{"cells":[{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"import collections, re\nfrom pathlib import Path\nfrom matplotlib import pyplot as plt\n\n\nPROJECT_DIR = Path('..')\nINPUT_DIR = PROJECT_DIR / 'input' / 'bms-molecular-translation'\nTRAIN_LABELS_PATH = INPUT_DIR / 'train_labels.csv'","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def count_elements(input_file_path):\n    line_regex = re.compile('^[0-9a-f]+,\"?InChI=1S/([^/]+)')\n    element_regex = re.compile('([A-Z][a-z]?)[0-9]*')\n    formulas = []\n    formula_count = 0\n    with open(input_file_path, 'r') as f:\n        first_line = True\n        for line in f.readlines():\n            if first_line:\n                # skip the header line\n                first_line = False\n                continue\n            match = line_regex.match(line)\n            if not match:\n                print('Warning - line not matched:', line)\n                continue\n            formula = match.groups(0)[0]\n            formula_count += 1\n            formulas.append(formula)\n    # Count the elements: increment an element's count each time an element occurs in a formula.\n    element_counts = collections.Counter()\n    for formula in formulas:\n        elements = element_regex.findall(formula)\n        if not 'C' in elements:\n            print('NOTE: no carbon in:', formula)\n        element_counts.update(elements)\n    # Convert counts to a list of (element, frequency) pairs.\n    element_counts = list(element_counts.items())\n    # Sort counts by descending frequency.\n    element_counts = sorted(element_counts, key=lambda pair: -pair[1])\n    return formula_count, element_counts\n\n\nTRAIN_FORMULA_COUNT, TRAIN_ELEMENT_COUNTS = count_elements(TRAIN_LABELS_PATH)\nprint(f'Training data element counts: {TRAIN_ELEMENT_COUNTS} in {TRAIN_FORMULA_COUNT} formulas.')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def show_element_counts(element_counts, title, log=False):\n    elements = [pair[0] for pair in element_counts]\n    counts = [pair[1] for pair in element_counts]\n    if not log:\n        counts = [count/10**6 for count in counts]\n    y_pos = [-i for i, _ in enumerate(counts)]\n    plt.figure(figsize=(10, 8), facecolor='#eef')\n    plt.title(title)\n    plt.xlabel('No. of formulas' + (not log and ' (millions)' or ''))\n    plt.ylabel('Element')\n    plt.barh(y_pos, counts, log=log, fc='gold')\n    plt.yticks(y_pos, elements)\n    plt.show()\n    print('Done.')\n    return\n\n\nshow_element_counts(TRAIN_ELEMENT_COUNTS, 'Training Data Element Counts', log=False)","execution_count":null,"outputs":[]}],"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":4,"nbformat_minor":4}