{"id":20560070,"url":"https://github.com/cannylab/scones","last_synced_at":"2025-09-15T09:40:50.908Z","repository":{"id":78184828,"uuid":"356106372","full_name":"CannyLab/scones","owner":"CannyLab","description":"Source Code for \"Scones: Towards Conversational Authoring of Sketches\"","archived":false,"fork":false,"pushed_at":"2021-04-09T20:21:01.000Z","size":19,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":6,"default_branch":"main","last_synced_at":"2025-01-16T19:04:33.285Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"bsd-3-clause","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/CannyLab.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2021-04-09T02:15:02.000Z","updated_at":"2021-08-08T00:46:45.000Z","dependencies_parsed_at":"2023-03-01T11:00:46.139Z","dependency_job_id":null,"html_url":"https://github.com/CannyLab/scones","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/CannyLab%2Fscones","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/CannyLab%2Fscones/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/CannyLab%2Fscones/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/CannyLab%2Fscones/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/CannyLab","download_url":"https://codeload.github.com/CannyLab/scones/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":242168255,"owners_count":20083048,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-11-16T03:53:09.364Z","updated_at":"2025-03-06T07:42:44.868Z","avatar_url":"https://github.com/CannyLab.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Source Code for \"Scones: Towards Conversational Authoring of Sketches\"\n## Paper\nPlease find our paper [here](https://dl.acm.org/doi/abs/10.1145/3377325.3377485).\n\n## Composition Proposer\n\n### Prerequisites\nTo train new models and/or run inference on published pre-trained checkpoints for the composition proposer, you will need the following prerequisites:\n- Tensorflow 2.3.1\n- Tensorflow Estimator 2.3.0\n- Huggingface Transformers 3.1.0\n- nltk 3.5\n- tqdm 4.50.2\n- numpy 1.18.5\n- Data and Pre-trained GLoVe embedding listed in the section below\n\n### Data\nFor Training/Eval only: Scones is trained on CoDraw data, which can be downloaded in JSON format from [here](https://drive.google.com/file/d/0B-u9nH58139bTy1XRFdqaVEzUGs/view). Please download the json file and move it into the *data/* folder at the root of the repo\n\nFor Training/Eval/Prediction only (generating new scenes based no captions): Scones preprocesses text tokens into GLoVe vector. We use 300-d GLoVe vectors trained on Common Crawl with 42B tokens. Please download the file *glove.42B.300d.zip* from [here](https://nlp.stanford.edu/projects/glove/) and extract the file *glove.42B.300d.txt* into the *data/* folder.\n\n### Pretrained Model\nA new pretrained model that uses the huggingface GPT-2 implementation can be downloaded from [here](https://drive.google.com/file/d/1Anny8fyV46jwnXgiJ4YveR0HvcvAdWUP/view?usp=sharing). This model achieves **3.53** for the similarity metric on CoDraw dataset's test set. \n\n### Training\nTo train the model, simply run\n`python train_state.py`\n\n### Eval\nTo run evaluation on the test set, change the *model_ckpt* variable in *run_eval.py* to the desired checkpoint path. Then, run:\n`python run_eval.py`\n\n### Generation\nComing soon.\n\n## Object Sketchers\nComing soon.","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcannylab%2Fscones","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fcannylab%2Fscones","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcannylab%2Fscones/lists"}