{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.13","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":56537,"databundleVersionId":8015876,"sourceType":"competition"}],"dockerImageVersionId":30699,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import pandas as pd\nfrom tqdm import tqdm\nimport numpy as np\nfrom sklearn.decomposition import PCA\nimport time\npd.set_option('io.hdf.default_format','table')","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2024-04-24T16:36:07.707941Z","iopub.execute_input":"2024-04-24T16:36:07.708324Z","iopub.status.idle":"2024-04-24T16:36:07.714519Z","shell.execute_reply.started":"2024-04-24T16:36:07.708297Z","shell.execute_reply":"2024-04-24T16:36:07.713252Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def rearrange_df(data, leave_out_cols):\n    leave_out_df = pd.concat([data.pop(col) for col in leave_out_cols], axis=1)\n    data = pd.concat([data, leave_out_df], axis=1)\n    return data\n\nTRAIN = '/kaggle/input/leap-atmospheric-physics-ai-climsim/train.csv'\nTEST = '/kaggle/input/leap-atmospheric-physics-ai-climsim/test.csv'\nSUB = '/kaggle/input/leap-atmospheric-physics-ai-climsim/sample_submission.csv'\n\n\nX_LEAVE_OUT_COLS = ['pbuf_SOLIN', 'pbuf_SHFLX', 'pbuf_LHFLX', 'cam_in_LWUP', 'state_ps']\nY_LEAVE_OUT_COLS = ['cam_out_SOLLD', 'cam_out_SOLL', 'cam_out_SOLS', 'cam_out_SOLSD', 'cam_out_FLWDS', 'cam_out_NETSW']\nX_COLS = rearrange_df(pd.read_csv(TEST, nrows=1).drop('sample_id', axis=1), X_LEAVE_OUT_COLS).columns.tolist()\nY_COLS = rearrange_df(pd.read_csv(SUB, nrows=1).drop('sample_id', axis=1), Y_LEAVE_OUT_COLS).columns.tolist()\n\n# For debugging notebook, change to 500_000 before running notebook\nCHECK_ROWS = 500_000","metadata":{"execution":{"iopub.status.busy":"2024-04-24T16:36:07.717023Z","iopub.execute_input":"2024-04-24T16:36:07.718068Z","iopub.status.idle":"2024-04-24T16:36:07.807182Z","shell.execute_reply.started":"2024-04-24T16:36:07.718039Z","shell.execute_reply":"2024-04-24T16:36:07.806116Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Why?\n\nData will be biggest bottleneck in the contest, as most of us do not have the hardware capable of working with it efficiently. And even using methods like processing the data chunkwise, we still waste a lot of time moving the data between disk, cpu and gpu. But the data is also extremely informationally sparse, with many columns having a single unique value, and the rest having very low standard deviations. This also makes it harder for our models to learn. Thus it becomes very important to reduce the size of the data by any means, hopefully while losing as less information as possible.\n\n# Downcasting\nAll the columns are float64, but most of the values are within the float32 range. Let's see how much info we lose if we downcast it to float32, which should halve the memory usage required. We will use MSE as a metric to see how much, and if the data loses any information.","metadata":{}},{"cell_type":"code","source":"start = time.time()\n\n# For example, let's consider 500,000 rows\nchunk = pd.read_csv(TRAIN, nrows=CHECK_ROWS, usecols=X_COLS+Y_COLS)\noriginal_mem_used = chunk.memory_usage(deep=True).sum() / (1024*1024)\nprint(f'Memory used by the original chunk: {original_mem_used} MB')\n\nchunk_32 = chunk.astype('float32')\nf32_mem_used = chunk_32.memory_usage(deep=True).sum() / (1024*1024)\nprint(f'Memory used by the downcasted chunk: {f32_mem_used} MB')\nprint('-'*80)\n\nprint('Mean MSE over all features between original and downcasted Dataset: ',\n     ((chunk - chunk_32)** 2).mean().mean())\n\nprint('Max MSE over all data values between original and downcasted Dataset: ',\n     ((chunk - chunk_32) ** 2).max().max())\nprint('-'*80)\n\nprint(f\"Took {time.time() - start}s to finish this cell\")\nprint(f'Total Memory usage reduced from original: {original_mem_used/f32_mem_used}x')","metadata":{"execution":{"iopub.status.busy":"2024-04-24T16:36:07.808438Z","iopub.execute_input":"2024-04-24T16:36:07.808736Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Alright! A mean MSE loss of 0.000000005 over the dataset, and the largest value changed in the notebook had a MSE of just 0.000015. We can make do with that, almost no loss in data for a whopping 50% size reduction.\n\nNext we test some dimensionality reduction techniques. The data is extremely sparse, so reducing the dimensions while preserving information should bump up speed by a fair amount, and also make it easier for our models to learn. \n\nLet's start with the PCA. Although PCA is usually only done on the predictor variables, our target data is also somewhat unusual with over 370 variables, most of them also very sparse. So we will try to reduce the target variables as well, and then inverse_transform the predicted target to original dimensions. We must also scale our data before using PCA, as is the standard practice.\n\n![image.png](attachment:00379743-9770-47f7-bc08-0a1fdab58bc8.png)","metadata":{},"attachments":{"00379743-9770-47f7-bc08-0a1fdab58bc8.png":{"image/png":"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"}}},{"cell_type":"markdown","source":"**Note**:\nWhen testing this, I realized keeping a few relatively high-variance columns in the PCA significantly makes it harder. It needs more PCA components to preserve variance, and the inverse_transform for these columns is also bad. Since our main goal is to reduce size and maintain information, I will simply be skipping these columns from PCA process. All the other columns get scaled and PCA applied on them, these 'leave out columns' will only be scaled, and then concatenated again to the other columns post-PCA before being fed to the model.","metadata":{}},{"cell_type":"code","source":"from sklearn.pipeline import make_pipeline\nfrom sklearn.preprocessing import StandardScaler, RobustScaler\nfrom sklearn.decomposition import IncrementalPCA as IPCA\nimport plotly.express as px\nimport gc\n\nclass CustomPipeline:\n    def __init__(self, all_cols, leave_out_cols, pca_components, batch_size):\n        self.cols = all_cols\n        self.scaler = StandardScaler()\n        self.leave_out_cols = leave_out_cols\n        self.pca = IPCA(n_components=pca_components, batch_size=batch_size)\n    \n    def partial_fit(self, data):\n        self.scaler.partial_fit(data)\n        temp_data = pd.DataFrame(self.scaler.transform(data), columns=self.cols).astype('float32')\n        temp_data.drop(self.leave_out_cols, axis=1, inplace=True)\n        self.pca.partial_fit(temp_data.to_numpy())\n        gc.collect()\n        \n    def transform(self, data):\n        temp_data = pd.DataFrame(self.scaler.transform(data).astype('float32'), columns=self.cols)      \n        backup_cols = temp_data[self.leave_out_cols]\n        temp_data.drop(self.leave_out_cols, axis=1, inplace=True)\n        temp_data = self.pca.transform(temp_data.to_numpy())\n        return np.hstack((temp_data, backup_cols.to_numpy())).astype('float32')\n    \n    def inverse_transform(self, data):\n        backup_cols = data[:, -len(self.leave_out_cols):].copy()\n        data = data[:, :-len(self.leave_out_cols)].copy()\n        data = self.pca.inverse_transform(data)\n        data = pd.DataFrame(np.hstack((data, backup_cols)), columns=self.cols)\n        return self.scaler.inverse_transform(data).astype('float32')\n        \n\nstart = time.time()\n\nchunk_x = chunk_32[X_COLS].copy(deep=True)\nchunk_y = chunk_32[Y_COLS].copy(deep=True)\n\npipeline_x = CustomPipeline(X_COLS, X_LEAVE_OUT_COLS, 200, CHECK_ROWS)\npipeline_y = CustomPipeline(Y_COLS, Y_LEAVE_OUT_COLS, 270, CHECK_ROWS)\n\npipeline_x.partial_fit(chunk_x.copy())\npipeline_y.partial_fit(chunk_y.copy())\n\ntransformed_x = pd.DataFrame(pipeline_x.transform(chunk_x)).astype('float32')\ngc.collect()\ntransformed_y = pd.DataFrame(pipeline_y.transform(chunk_y)).astype('float32')\ngc.collect()\n\nx_pca_components = pipeline_x.pca.explained_variance_ratio_\ny_pca_components = pipeline_y.pca.explained_variance_ratio_\n\nprint(f'PCA components to preserve {x_pca_components.sum()*100}% variance on predictors: {len(x_pca_components)}')\nprint(f'PCA components to preserve {y_pca_components.sum()*100}% variance on targets: {len(y_pca_components)}')\nprint('-'*80)\n\ntransformedx_mem_used = transformed_x.memory_usage(deep=True).sum() / (1024*1024)\ntransformedy_mem_used = transformed_y.memory_usage(deep=True).sum() / (1024*1024)\n\nprint(f'Memory used by the original Data: {original_mem_used} MB')\nprint(f'Memory used by the transformed Data: {transformedx_mem_used+transformedy_mem_used} MB')\nprint(f'Total Memory usage reduced from original: {original_mem_used/(transformedx_mem_used+transformedy_mem_used)}x')\nprint('-'*80)\n\ninvtransformed_x = pipeline_x.inverse_transform(transformed_x.to_numpy())\ninvtransformed_y = pipeline_y.inverse_transform(transformed_y.to_numpy())\n\ngc.collect()\n\nfig = px.line(((invtransformed_y - chunk_32[Y_COLS]) ** 2).mean())\nfig.show()\n\nprint(f\"Took {time.time() - start}s to finish this cell\")\nprint(f'Total Memory usage reduced from original: {original_mem_used/(transformedx_mem_used+transformedy_mem_used)}x')\n\ntotal_data = pd.concat((transformed_x, transformed_y), axis=1).astype('float32')\nfinal_shape = total_data.shape","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Wonderful! So we have managed to reduce our datasize by almost 4x, while preserving almost 99.9% variance (perhaps more, considering the left-out columns had high variance themselves. The MSE after the inverse_transform is also phenomenal, almost neglible. Thus we can be fairly confident that we can inverse transform our predicted variables post-prediction into the same dimension space. \n\nFor rapid prototyping of your model, you could perhaps even reduce the variance threshold down to 95% by setting the X and Y PCA components to 50 and 165 respectively, which reduces the dataset size by about 9x. At that point, you should be able to load the entire training dataset on the kaggle 2xT4 GPU.\n\nAnother point to be noted is, a lot of the variance in the target/Y-data seems to be just noise. As even with just 10% variance preserved, inverse_transform is pretty much able to reconstruct the data again. This does NOT hold for the predictor variables, they lose inversability fast as they lose preserved variance. Let's try fitting these in a model now. To make this fast, I will be going with 95% variance preserved, for the final, we should probably ramp it up to 99.9% or so.","metadata":{}},{"cell_type":"code","source":"from sklearn.neighbors import KNeighborsRegressor\nfrom sklearn.model_selection import cross_val_score, KFold\nfrom sklearn.metrics import r2_score\n\ndef get_score_from_data(data_x, data_y, pipeline):\n    start = time.time()\n    cv = KFold(n_splits=3, random_state=42, shuffle=True)\n    model = KNeighborsRegressor(n_neighbors=30)\n    for train_index, test_index in cv.split(data_x):\n        X_train, X_test = data_x.iloc[train_index], data_x.iloc[test_index]\n        y_train, y_test = data_y.iloc[train_index], data_y.iloc[test_index]\n        chunk_y_test = chunk_y.iloc[test_index]\n\n        model.fit(X_train, y_train)\n        y_pred = model.predict(X_test)\n        # If pipeline is not provided, then its original data, else its transformed\n        # and needs inverse transforming.\n        if pipeline is not None:\n            y_pred = pipeline.inverse_transform(y_pred)\n\n        scores = r2_score(chunk_y_test, y_pred)\n        print(\"Average over all Columns R2 score this split: \", np.mean(scores))\n        print('-'*80)\n    \n\n    print(f\"Total Time Taken: {time.time() - start} \")\n\n\nprint('-'*80)\nprint(\"Fitting on KNeighborsRegressor on the transformed data:\")\nget_score_from_data(transformed_x, transformed_y, pipeline_y)\n\n_ = gc.collect()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Now to compare, let's train the same model on our original data and check what the R2-score on that would be.","metadata":{}},{"cell_type":"code","source":"get_score_from_data(chunk_x, chunk_y, None)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Wonderful! Almost 3x less time, on 4x less dimension space, and our KNearestRegressor even shows a very slightly better CV score on the transformed data.\n\nLet's transform the entire data now, and save it. ","metadata":{}},{"cell_type":"code","source":"def get_entire_train_data_chunks():\n    reader = pd.read_csv(TRAIN, chunksize=CHECK_ROWS)\n    for chunk in reader:\n        chunk = chunk[X_COLS+Y_COLS].astype('float32')\n        yield chunk","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Remove all the old data to free up memory space\ndel chunk, chunk_32, chunk_x, chunk_y, transformed_x, transformed_y, total_data, invtransformed_x, invtransformed_y\n_ = gc.collect()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"fpipeline_x = CustomPipeline(X_COLS, X_LEAVE_OUT_COLS, 200, CHECK_ROWS)\nfpipeline_y = CustomPipeline(Y_COLS, Y_LEAVE_OUT_COLS, 270, CHECK_ROWS)\n\ncounter = 0\n\nfor chunk in get_entire_train_data_chunks():\n    _ = gc.collect()\n    start = time.time()\n\n    chunk_x = chunk[X_COLS].copy(deep=True)\n    chunk_y = chunk[Y_COLS].copy(deep=True)\n    \n    fpipeline_x.partial_fit(chunk_x)\n    fpipeline_y.partial_fit(chunk_y)\n    \n    print(f'Finished {counter+1} chunk in {time.time() - start} seconds.')\n    counter += 1\n\n","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from joblib import dump, load\n\n# Saving the fitted Pipelines to inverse transform later when you need it.\n\n!mkdir '/kaggle/working/models'\ndump(fpipeline_x, '/kaggle/working/models/final_pipelinex.joblib')\ndump(fpipeline_y, '/kaggle/working/models/final_pipeliney.joblib')","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import h5py\nimport os\n\nh5_filename = '/kaggle/working/train_data_transformed.h5'\nwith h5py.File(h5_filename, 'w') as f:\n    dset = f.create_dataset('train_transformed', (0, final_shape[1]), maxshape=(None, final_shape[1]), dtype='float32',\n                           compression='gzip', compression_opts=9)\n    \ncounter = 0\n# Now to transform and save the training data.\nfor chunk in get_entire_train_data_chunks():\n    _ = gc.collect()\n    start = time.time()\n\n    chunk_x = chunk[X_COLS].copy(deep=True)\n    chunk_y = chunk[Y_COLS].copy(deep=True)\n    \n    transformed_x = fpipeline_x.transform(chunk_x)\n    transformed_y = fpipeline_y.transform(chunk_y)\n    \n    total_data = np.concatenate((transformed_x, transformed_y), axis=1).astype('float32')\n    with h5py.File(h5_filename, 'a') as f:\n        dset = f['train_transformed']\n        new_size = dset.shape[0] + total_data.shape[0]\n        dset.resize(new_size, axis=0)\n        dset[dset.shape[0] - total_data.shape[0]:] = total_data\n        f.flush()\n        \n        print(f'File size and shape after append: {os.path.getsize(h5_filename) / (1024*1024)} MB and {dset.shape}')\n\n    \n    print(f'Finished {counter+1} chunk in {time.time() - start} seconds.')\n    counter += 1\n    _ = gc.collect()\n\n","metadata":{"trusted":true},"execution_count":null,"outputs":[]}]}