{ "id": 403321, "title": "[Compilation] Good Starter Notebooks", "url": "/competitions/benetech-making-graphs-accessible/discussion/403321", "author_name": "", "post_date": "2023-04-22T12:38:55.702419900Z", "votes": 10, "comment_count": 1, "views": 0, "content": "
    \n

  1. Classification Task - Inference by 2bibanh. Their contribution to this competition includes a few other notebooks.

    2. \n

  2. Simple EDA and flattening JSON data with pandas by @kimtaehun

    3. \n

  3. Benetech | Simple OCR Network by @alejopaullier

    4. \n

  4. Spell check on tick labels using BERT by @eleonoraricci

    5. \n

  5. Pix2Struct PyTorch + AMP + W&B by @heyytanay

    6. \n

  6. MGA ~ Graphs with interactive annotations by @leonidkulyk. Quite a few high scoring notebooks in this competition are based on this one.

    7. \n

  7. Trying-Out-DePlot by @gomaki

    8. \n

  8. Let's Plot Charts by @mpwolke

    9. \n

  9. Predict plot-bb coordinates by @yoshikuwano

    10. \n

  10. Start Here - Challenge Benetech by @durvalslompojunior

    11. \n

  11. Tensorflow full pipeline by @tasnimnishatislam

    12. \n

  12. DotSolver: A Rule-based Solution for Dot Plot by @awqatak

    13. \n

  13. Benetech: (Vertical + Horizontal) Bar Generator by @brendanartley

    14. \n
\n

This is a unique competition, so happy learning. One question that plagues me is, do we even need machine learning to do this?
\nDo let me know your thoughts.
\nAll due credits to the original contributors. Please upvote their notebooks if you find it useful.

\n

BONUS : The details about competition metric can be found here. Thanks @ryanholbrook

\n

BONUS 2 : Just found out ChartGPT, which can be very helpful in this competition.

\n

Thanks

", "messages": [ { "id": "2230504", "postDate": "04/22/2023 12:38:55", "content": "
    \n

  1. Classification Task - Inference by 2bibanh. Their contribution to this competition includes a few other notebooks.

    2. \n

  2. Simple EDA and flattening JSON data with pandas by @kimtaehun

    3. \n

  3. Benetech | Simple OCR Network by @alejopaullier

    4. \n

  4. Spell check on tick labels using BERT by @eleonoraricci

    5. \n

  5. Pix2Struct PyTorch + AMP + W&B by @heyytanay

    6. \n

  6. MGA ~ Graphs with interactive annotations by @leonidkulyk. Quite a few high scoring notebooks in this competition are based on this one.

    7. \n

  7. Trying-Out-DePlot by @gomaki

    8. \n

  8. Let's Plot Charts by @mpwolke

    9. \n

  9. Predict plot-bb coordinates by @yoshikuwano

    10. \n

  10. Start Here - Challenge Benetech by @durvalslompojunior

    11. \n

  11. Tensorflow full pipeline by @tasnimnishatislam

    12. \n

  12. DotSolver: A Rule-based Solution for Dot Plot by @awqatak

    13. \n

  13. Benetech: (Vertical + Horizontal) Bar Generator by @brendanartley

    14. \n
\n

This is a unique competition, so happy learning. One question that plagues me is, do we even need machine learning to do this?
\nDo let me know your thoughts.
\nAll due credits to the original contributors. Please upvote their notebooks if you find it useful.

\n

BONUS : The details about competition metric can be found here. Thanks @ryanholbrook

\n

BONUS 2 : Just found out ChartGPT, which can be very helpful in this competition.

\n

Thanks

", "rawMarkdown": "1. [Classification Task - Inference](https://www.kaggle.com/code/bibanh/simple-submission-classification-task-inference) by 2bibanh. Their contribution to this competition includes a few other notebooks.\n\n2. [Simple EDA and flattening JSON data with pandas](https://www.kaggle.com/code/kimtaehun/simple-eda-and-flattening-json-data-with-pandas) by @kimtaehun\n\n3. [Benetech | Simple OCR Network](https://www.kaggle.com/code/alejopaullier/benetech-simple-ocr-network) by @alejopaullier\n\n4. [Spell check on tick labels using BERT](https://www.kaggle.com/code/eleonoraricci/spell-check-on-tick-labels-using-bert) by @eleonoraricci\n\n5. [Pix2Struct PyTorch + AMP + W&B](https://www.kaggle.com/code/heyytanay/training-pix2struct-pytorch-amp-w-b) by @heyytanay\n\n6. [MGA ~ Graphs with interactive annotations](https://www.kaggle.com/code/leonidkulyk/eda-mga-graphs-with-interactive-annotations) by @leonidkulyk. Quite a few high scoring notebooks in this competition are based on this one.\n\n7. [Trying-Out-DePlot](https://www.kaggle.com/code/gomaki/deplot-trying-out) by @gomaki\n\n8. [Let's Plot Charts](https://www.kaggle.com/code/mpwolke/let-s-plot-charts/notebook) by @mpwolke\n\n9. [Predict plot-bb coordinates](https://www.kaggle.com/code/yoshikuwano/predict-plot-bb-coordinates) by @yoshikuwano\n\n10. [Start Here - Challenge Benetech](https://www.kaggle.com/code/durvalslompojunior/start-here-challenge-benetech) by @durvalslompojunior\n\n11. [Tensorflow full pipeline](https://www.kaggle.com/code/tasnimnishatislam/tensorflow-full-pipeline) by @tasnimnishatislam\n\n12. [DotSolver: A Rule-based Solution for Dot Plot](https://www.kaggle.com/code/awqatak/dotsolver-a-rule-based-solution-for-dot-plot) by @awqatak\n\n13. [Benetech: (Vertical + Horizontal) Bar Generator](https://www.kaggle.com/code/brendanartley/benetech-vertical-horizontal-bar-generator) by @brendanartley \n\n\nThis is a unique competition, so happy learning. One question that plagues me is, do we even need machine learning to do this?\nDo let me know your thoughts.\nAll due credits to the original contributors. Please upvote their notebooks if you find it useful.\n\n**BONUS** : The details about competition metric can be found [here](https://www.kaggle.com/code/ryanholbrook/competition-metric-benetech-mixed-match). Thanks @ryanholbrook\n\n**BONUS 2** : Just found out [ChartGPT](https://www.chartgpt.dev/), which can be very helpful in this competition.\n\nThanks", "votes": null }, { "id": "2230520", "postDate": "04/22/2023 13:00:51", "content": "

Thanks for the mention @kritidoneria!

", "rawMarkdown": "Thanks for the mention @kritidoneria!", "votes": null } ], "comments": [ { "id": 2230520, "author_name": "heyytanay", "author_url": "", "post_date": "04/22/2023 13:00:51", "content": "

Thanks for the mention @kritidoneria!

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "2230504": "1. [Classification Task - Inference](https://www.kaggle.com/code/bibanh/simple-submission-classification-task-inference) by 2bibanh. Their contribution to this competition includes a few other notebooks.\n\n2. [Simple EDA and flattening JSON data with pandas](https://www.kaggle.com/code/kimtaehun/simple-eda-and-flattening-json-data-with-pandas) by @kimtaehun\n\n3. [Benetech | Simple OCR Network](https://www.kaggle.com/code/alejopaullier/benetech-simple-ocr-network) by @alejopaullier\n\n4. [Spell check on tick labels using BERT](https://www.kaggle.com/code/eleonoraricci/spell-check-on-tick-labels-using-bert) by @eleonoraricci\n\n5. [Pix2Struct PyTorch + AMP + W&B](https://www.kaggle.com/code/heyytanay/training-pix2struct-pytorch-amp-w-b) by @heyytanay\n\n6. [MGA ~ Graphs with interactive annotations](https://www.kaggle.com/code/leonidkulyk/eda-mga-graphs-with-interactive-annotations) by @leonidkulyk. Quite a few high scoring notebooks in this competition are based on this one.\n\n7. [Trying-Out-DePlot](https://www.kaggle.com/code/gomaki/deplot-trying-out) by @gomaki\n\n8. [Let's Plot Charts](https://www.kaggle.com/code/mpwolke/let-s-plot-charts/notebook) by @mpwolke\n\n9. [Predict plot-bb coordinates](https://www.kaggle.com/code/yoshikuwano/predict-plot-bb-coordinates) by @yoshikuwano\n\n10. [Start Here - Challenge Benetech](https://www.kaggle.com/code/durvalslompojunior/start-here-challenge-benetech) by @durvalslompojunior\n\n11. [Tensorflow full pipeline](https://www.kaggle.com/code/tasnimnishatislam/tensorflow-full-pipeline) by @tasnimnishatislam\n\n12. [DotSolver: A Rule-based Solution for Dot Plot](https://www.kaggle.com/code/awqatak/dotsolver-a-rule-based-solution-for-dot-plot) by @awqatak\n\n13. [Benetech: (Vertical + Horizontal) Bar Generator](https://www.kaggle.com/code/brendanartley/benetech-vertical-horizontal-bar-generator) by @brendanartley \n\n\nThis is a unique competition, so happy learning. One question that plagues me is, do we even need machine learning to do this?\nDo let me know your thoughts.\nAll due credits to the original contributors. Please upvote their notebooks if you find it useful.\n\n**BONUS** : The details about competition metric can be found [here](https://www.kaggle.com/code/ryanholbrook/competition-metric-benetech-mixed-match). Thanks @ryanholbrook\n\n**BONUS 2** : Just found out [ChartGPT](https://www.chartgpt.dev/), which can be very helpful in this competition.\n\nThanks", "2230520": "Thanks for the mention @kritidoneria!" }, "source": "meta" }