U Berkeley, Mitchell Wortsman, Ludwig Schmidt, Christoph Schuhmann, Romain Beaumont, Robert Kaczmarczyk, Clayton Mullis, Theo Coombes, Jenia Jitsev, Patrick Schramowski, Vencu, Ade Gordon, and 9 more
We lifted 18 functions out of this paper's own repositories and ran 14 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| rvencu/crawlingathome-gpu-hcloud | canonical | 7 of 8 |
| rom1504/clip-retrieval | canonical | 4 of 7 |
| copy not recorded | — | 3 of 3 |
| Function | Status | Where it lives |
|---|---|---|
| config | Ran | rvencu/crawlingathome-gpu-hcloud/gpu_inference.py code served (permissive licence) · get_code("6b0f6ffaf5450f20") |
| config | Ran | rvencu/crawlingathome-gpu-hcloud/ccpp.py code served (permissive licence) · get_code("19fb2294a9c2a40c") |
| convert_to_custom_text_state_dict | Ran | this paper's copy was not recorded; identical code first harvested from deep-real/dcp pointer only · get_code("f19962ebb134b3d7") |
| count_samples | Ran | rom1504/clip-retrieval/clip_retrieval/clip_back_prepro/parquet_to_arrow.py code served (permissive licence) · get_code("5857f996a53e3d10") |
| df_clipfilter | Ran | rvencu/crawlingathome-gpu-hcloud/gpu_inference.py code served (permissive licence) · get_code("aad8a39d981e52e6") |
| file_to_count | Ran | rom1504/clip-retrieval/clip_retrieval/clip_back_prepro/parquet_to_arrow.py code served (permissive licence) · get_code("54a08251b7e05489") |
| filter | Ran | rvencu/crawlingathome-gpu-hcloud/gpu_inference.py code served (permissive licence) · get_code("0f5ca6c21457febb") |
| get_cast_dtype | Ran | this paper's copy was not recorded; identical code first harvested from zsxkib/PuLID pointer only · get_code("dcd422d66b0581d8") |
| get_input_dtype | Ran | this paper's copy was not recorded; identical code first harvested from deep-real/dcp pointer only · get_code("b476c8cfbf0f1f47") |
| metric_to_average | Ran | rom1504/clip-retrieval/clip_retrieval/clip_back.py code served (permissive licence) · get_code("42824ddb4bfbec48") |
| move_files_to_folders | Ran | rvencu/crawlingathome-gpu-hcloud/postCLIP_staging/movefiles.py code served (permissive licence) · get_code("a824d70f7658bab0") |
| normalized | Ran | rom1504/clip-retrieval/clip_retrieval/clip_back.py code served (permissive licence) · get_code("1472ce0e9f26f1c7") |
| process_img_content | Ran | rvencu/crawlingathome-gpu-hcloud/dbdl.py code served (permissive licence) · get_code("68deb34ae97e2fa1") |
| remove_bad_chars | Ran | rvencu/crawlingathome-gpu-hcloud/ccpp.py code served (permissive licence) · get_code("a487a6f248dc7148") |
| create_file_list | Not yet run | rvencu/crawlingathome-gpu-hcloud/postCLIP_staging/movefiles.py code served (permissive licence) · get_code("e47c15572e3af93b") |
| download_image | Not yet run | rom1504/clip-retrieval/clip_retrieval/clip_back.py code served (permissive licence) · get_code("ef049563c3046681") |
| get_old_to_new_mapping | Not yet run | rom1504/clip-retrieval/clip_retrieval/ivf_metadata_ordering.py code served (permissive licence) · get_code("5261e2f4314f6275") |
| search_to_new_ids | Not yet run | rom1504/clip-retrieval/clip_retrieval/ivf_metadata_ordering.py code served (permissive licence) · get_code("c315c90997b8fc3c") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Groundbreaking language-vision architectures like CLIP and DALL-E proved the utility of training on large amounts of noisy image-text data, without relying on expensive accurate labels used in standard vision unimodal supervised learning. The resulting models showed capabilities of strong text-guided image generation and transfer to downstream tasks, while performing remarkably at zero-shot classification with noteworthy out-of-distribution robustness. Since then, large-scale language-vision models like ALIGN, BASIC, GLIDE, Flamingo and Imagen made further improvements. Studying the training and capabilities of such models requires datasets containing billions of image-text pairs. Until now, no datasets of this size have been made openly available for the broader research community. To address this problem and democratize research on large-scale multi-modal models, we present LAION-5B -a dataset consisting of 5.85 billion CLIP-filtered image-text pairs, of which 2.32B contain English language. We show successful replication and fine-tuning of foundational models like CLIP, GLIDE and Stable Diffusion using the dataset, and discuss further experiments enabled with an openly available dataset of this scale. Additionally we provide several nearest neighbor indices, an improved web-interface for dataset exploration and subset generation, and detection scores for watermark, NSFW, and toxic content detection. 1
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2210.08402")
get_code_for_paper("2210.08402")
have("2210.08402")
Connect an agent — have() is free.