{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))\n\n# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import os, gc, pickle\nimport pandas as pd\nimport matplotlib.pyplot as plt\nimport numpy as np\nfrom colorama import Fore, Back, Style\nfrom matplotlib.ticker import MaxNLocator\n\nfrom sklearn.base import BaseEstimator, TransformerMixin\nfrom sklearn.model_selection import KFold, GroupKFold\nfrom sklearn.preprocessing import StandardScaler\n, scale\nfrom sklearn.decomposition import PCA, TruncatedSVD\nfrom sklearn.dummy import DummyRegressor\nfrom sklearn.pipeline import make_pipeline, Pipeline\nfrom sklearn.linear_model import Ridge, LinearRegression, Lasso\nfrom sklearn.metrics import mean_squared_error\n\nimport scipy\nimport scipy.sparse\n\nimport gc\nimport pickle\nimport warnings\nimport random\nwarnings.filterwarnings('ignore')\n\ndef seed_everything(seed):\n    random.seed(seed)\n    np.random.seed(seed)\n    os.environ['PYTHONHASHSEED'] = str(seed)\n\nseed_everything(666)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"DATA_DIR = \"/kaggle/input/open-problems-multimodal\"\nFP_CELL_METADATA = os.path.join(DATA_DIR,\"metadata.csv\")\n\nFP_CITE_TRAIN_INPUTS = os.path.join(DATA_DIR,\"train_cite_inputs.h5\")\nFP_CITE_TRAIN_TARGETS = os.path.join(DATA_DIR,\"train_cite_targets.h5\")\nFP_CITE_TEST_INPUTS = os.path.join(DATA_DIR,\"test_cite_inputs.h5\")\n\nFP_MULTIOME_TRAIN_INPUTS = os.path.join(DATA_DIR,\"train_multi_inputs.h5\")\nFP_MULTIOME_TRAIN_TARGETS = os.path.join(DATA_DIR,\"train_multi_targets.h5\")\nFP_MULTIOME_TEST_INPUTS = os.path.join(DATA_DIR,\"test_multi_inputs.h5\")\n\nFP_SUBMISSION = os.path.join(DATA_DIR,\"sample_submission.csv\")\nFP_EVALUATION_IDS = os.path.join(DATA_DIR,\"evaluation_ids.csv\")\n\nCROSS_VALIDATE = True\nSUBMIT = True","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def correlation_score(y_true, y_pred):\n    \"\"\"Scores the predictions according to the competition rules. \n    \n    It is assumed that the predictions are not constant.\n    \n    Returns the average of each sample's Pearson correlation coefficient\"\"\"\n    if type(y_true) == pd.DataFrame: y_true = y_true.values\n    if type(y_pred) == pd.DataFrame: y_pred = y_pred.values\n    corrsum = 0\n    for i in range(len(y_true)):\n        corrsum += np.corrcoef(y_true[i], y_pred[i])[1, 0]\n    return corrsum / len(y_true)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import scipy\nimport scipy.sparse\n\nimport gc\nimport pickle\nimport warnings\nwarnings.filterwarnings('ignore')","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"%%time\ntrain_inputs = scipy.sparse.load_npz(\"../input/multimodal-single-cell-as-sparse-matrix/train_multi_inputs_values.sparse.npz\")","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_inputs = train_inputs.astype('float16', copy=False)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_targets = scipy.sparse.load_npz(\"../input/multimodal-single-cell-as-sparse-matrix/train_multi_targets_values.sparse.npz\")","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"with open('../input/fork-of-msci-multiome-randomsampling-sp-6b182b/pca.pkl', 'rb') as f: \n    pca = pickle.load(f)\n\nwith open('../input/fork-of-msci-multiome-randomsampling-sp-6b182b/pca2.pkl', 'rb') as f: \n    pca2 = pickle.load(f)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"pca\npca2","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_inputs = pca.fit_transform(train_inputs)\nprint(pca.explained_variance_ratio_.sum())\ntrain_target = pca2.fit_transform(train_targets)\nprint(pca2.explained_variance_ratio_.sum())","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def save(name, model):\n    with open(name, 'wb') as f:\n        pickle.dump(model, f)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# read metadata\nmetadata_df = pd.read_csv(FP_CELL_METADATA, index_col='cell_id')\nmetadata_df = metadata_df[metadata_df.technology==\"multiome\"]\nmetadata_df.shape","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"cell_index = np.load('../input/multiome-cell-index/multiome_cell_index.npy')\ncell_index = cell_index.tolist()\n\n# based on multome cell_ids, get the corresponding meta data\nmeta = metadata_df.reindex(cell_index)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from sklearn.preprocessing import QuantileTransformer\n\nqt = QuantileTransformer(n_quantiles = 100, output_distribution = 'normal', random_state = 42)\n\n\nmulti_test_x = scipy.sparse.load_npz(\"../input/multimodal-single-cell-as-sparse-matrix/test_multi_inputs_values.sparse.npz\")\nmulti_test_x = pca.transform(multi_test_x)\n\n\n\n\nqt.fit(pd.concat([pd.DataFrame(train_inputs), pd.DataFrame(multi_test_x)]))\ntrain_inputs = qt.transform(train_inputs)\nmulti_test_x = qt.transform(multi_test_x)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import lightgbm as lgb\nfrom sklearn.multioutput import MultiOutputRegressor\n\nparams = {\n     'learning_rate': 0.1, \n     'metric': 'mae', \n     \"seed\": 42,\n    'reg_alpha': 0.0014, \n    'reg_lambda': 0.2, \n    'colsample_bytree': 0.8, \n    'subsample': 0.5, \n    'max_depth': 10, \n    'num_leaves': 722, \n    'min_child_samples': 83, \n    }\n","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"%%time\n# Cross-validation\n\nnp.random.seed(42)\nall_row_indices = np.arange(train_inputs.shape[0])\nnp.random.shuffle(all_row_indices)\n\n# kf = KFold(n_splits=5, shuffle=True, random_state=42)\n\nkf = GroupKFold(n_splits=3)\n\nindex = 0\nscore = []\n\n# kf = KFold(n_splits=5, shuffle=True, random_state=1)\nscore_list = []\nfor fold, (idx_tr, idx_va) in enumerate(kf.split(train_inputs, groups=meta.donor)):\n    model = None\n    gc.collect()\n    X_tr = train_inputs[idx_tr] # creates a copy, https://numpy.org/doc/stable/user/basics.copies.html\n    y_tr = train_target[idx_tr]\n    del idx_tr\n\n    model = MultiOutputRegressor(lgb.LGBMRegressor(**params, n_estimators=128))\n    model.fit(X_tr, y_tr)\n    filename = f\"model{fold}.pkl\"\n    with open(filename, 'wb') as file:\n        pickle.dump(model, file)\n    del X_tr, y_tr\n    gc.collect()\n\n    # We validate the model\n    X_va = train_inputs[idx_va]\n    y_va = train_target[idx_va]\n    del idx_va\n    y_va_pred = model.predict(X_va)\n    mse = mean_squared_error(y_va, y_va_pred)\n    corrscore = correlation_score(y_va, y_va_pred)\n    del X_va, y_va\n\n    print(f\"Fold {fold}: mse = {mse:.5f}, corr =  {corrscore:.3f}\")\n    score_list.append((mse, corrscore))\n\n# Show overall score\nresult_df = pd.DataFrame(score_list, columns=['mse', 'corrscore'])\nprint(f\"{Fore.GREEN}{Style.BRIGHT}{train_inputs.shape} Average  mse = {result_df.mse.mean():.5f}; corr = {result_df.corrscore.mean():.3f}{Style.RESET_ALL}\")","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"del train_target, train_inputs, train_targets\ngc.collect()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"n = 1\ntest_len = multi_test_x.shape[0]\nd = test_len//n\nx = []\nfor i in range(n):\n    x.append(multi_test_x[i*d:i*d+d])\ndel multi_test_x\ngc.collect()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"preds = np.zeros((test_len, 23418), dtype='float16')\nfor i,xx in enumerate(x):\n    for ind in range(index):\n        print(ind, end=' ')\n        for fold in range(3):\n            with open(f'model{fold}.pkl', 'rb') as file:\n                model = pickle.load(file)\n            preds[i*d:i*d+d,:] += (model.predict(xx)@pca2.components_)/index\n        preds[i*d:i*d+d,:] /= 3\n        gc.collect()\n    print('')\n    del xx\ngc.collect()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"del x\ngc.collect()","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"np.save('preds.npy', preds)","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# %%time\n# Read the table of rows and columns required for submission\neval_ids = pd.read_parquet(\"../input/multimodal-single-cell-as-sparse-matrix/evaluation.parquet\") \n# Convert the string columns to more efficient categorical types\neval_ids.cell_id = eval_ids.cell_id.astype(pd.CategoricalDtype())\neval_ids.gene_id = eval_ids.gene_id.astype(pd.CategoricalDtype())","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Prepare an empty series which will be filled with predictions\nsubmission = pd.Series(name='target',\n                       index=pd.MultiIndex.from_frame(eval_ids), \n                       dtype=np.float32)\nsubmission","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# %%time\ny_columns = np.load(\"../input/multimodal-single-cell-as-sparse-matrix/train_multi_targets_idxcol.npz\",\n                   allow_pickle=True)[\"columns\"]\n\ntest_index = np.load(\"../input/multimodal-single-cell-as-sparse-matrix/test_multi_inputs_idxcol.npz\",\n                    allow_pickle=True)[\"index\"]","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"cell_dict = dict((k,v) for v,k in enumerate(test_index)) \nassert len(cell_dict)  == len(test_index)\n\ngene_dict = dict((k,v) for v,k in enumerate(y_columns))\nassert len(gene_dict) == len(y_columns)","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"eval_ids_cell_num = eval_ids.cell_id.apply(lambda x:cell_dict.get(x, -1))\neval_ids_gene_num = eval_ids.gene_id.apply(lambda x:gene_dict.get(x, -1))\n\nvalid_multi_rows = (eval_ids_gene_num !=-1) & (eval_ids_cell_num!=-1)","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"submission.iloc[valid_multi_rows] = preds[eval_ids_cell_num[valid_multi_rows].to_numpy(),\neval_ids_gene_num[valid_multi_rows].to_numpy()]","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"del eval_ids_cell_num, eval_ids_gene_num, valid_multi_rows, eval_ids, test_index, y_columns\ngc.collect()","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"submission.to_csv(\"submission_M_lgbm_QuantileTransformer.csv\")","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"submission.reset_index(drop=True, inplace=True)\nsubmission.index.name = 'row_id'","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}