{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"id": "96b34bd1",
"metadata": {},
"outputs": [],
"source": [
"import os\n",
"import pandas as pd\n",
"import numpy as np\n",
"import matplotlib.pyplot as plt\n",
"import seaborn as sns\n",
"import xgboost as xgb\n",
"import sklearn"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "37b99e9f",
"metadata": {},
"outputs": [],
"source": [
"model_name = 'models/xgboost_first_model.json'\n",
"xgb_model1 = xgb.XGBClassifier()\n",
"xgb_model1.load_model(model_name)\n",
"model_name = 'models/xgboost_second_model.json'\n",
"xgb_model2 = xgb.XGBClassifier()\n",
"xgb_model2.load_model(model_name)\n",
"model_name = 'models/xgboost_third_model.json'\n",
"xgb_model3 = xgb.XGBClassifier()\n",
"xgb_model3.load_model(model_name)"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "40aa73e8",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"2022\n",
"2023\n",
"2024\n",
"2025\n"
]
}
],
"source": [
"m3_pred = []\n",
"m2_pred = []\n",
"m1_pred = []\n",
"bird_list = []\n",
"year_list = []\n",
"dur_list = []\n",
"tim_list = []\n",
"fil_list = []\n",
"aug_list = []\n",
"\n",
"for year in [2022,2023,2024,2025]:\n",
"# for year in [2022,2023,2024]:\n",
" print(year)\n",
" birds = os.listdir(f'/mnt/d/acoustics-data/birdclef-{year}/train_audio')\n",
" for bird in birds:\n",
" file = f'/mnt/d/acoustics-data/processed/birdclef-{year}/{bird}_features.npy'\n",
" try:\n",
" X = np.load(file,)\n",
" n = X.shape[0]\n",
" bird_list += [bird for _ in range(n)]\n",
" year_list += [year for _ in range(n)]\n",
" y_pred_proba = xgb_model1.predict_proba(X)[:,1]\n",
" m1_pred += list(y_pred_proba)\n",
" y_pred_proba = xgb_model2.predict_proba(X)[:,1]\n",
" m2_pred += list(y_pred_proba)\n",
" y_pred_proba = xgb_model3.predict_proba(X)[:,1]\n",
" m3_pred += list(y_pred_proba)\n",
" tab = pd.read_csv(f'/mnt/d/acoustics-data/processed/birdclef-{year}/{bird}_slices.csv')\n",
" dur_list += tab['duration'].tolist()\n",
" tim_list += tab['start_time'].tolist()\n",
" fil_list += tab['file_name'].tolist()\n",
" aug_list += tab['is_augmented'].tolist()\n",
" except:\n",
" pass"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "876d3800",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" bird | \n",
" year | \n",
" model_1 | \n",
" model_2 | \n",
" model_3 | \n",
" duration | \n",
" time | \n",
" file_name | \n",
" is_augmented | \n",
"
\n",
" \n",
" \n",
" \n",
" | 0 | \n",
" afrsil1 | \n",
" 2022 | \n",
" 0.520478 | \n",
" 0.992990 | \n",
" 0.997792 | \n",
" 3.5004 | \n",
" 0.0 | \n",
" XC207432.ogg | \n",
" 0 | \n",
"
\n",
" \n",
" | 1 | \n",
" afrsil1 | \n",
" 2022 | \n",
" 0.512172 | \n",
" 0.992990 | \n",
" 0.998557 | \n",
" 3.5004 | \n",
" 0.0 | \n",
" XC207432.ogg | \n",
" 1 | \n",
"
\n",
" \n",
" | 2 | \n",
" afrsil1 | \n",
" 2022 | \n",
" 0.374894 | \n",
" 0.985676 | \n",
" 0.996787 | \n",
" 3.5004 | \n",
" 0.0 | \n",
" XC207432.ogg | \n",
" 1 | \n",
"
\n",
" \n",
" | 3 | \n",
" afrsil1 | \n",
" 2022 | \n",
" 0.348209 | \n",
" 0.987380 | \n",
" 0.994968 | \n",
" 3.5004 | \n",
" 0.0 | \n",
" XC207432.ogg | \n",
" 1 | \n",
"
\n",
" \n",
" | 4 | \n",
" afrsil1 | \n",
" 2022 | \n",
" 0.469575 | \n",
" 0.984070 | \n",
" 0.995817 | \n",
" 3.5004 | \n",
" 0.0 | \n",
" XC207432.ogg | \n",
" 1 | \n",
"
\n",
" \n",
"
\n",
"