Trying formatting with an older yapf version 0.17.0 to see if it passes Travis CI

            cid_list = np.append(cid_list, df["PUBCHEM_CID"].as_matrix())

            assay_no = assay_no + 1

            if assay_no % 100 == 0:

              print("Parsed: {0} of: {1}".format(assay_no,

                                                 len(assays_to_parse)))

              print(

                  "Parsed: {0} of: {1}".format(assay_no, len(assays_to_parse)))

    print("Convert to CID set")

    cid_set = np.unique(cid_list)

      assay_names.append(assay_name)

      assay_results.append(assay_results_array)

      assay_end = time.time()

      print("Parsed: {0} in {1} seconds".format(assay_name,

                                                assay_end - assay_start))

      print("Parsed: {0} in {1} seconds".format(assay_name, assay_end -

                                                assay_start))

    # Now, write out the results csv, going line by line through all molecule results

    assay_results_len = len(assay_results)

    all_assay_end = time.time()

    print("Parsed all assays in: {} hours".format(

        (all_assay_end - all_assay_start) / 3600))

    print("Parsed all assays in: {} hours".format((

        all_assay_end - all_assay_start) / 3600))

    smiles_start = time.time()

    os.remove(path_final)

    smiles_end = time.time()

    print("Smiles joined and gzip in: {} hours".format(

        (smiles_end - smiles_start) / 3600))

    print("Smiles joined and gzip in: {} hours".format((

        smiles_end - smiles_start) / 3600))

    print("Finished creating dataset: {} in: {} hours".format(

        file_name, (smiles_end - all_assay_start) / 3600))

  overall_end = time.time()

  secs_elapsed = overall_end - overall_start

  print(

      "Parsed all smiles in: {0} seconds, or {1} minutes, or {2} hours".format(

          secs_elapsed, secs_elapsed / 60, secs_elapsed / 3600))

  print("Parsed all smiles in: {0} seconds, or {1} minutes, or {2} hours".

        format(secs_elapsed, secs_elapsed / 60, secs_elapsed / 3600))

  print("Total length of: {}".format(len(keys)))

  with open(os.path.join(data_dir, "/pubchemsmiles_tuple.pickle"), "wb") as f:

    pickle.dump((keys, values), f)

  data_dir = deepchem.utils.get_data_dir()

  if reload:

    save_dir = os.path.join(

        data_dir,

        assay_file_name.split(".")[0] + featurizer + "/" + split)

        data_dir, assay_file_name.split(".")[0] + featurizer + "/" + split)

  dataset_file = os.path.join(data_dir, assay_file_name)