
+
+# === 2. Padronizar ===
+scaler = StandardScaler()
+x_train_scaled = scaler.fit_transform(x_train_filtered)
+x_val_scaled = scaler.transform(x_val_filtered)
+
+# === 3. Treinar modelo ===
+# model = LGBMClassifier(
+#     objective='binary',
+#     n_estimators=2000,
+#     learning_rate=0.05,
+#     num_leaves=31,
+#     random_state=42
+# )
+
+# model.fit(
+#     x_train_scaled, y_train,
+#     eval_set=[(x_val_scaled, y_val)],
+#     eval_metric='auc'
+# )
+
+# print("Best iteration:", model.best_iteration_)
+
+#model = Training(x_train_clean, y_train, x_val_clean, y_val)
+#print("Finished Training. 3!!")
+# import numpy as np
+# import pandas as pd
+
+# # Converter se ainda estiver em numpy array
+# x_train_df = pd.DataFrame(x_train_clean)
+
+# # 1️⃣ Correlação média entre cada feature e o target
+# corrs = []
+# for col in x_train_df.columns:
+#     try:
+#         corrs.append(abs(np.corrcoef(x_train_df[col], y_train)[0,1]))
+#     except:
+#         corrs.append(0)
+# mean_corr = np.nanmean(corrs)
+# print("Correlação média com o target:", mean_corr)
+
+# # 2️⃣ Contar quantas colunas têm correlação > 0.05
+# useful = np.sum(np.array(corrs) > 0.05)
+# print("Features com correlação > 0.05:", useful, "/", len(corrs))
+
+
+# A litle degub made by GPT
+# print("Best iteration:", model.best_iteration_)
+# print("Train AUC:", model.best_score_['training']['auc'])
+# print("Valid AUC:", model.best_score_['valid_0']['auc'])
+import numpy as np
+
+if hasattr(x_train_clean, "todense"):
+    x_train_clean = np.array(x_train_clean.todense(), dtype=np.float32)
+    x_val_clean = np.array(x_val_clean.todense(), dtype=np.float32)
+else:
+    x_train_clean = np.array(x_train_clean, dtype=np.float32)
+    x_val_clean = np.array(x_val_clean, dtype=np.float32)
+print("Treino - tipo:", type(x_train_clean))
+print("Treino - shape:", x_train_clean.shape)
+print("Treino - dtype:", x_train_clean.dtype)
+print("Valores únicos de y:", np.unique(y_train, return_counts=True))
+
+#model = Training(x_train_clean, y_train, x_val_clean, y_val)
+from lightgbm import LGBMClassifier
+
+model = LGBMClassifier(
+    objective='binary',
+    learning_rate=0.01,
+    n_estimators=2000,
+    random_state=42
+)
+model.fit(x_train_clean, y_train)
+train_score = model.score(x_train_clean, y_train)
+print("Treino score:", train_score)
+
+from sklearn.metrics import accuracy_score
+
+y_pred = model.predict(x_val_clean)
+acc = accuracy_score(y_val, y_pred)
+print(f"Acurácia: {acc:.4f}")
+
+
+

requirements.txt

@@ -0,0 +1,5 @@
+flask
+pandas
+joblib
+scikit-learn
+lightgbm