MCPcopy Create free account

hub / github.com/huggingface/smollm / functions

Functions1,907 in github.com/huggingface/smollm

⨍ Functions 1,907◇ Types & classes 810↳ Endpoints 7
↓ 1 callersFunctionlora_unload
(model)
vision/m4/training/utils.py:1023
↓ 1 callersFunctionmain
()
tools/smolvlm_local_inference/SmolVLM_video_inference.py:133
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/01_tar_datasets_with_jpeg/python_scripts/06_convert_lrv_per_shard_idx.py:33
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/01_tar_datasets_with_jpeg/python_scripts/12_convert_llavar_per_shard_idx.py:33
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/01_tar_datasets_with_jpeg/python_scripts/07_convert_llava_per_shard_idx.py:33
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/01_tar_datasets_with_jpeg/python_scripts/19_convert_websight_mix_per_shard_idx.py:36
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/01_tar_datasets_with_jpeg/python_scripts/13_convert_vqav2_task_finetuning_per_shard_idx.py:33
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/01_tar_datasets_with_jpeg/python_scripts/17_convert_websight_v02_per_shard_idx.py:33
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/01_tar_datasets_with_jpeg/python_scripts/02_convert_cm4_per_shard_idx.py:34
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/01_tar_datasets_with_jpeg/python_scripts/09_convert_ultrachat_per_shard_idx.py:33
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/01_tar_datasets_with_jpeg/python_scripts/16_convert_image_website_code_per_shard_idx.py:33
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/01_tar_datasets_with_jpeg/python_scripts/01_convert_coco_per_shard_idx.py:34
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/01_tar_datasets_with_jpeg/python_scripts/08_convert_svit_per_shard_idx.py:33
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/01_tar_datasets_with_jpeg/python_scripts/10_convert_m3it_per_shard_idx.py:33
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/01_tar_datasets_with_jpeg/python_scripts/18_convert_sft_per_shard_idx.py:36
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/01_tar_datasets_with_jpeg/python_scripts/03_convert_laoin_per_shard_idx.py:36
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/01_tar_datasets_with_jpeg/python_scripts/15_convert_docvqa_per_shard_idx.py:33
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/01_tar_datasets_with_jpeg/python_scripts/04_convert_cm4_per_shard_idx.py:36
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/01_tar_datasets_with_jpeg/python_scripts/14_convert_tikz_per_shard_idx.py:33
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/01_tar_datasets_with_jpeg/python_scripts/11_convert_spot_difference_per_shard_idx.py:33
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/01_tar_datasets_with_jpeg/python_scripts/05_convert_scaled_laion_per_shard_idx.py:36
↓ 1 callersFunctionmain
()
vision/data/datasets_processing_scripts/clean_m4_prelimenary_experiments/python_scripts/02_add_html_back.py:56
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/clean_m4_prelimenary_experiments/python_scripts/04_get_banned_urls.py:54
↓ 1 callersFunctionmain
()
vision/data/datasets_processing_scripts/clean_m4_prelimenary_experiments/python_scripts/05_filter_cm4.py:56
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/clean_m4_prelimenary_experiments/python_scripts/03_clean_v2.py:53
↓ 1 callersFunctionmain
(args)
vision/data/datasets_processing_scripts/enwiki/python_scripts/01_extract_text_and_urls_from_wikipedia_web_documents.py:51
↓ 1 callersFunctionmain
()
vision/data/datasets_processing_scripts/build_concatenation_datasets_sft/viz_tool.py:35
↓ 1 callersFunctionmain
()
vision/experiments/pretraining/vloom/tr_343_smolvlm_05b_1st_stage/resize_embed_for_eou.py:16
↓ 1 callersFunctionmain
()
vision/experiments/pretraining/vloom/tr_341_smolvlm_025b_1st_stage/resize_embed_for_eou.py:16
↓ 1 callersFunctionmain
()
vision/experiments/pretraining/vloom/tr_347_smolvlm_500M_2nd_stage/resize_embed_for_eou.py:16
↓ 1 callersFunctionmain
()
vision/experiments/pretraining/vloom/tr_348_smolvlm_2B/resize_embed_for_eou.py:16
↓ 1 callersFunctionmain
()
vision/experiments/pretraining/vloom/tr_345_vsmollm2_256M_2nd_stage/resize_embed_for_eou.py:16
↓ 1 callersFunctionmain
()
vision/experiments/pretraining/vloom/tr_349_vsmollm2_500M_3rd_stage/resize_embed_for_eou.py:16
↓ 1 callersFunctionmain
()
vision/experiments/pretraining/vloom/tr_350_smolvlm_2B_2nd_stage/resize_embed_for_eou.py:16
↓ 1 callersFunctionmain
()
vision/experiments/pretraining/vloom/tr_346_vsmollm2_256M_3rd_stage/resize_embed_for_eou.py:16
↓ 1 callersFunctionmain
()
vision/smolvlm2/scripts/create_mixture.py:172
↓ 1 callersFunctionmain
()
vision/smolvlm2/scripts/sort.py:116
↓ 1 callersFunctionmain
()
vision/m4/utils/datasets/get_self_contained_ds.py:88
↓ 1 callersFunctionmain
()
vision/m4/scripts/s3-upload-checkpoints.py:124
↓ 1 callersFunctionmain
()
vision/m4/scripts/resize_embed_for_eou.py:18
↓ 1 callersFunctionmain
()
vision/m4/scripts/convert-checkpoints.py:111
↓ 1 callersFunctionmain
()
vision/m4/scripts/s3_checkpoint_download_convert_upload.py:99
↓ 1 callersFunctionmain
()
vision/m4/scripts/schedule-evals.py:75
↓ 1 callersFunctionmain
()
vision/m4/scripts/merge_lora_and_save.py:28
↓ 1 callersFunctionmain
()
vision/m4/scripts/cleanup-checkpoints.py:110
↓ 1 callersFunctionmain
(args)
vision/m4/evaluation/launch.py:43
↓ 1 callersFunctionmain
()
vision/m4/evaluation/generation/generate.py:182
↓ 1 callersFunctionmain
()
vision/m4/evaluation/generation/deprecated_generation/log_generation.py:62
↓ 1 callersFunctionmain
()
vision/m4/evaluation/generation/deprecated_generation/launch_generation.py:16
↓ 1 callersFunctionmain
()
vision/m4/evaluation/scripts/sync_evaluations_on_wandb.py:36
↓ 1 callersFunctionmain
()
vision/m4/evaluation/scripts/mmmu_submission_format.py:36
↓ 1 callersFunctionmain
()
vision/m4/evaluation/scripts/mmbench_submission_format.py:62
↓ 1 callersFunctionmain
()
vision/m4/evaluation/scripts/create_sample_evaluation_datasets.py:35
↓ 1 callersFunctionmain
()
vision/m4/evaluation/scripts/visualize_generations.py:131
↓ 1 callersFunctionmain
(args)
text/finetuning/train.py:54
↓ 1 callersFunctionmain
(args)
text/data/fineweb-edu/run_edu_bert.py:7
↓ 1 callersFunctionmain
(args)
text/data/fineweb-edu/train_edu_bert.py:47
↓ 1 callersFunctionmake_question_answer_visual7w
(question, multiple_choices, answer)
vision/data/datasets_processing_scripts/build_concatenation_datasets_sft/build_ds_sft.py:469
↓ 1 callersFunctionmake_supervised_data_module
Creates train_dataset, eval_dataset, and data_collator. If data_args.packed => we do integer-coded subseq_ids approach, else => normal ap
vision/smolvlm2/smolvlm/datasets/builder.py:459
↓ 1 callersMethodmake_tree
(self, selectolax_node, page_url, path_in_tree=[])
vision/m4/sourcing/data_collection/processors/pre_extraction_simplificator.py:38
↓ 1 callersFunctionmerge_consecutive_END_OF_DOCUMENT_TOKEN_TO_BE_REPLACED
(texts, images, metadata)
vision/data/datasets_processing_scripts/build_webdocs_dataset/python_scripts/13_final_processing.py:144
↓ 1 callersFunctionmodel_generation
( prompts, all_images, all_image_paths, tokenizer, model, accelerator, image_size,
vision/m4/evaluation/generation/generate.py:74
↓ 1 callersFunctionngram_conv
(ids: np.ndarray)
vision/m4/sourcing/processing/app.py:25
↓ 1 callersMethodnormalization
( text, remove_non_printing_characters, strip, lower_case, standardize
vision/m4/sourcing/data_collection/processors/web_document_filtering.py:368
↓ 1 callersFunctionnormalized_levenshtein
(s1, s2)
vision/m4/evaluation/custom_metrics/doc_vqa_metrics.py:26
↓ 1 callersFunctionopt_in_out_task
(image_urls: List[str], session, semaphore, limiter)
vision/data/datasets_processing_scripts/build_laion_coco_dataset/python_scripts/02_01_find_opt_out.py:59
↓ 1 callersFunctionopt_in_out_task
(image_urls: List[str], session, semaphore, limiter)
vision/data/datasets_processing_scripts/build_laion_dataset/python_scripts/04_03_find_opt_out_images_laion.py:57
↓ 1 callersFunctionopt_in_out_task
(image_urls: List[str], session, semaphore, limiter)
vision/data/datasets_processing_scripts/build_webdocs_dataset/python_scripts/12_01_find_opt_out_images.py:59
↓ 1 callersFunctionopt_in_out_task
(image_urls: List[str], session, semaphore, limiter)
vision/data/datasets_processing_scripts/build_webdocs_dataset/python_scripts/15_02_find_opt_out.py:59
↓ 1 callersMethodoverlap_train_eval_deduplication
(self, reload_files=False, reload_to_be_deduplicated=False)
vision/m4/sourcing/data_collection/processors/web_document_image_deduplication.py:174
↓ 1 callersMethodoverride_vision_model_wrapper
(cls, model, config, vision_model_name, vision_model_params, torch_dtype)
vision/m4/models/vgpt2/modeling_vgpt2.py:593
↓ 1 callersFunctionparallel_opt_in_out_task
(urls: list)
vision/data/datasets_processing_scripts/build_laion_coco_dataset/python_scripts/02_01_find_opt_out.py:66
↓ 1 callersFunctionparallel_opt_in_out_task
(urls: list)
vision/data/datasets_processing_scripts/build_laion_dataset/python_scripts/04_03_find_opt_out_images_laion.py:64
↓ 1 callersFunctionparallel_opt_in_out_task
(urls: list)
vision/data/datasets_processing_scripts/build_webdocs_dataset/python_scripts/12_01_find_opt_out_images.py:66
↓ 1 callersFunctionparallel_opt_in_out_task
(urls: list)
vision/data/datasets_processing_scripts/build_webdocs_dataset/python_scripts/15_02_find_opt_out.py:66
↓ 1 callersMethodparallelize
(self, device_map=None)
vision/m4/models/vgpt2/modeling_vgpt2.py:786
↓ 1 callersFunctionparse
(e)
vision/data/datasets_processing_scripts/build_concatenation_datasets_sft/build_ds_sft.py:3986
↓ 1 callersMethodparse
(cls)
vision/m4/evaluation/generation/deprecated_generation/log_generation.py:24
↓ 1 callersFunctionparse_open_response_mmmu
Parse the prediction from the generated response. Return a list of predicted strings or numbers
vision/m4/evaluation/custom_metrics/utils.py:466
↓ 1 callersFunctionpick_video_reader
Pick decord if installed, otherwise torchvision.
vision/smolvlm2/smolvlm/mm_utils.py:155
↓ 1 callersMethodpre_download_image_map
(self, batch: Dict)
vision/m4/sourcing/pmd/loader_builder.py:128
↓ 1 callersMethodpredict
(self, **kwargs)
vision/m4/models/vgpt2/evaluation_classification_in_context_vgpt2.py:437
↓ 1 callersMethodprepare_dataset
Prepare batch of examples.
vision/m4/models/vgpt2/evaluation_image_caption_matching_vgpt2.py:82
↓ 1 callersMethodprepare_dataset
Prepare batch of examples. Each example (X, y) where y is among (y1, y2, ..., yN) - the labels options - is turned into [(X,
vision/m4/models/vgpt2/evaluation_classification_in_context_vgpt2.py:95
↓ 1 callersMethodprepare_dataset
Prepare batch of examples.
vision/m4/models/vgpt2/evaluation_captioning_in_context_vgpt2.py:143
↓ 1 callersMethodprepare_dataset
Prepare batch of examples.
vision/m4/models/vgpt2/evaluation_perplexity_in_context_vgpt2.py:250
↓ 1 callersMethodprepare_dataset
Prepare batch of examples.
vision/m4/models/vgpt2/evaluation_open_ended_vqa_in_context_vgpt2.py:188
↓ 1 callersMethodprepare_image_caption_pair_ds
(self, exs: Dict)
vision/m4/models/vgpt2/evaluation_perplexity_in_context_vgpt2.py:80
↓ 1 callersFunctionprepare_inputs_for_generation
(input_ids, past_key_values=None, **kwargs)
vision/m4/models/vmistral/modeling_vmistral.py:222
↓ 1 callersFunctionprepare_math_vista_mcq
(split)
vision/data/datasets_processing_scripts/create_evaluation_datasets/create_math_vista_mcq.py:61
↓ 1 callersFunctionprepare_math_vista_openended
(split)
vision/data/datasets_processing_scripts/create_evaluation_datasets/create_math_vista_open_ended.py:57
↓ 1 callersFunctionprepare_mmmu_mcq
(split)
vision/data/datasets_processing_scripts/create_evaluation_datasets/create_mmmu_mcq.py:83
↓ 1 callersFunctionprepare_mmmu_openended
(split)
vision/data/datasets_processing_scripts/create_evaluation_datasets/create_mmmu_open_ended.py:72
↓ 1 callersFunctionprepare_model
Loads and configures the smolVLM model (Idefics3ForConditionalGeneration), applying rope scaling if needed, plus optional bitsandbytes quant
vision/smolvlm2/smolvlm/train/train.py:153
↓ 1 callersFunctionprepare_model
(model, accelerator)
vision/m4/evaluation/generation/generate.py:55
↓ 1 callersMethodprepare_webdoc_ds
(self, exs: Dict)
vision/m4/models/vgpt2/evaluation_perplexity_in_context_vgpt2.py:127
↓ 1 callersMethodprint_stats
(self)
vision/data/datasets_processing_scripts/build_concatenation_datasets_sft/merge_on_image_individual_dataset.py:135
↓ 1 callersMethodprocessDigitArticle
(self, in_text)
vision/m4/evaluation/custom_metrics/utils.py:358
↓ 1 callersMethodprocessPunctuation
(self, in_text)
vision/m4/evaluation/custom_metrics/utils.py:348
↓ 1 callersFunctionprocess_docs
(documents: List[str], subshard_idx: int)
vision/m4/sourcing/processing/extracting_ngrams/extract_documents_ngrams.py:41
← previousnext →801–900 of 1,907, ranked by callers