parse_kaggle_2022_survey.py

import csv

groups = {
    "TensorFlow/Keras": ("Keras", "TensorFlow"),
    "PyTorch/PyTorch Lightning/Fast.ai": ("PyTorch", "PyTorch Lightning", "Fast.ai"),
}
counts = {name: 0 for name in groups}
total = 0
answer_indices = []
with open("kaggle_survey_2022_responses.csv", newline="\n") as f:
    reader = csv.reader(f, delimiter=",", quotechar='"')
    for row_i, row in enumerate(reader):
        if row_i == 0:
            for col_i in range(1, 16):
                answer_indices.append(row.index(f"Q17_{col_i}"))
        if row_i > 1:
            use_group = {name: False for name in groups}
            use_any = False
            for answer_i in answer_indices:
                answer = row[answer_i].strip()
                if answer and answer != "None":
                    use_any = True
                else:
                    continue
                use_other = True
                for name, group in groups.items():
                    if answer in group:
                        use_group[name] = True
                        use_other = False

                if use_other:
                    if answer not in counts:
                        counts[answer] = 0
                    counts[answer] += 1

            for name in use_group:
                if use_group[name]:
                    counts[name] += 1

            if use_any:
                total += 1

for k, v in sorted(counts.items(), key=lambda x: -x[1]):
    print(f"{k}: {100 * v/total :.2f} %")
print("-")
print(f"N={total}")