├── .gitignore ├── .gitmodules ├── CHANGE_LOG.md ├── LICENSE ├── MANIFEST.in ├── README.md ├── examples ├── bert │ ├── README.md │ ├── bert_ft_model.py │ ├── finetune_bert_adapter_boolq.py │ ├── finetune_bert_boolq.py │ ├── finetune_distill_boolq.py │ ├── inference_bert.py │ ├── inference_distill.py │ ├── scripts │ │ ├── finetune_adapter_boolq.sh │ │ ├── finetune_boolq.sh │ │ ├── finetune_distill_boolq.sh │ │ ├── inference_bert.sh │ │ └── inference_distill.sh │ └── transform_param.py ├── cait │ ├── README.md │ ├── cait_ft_model.py │ ├── config │ │ └── cait_config.py │ ├── finetune_vit_cifar10.py │ ├── scripts │ │ └── finetune_cifar10.sh │ └── transform_param.py ├── chatglm │ ├── .env │ ├── README.md │ ├── chat.py │ ├── chat_model.py │ ├── chat_sat.py │ ├── finetune_chatglm.py │ ├── inference_adgen.py │ ├── inference_chatglm.py │ ├── scripts │ │ ├── finetune_adgen_lora.sh │ │ └── finetune_adgen_ptuning.sh │ └── transform_param.py ├── chatglm2 │ ├── README.md │ ├── chat.py │ ├── chat_model.py │ ├── chat_sat.py │ ├── finetune_chatglm2.py │ ├── inference_adgen.py │ ├── scripts │ │ ├── finetune_adgen_lora.sh │ │ └── finetune_adgen_ptuning.sh │ ├── transform_param.py │ └── transform_param_newmlp.py ├── chatglm3 │ ├── README.md │ ├── chat_sat.py │ └── transform_param.py ├── clip │ ├── README.md │ ├── clip_finetune_model.py │ ├── config │ │ ├── clip_config.py │ │ ├── model_clip.sh │ │ └── model_clip_ftwp.sh │ ├── finetune_clip_cifar10.py │ ├── finetune_clip_cifar10wp.py │ ├── inference_clip.py │ ├── scripts │ │ ├── finetune_cifar10.sh │ │ ├── finetune_cifar10wp.sh │ │ └── inference_clip.sh │ ├── transform_param.py │ └── transform_param_new.py ├── cogview │ ├── inference_cogview.py │ ├── inference_cogview_caps.py │ ├── pretrain_gpt2.py │ └── scripts │ │ ├── ds_config.json │ │ ├── ds_config_zero.json │ │ ├── large_scale_select.sh │ │ ├── pretrain_multiple_nodes.sh │ │ └── text2image_cogview.sh ├── deit │ ├── README.md │ ├── config │ │ ├── deit_base_config.py │ │ ├── deit_small_config.py │ │ └── deit_tiny_config.py │ ├── finetune_vit_cifar10.py │ ├── scripts │ │ └── finetune_cifar10.sh │ ├── transform_param.py │ └── vit_ft_model.py ├── dpr │ ├── inference_dpr_ctx.py │ ├── inference_dpr_question_encoder.py │ ├── inference_dpr_reader.py │ └── scripts │ │ ├── inference_dpr_ctx.sh │ │ ├── inference_dpr_question_encoder.sh │ │ └── inference_dpr_reader.sh ├── eva2 │ ├── README.md │ ├── inference_eva2.py │ └── transform_param.py ├── eva2clip │ ├── README.md │ ├── transform_param.py │ └── transforms.py ├── glm-130b │ └── README.md ├── glm │ ├── config │ │ ├── model_glm_10B.sh │ │ ├── model_glm_10B_chinese.sh │ │ ├── model_glm_large_chinese.sh │ │ └── model_glm_roberta_large.sh │ ├── finetune_glm_boolq.py │ ├── finetune_glm_sst2.py │ ├── inference_glm.py │ └── scripts │ │ ├── ds_config_ft.json │ │ ├── finetune_boolq.sh │ │ ├── finetune_sst2.sh │ │ └── generate_glm.sh ├── glm4v │ ├── blip2_img_processor.py │ ├── chat_sat.py │ ├── chat_utils.py │ ├── chatglm4_chat.py │ └── cli_demo.py ├── gpt2 │ ├── README.md │ ├── inference_gpt2.py │ └── transform_param.py ├── gptneo │ ├── inference_gptneo.py │ └── scripts │ │ └── inference_gptneo.sh ├── llama │ ├── README.md │ ├── chat_sat.py │ ├── split_model.py │ └── transform_param.py ├── mae │ ├── README.md │ ├── config │ │ └── mae_config.py │ ├── finetune_mae_cifar10.py │ ├── inference_mae.ipynb │ ├── mae_finetune_model.py │ ├── models_mae.py │ ├── scripts │ │ └── finetune_cifar10.sh │ ├── transform_param.py │ └── util │ │ ├── crop.py │ │ ├── datasets.py │ │ ├── lars.py │ │ ├── lr_decay.py │ │ ├── lr_sched.py │ │ ├── misc.py │ │ └── pos_embed.py ├── mixtral-8x7b │ ├── chat_sat.py │ └── transform_param.py ├── roberta │ ├── README.md │ ├── config │ │ ├── model_roberta_base.sh │ │ └── model_roberta_large.sh │ ├── finetune_roberta_boolq.py │ ├── finetune_roberta_cb.py │ ├── finetune_roberta_copa.py │ ├── finetune_roberta_multirc.py │ ├── finetune_roberta_record.py │ ├── finetune_roberta_rte.py │ ├── finetune_roberta_wic.py │ ├── finetune_roberta_wsc.py │ ├── finetune_roberta_wsc_concat.py │ ├── inference_roberta.py │ ├── scripts │ │ ├── ds_config_ft.json │ │ ├── finetune_superglue.sh │ │ ├── finetune_superglue_multidevice.sh │ │ └── inference_roberta.sh │ └── transform_param.py ├── t5 │ ├── config │ │ ├── config_t5_large.json │ │ └── model_t5_large.sh │ ├── finetune_t5.py │ ├── inference_t5.py │ ├── scripts │ │ ├── config_t5_tmp.json │ │ ├── finetune_t5.sh │ │ └── generate_t5.sh │ └── test_t5.py ├── vit │ ├── README.md │ ├── config │ │ └── vit_base_config.py │ ├── finetune_vit_cifar10.py │ ├── scripts │ │ └── finetune_cifar10.sh │ ├── transform_param.py │ └── vit_ft_model.py └── yolos │ ├── README.md │ ├── config │ └── yolos_tiny_config.py │ ├── datasets_ │ ├── __init__.py │ ├── coco.py │ ├── coco_eval.py │ ├── transforms.py │ └── voc.py │ ├── infer_util.py │ ├── inference_yolos.py │ ├── models │ ├── __init__.py │ ├── backbone.py │ ├── detector.py │ ├── layers │ │ ├── __init__.py │ │ ├── drop.py │ │ ├── helper.py │ │ └── weight_init.py │ └── matcher.py │ ├── scripts │ └── finetune_coco.sh │ ├── train_yolos_coco.py │ ├── transform_param.py │ ├── util │ ├── __init__.py │ ├── box_ops.py │ ├── debug.py │ ├── misc.py │ ├── plot_utils.py │ └── scheduler.py │ └── visualization_new │ └── exp │ ├── gt_img.png │ ├── img.png │ └── pred_img.png ├── requirements.txt ├── sat ├── __init__.py ├── arguments.py ├── data_utils │ ├── __init__.py │ ├── configure_data.py │ ├── datasets.py │ ├── hf_dataset.py │ ├── jsonlds.py │ ├── samplers.py │ └── webds.py ├── generation │ ├── __init__.py │ ├── autoregressive_sampling.py │ ├── cuda2d_sampling.py │ ├── magnify.py │ ├── sampling_strategies │ │ ├── __init__.py │ │ ├── base_strategy.py │ │ ├── beam_search_strategy.py │ │ └── iterative_entfilter_strategy.py │ └── utils.py ├── helpers.py ├── model │ ├── __init__.py │ ├── attention │ │ ├── __init__.py │ │ └── memory_efficient_attention.py │ ├── base_model.py │ ├── cached_autoregressive_model.py │ ├── encoder_decoder_model.py │ ├── finetune │ │ ├── __init__.py │ │ ├── adapter.py │ │ ├── ffadd.py │ │ ├── lora.py │ │ ├── lora2.py │ │ ├── mlp_head.py │ │ └── prompt_tuning.py │ ├── mixins.py │ ├── official │ │ ├── __init__.py │ │ ├── bert_model.py │ │ ├── cait_model.py │ │ ├── chatglm2_model.py │ │ ├── chatglm3_model.py │ │ ├── chatglm4_model.py │ │ ├── chatglm_model.py │ │ ├── clip_model.py │ │ ├── cuda2d_model.py │ │ ├── distill_model.py │ │ ├── dpr_model.py │ │ ├── eva2_model.py │ │ ├── eva_clip_model.py │ │ ├── glm130B_model.py │ │ ├── glm4v_model.py │ │ ├── glm_model.py │ │ ├── gpt2_model.py │ │ ├── gptneo_model.py │ │ ├── llama_model.py │ │ ├── mae_model.py │ │ ├── mixtral_model.py │ │ ├── roberta_model.py │ │ ├── t5_model.py │ │ ├── vit_model.py │ │ └── yolos_model.py │ ├── position_embedding │ │ ├── __init__.py │ │ ├── rotary_embeddings.py │ │ ├── rotary_embeddings_original.py │ │ ├── sincos2d.py │ │ ├── triton_rotary.py │ │ ├── triton_rotary_embeddings.py │ │ └── vision_rotary_embeddings.py │ ├── registry.py │ └── transformer.py ├── mpu │ ├── __init__.py │ ├── cross_entropy.py │ ├── data.py │ ├── initialize.py │ ├── layers.py │ ├── mappings.py │ ├── operation.py │ └── utils.py ├── ops │ ├── __init__.py │ ├── csrc │ │ ├── adam │ │ │ ├── fused_ema_adam_frontend.cpp │ │ │ ├── multi_tensor_apply.cuh │ │ │ └── multi_tensor_ema_adam.cu │ │ └── includes │ │ │ ├── compat.h │ │ │ └── type_shim.h │ ├── fused_ema_adam.py │ ├── layernorm.py │ ├── local_attention_function.py │ ├── memory_efficient_attention.py │ ├── ops_builder │ │ ├── __init__.py │ │ ├── builder.py │ │ └── fused_ema_adam.py │ └── scaled_mask_softmax.py ├── quantization │ ├── __init__.py │ └── kernels.py ├── resources │ ├── __init__.py │ ├── download.py │ └── urls.py ├── tokenization │ ├── __init__.py │ ├── cogview │ │ ├── __init__.py │ │ ├── sp_tokenizer.py │ │ ├── templates.py │ │ ├── unified_tokenizer.py │ │ ├── vqvae │ │ │ ├── LICENSE │ │ │ ├── README.md │ │ │ ├── __init__.py │ │ │ ├── api.py │ │ │ ├── vqvae_diffusion.py │ │ │ └── vqvae_zc.py │ │ └── vqvae_tokenizer.py │ ├── embed_assets │ │ ├── chinese_sentencepiece │ │ │ ├── cog-pretrain.model │ │ │ └── cog-pretrain.vocab │ │ └── english_tokenizer │ │ │ ├── bert-base-uncased-vocab.txt │ │ │ ├── bert-large-uncased-vocab.txt │ │ │ ├── gpt2-merges.txt │ │ │ ├── gpt2-vocab.json │ │ │ ├── roberta-merges.txt │ │ │ └── roberta-vocab.json │ ├── glm │ │ ├── __init__.py │ │ ├── sp_tokenizer.py │ │ ├── tokenization.py │ │ ├── tokenization_gpt2.py │ │ └── tokenization_wordpiece.py │ ├── hf_tokenizer.py │ └── icetk_glm_130B │ │ ├── __init__.py │ │ ├── ice_tokenizer.py │ │ └── tokenizer.py ├── training │ ├── __init__.py │ ├── deepspeed_training.py │ ├── deepspeed_zero0.json │ ├── deepspeed_zero1.json │ ├── deepspeed_zero2.json │ ├── learning_rates.py │ ├── model_io.py │ └── utils.py └── transformer_defaults.py ├── setup.py ├── tests ├── deepspeed_test.json ├── sbatch_launch.sh ├── single_launch.sh ├── test_base_model.py ├── test_inference.py ├── test_jsonlds.py ├── test_list_info.py ├── test_mea.py ├── test_model_parallel.py ├── test_multi_webds.py ├── test_nested_model.py ├── test_read_img.py ├── test_remote_data.py ├── test_speed.py ├── test_train.py ├── test_train_dp.py ├── test_train_nested.py └── test_triton_rotary_embedding.py └── tutorials ├── inference.md ├── model_usage.md ├── speed_self_check.md └── training.md /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/.gitignore -------------------------------------------------------------------------------- /.gitmodules: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/.gitmodules -------------------------------------------------------------------------------- /CHANGE_LOG.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/CHANGE_LOG.md -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/LICENSE -------------------------------------------------------------------------------- /MANIFEST.in: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/MANIFEST.in -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/README.md -------------------------------------------------------------------------------- /examples/bert/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/bert/README.md -------------------------------------------------------------------------------- /examples/bert/bert_ft_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/bert/bert_ft_model.py -------------------------------------------------------------------------------- /examples/bert/finetune_bert_adapter_boolq.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/bert/finetune_bert_adapter_boolq.py -------------------------------------------------------------------------------- /examples/bert/finetune_bert_boolq.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/bert/finetune_bert_boolq.py -------------------------------------------------------------------------------- /examples/bert/finetune_distill_boolq.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/bert/finetune_distill_boolq.py -------------------------------------------------------------------------------- /examples/bert/inference_bert.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/bert/inference_bert.py -------------------------------------------------------------------------------- /examples/bert/inference_distill.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/bert/inference_distill.py -------------------------------------------------------------------------------- /examples/bert/scripts/finetune_adapter_boolq.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/bert/scripts/finetune_adapter_boolq.sh -------------------------------------------------------------------------------- /examples/bert/scripts/finetune_boolq.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/bert/scripts/finetune_boolq.sh -------------------------------------------------------------------------------- /examples/bert/scripts/finetune_distill_boolq.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/bert/scripts/finetune_distill_boolq.sh -------------------------------------------------------------------------------- /examples/bert/scripts/inference_bert.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/bert/scripts/inference_bert.sh -------------------------------------------------------------------------------- /examples/bert/scripts/inference_distill.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/bert/scripts/inference_distill.sh -------------------------------------------------------------------------------- /examples/bert/transform_param.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/bert/transform_param.py -------------------------------------------------------------------------------- /examples/cait/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/cait/README.md -------------------------------------------------------------------------------- /examples/cait/cait_ft_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/cait/cait_ft_model.py -------------------------------------------------------------------------------- /examples/cait/config/cait_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/cait/config/cait_config.py -------------------------------------------------------------------------------- /examples/cait/finetune_vit_cifar10.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/cait/finetune_vit_cifar10.py -------------------------------------------------------------------------------- /examples/cait/scripts/finetune_cifar10.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/cait/scripts/finetune_cifar10.sh -------------------------------------------------------------------------------- /examples/cait/transform_param.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/cait/transform_param.py -------------------------------------------------------------------------------- /examples/chatglm/.env: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm/.env -------------------------------------------------------------------------------- /examples/chatglm/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm/README.md -------------------------------------------------------------------------------- /examples/chatglm/chat.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm/chat.py -------------------------------------------------------------------------------- /examples/chatglm/chat_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm/chat_model.py -------------------------------------------------------------------------------- /examples/chatglm/chat_sat.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm/chat_sat.py -------------------------------------------------------------------------------- /examples/chatglm/finetune_chatglm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm/finetune_chatglm.py -------------------------------------------------------------------------------- /examples/chatglm/inference_adgen.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm/inference_adgen.py -------------------------------------------------------------------------------- /examples/chatglm/inference_chatglm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm/inference_chatglm.py -------------------------------------------------------------------------------- /examples/chatglm/scripts/finetune_adgen_lora.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm/scripts/finetune_adgen_lora.sh -------------------------------------------------------------------------------- /examples/chatglm/scripts/finetune_adgen_ptuning.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm/scripts/finetune_adgen_ptuning.sh -------------------------------------------------------------------------------- /examples/chatglm/transform_param.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm/transform_param.py -------------------------------------------------------------------------------- /examples/chatglm2/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm2/README.md -------------------------------------------------------------------------------- /examples/chatglm2/chat.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm2/chat.py -------------------------------------------------------------------------------- /examples/chatglm2/chat_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm2/chat_model.py -------------------------------------------------------------------------------- /examples/chatglm2/chat_sat.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm2/chat_sat.py -------------------------------------------------------------------------------- /examples/chatglm2/finetune_chatglm2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm2/finetune_chatglm2.py -------------------------------------------------------------------------------- /examples/chatglm2/inference_adgen.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm2/inference_adgen.py -------------------------------------------------------------------------------- /examples/chatglm2/scripts/finetune_adgen_lora.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm2/scripts/finetune_adgen_lora.sh -------------------------------------------------------------------------------- /examples/chatglm2/scripts/finetune_adgen_ptuning.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm2/scripts/finetune_adgen_ptuning.sh -------------------------------------------------------------------------------- /examples/chatglm2/transform_param.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm2/transform_param.py -------------------------------------------------------------------------------- /examples/chatglm2/transform_param_newmlp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm2/transform_param_newmlp.py -------------------------------------------------------------------------------- /examples/chatglm3/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm3/README.md -------------------------------------------------------------------------------- /examples/chatglm3/chat_sat.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm3/chat_sat.py -------------------------------------------------------------------------------- /examples/chatglm3/transform_param.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/chatglm3/transform_param.py -------------------------------------------------------------------------------- /examples/clip/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/clip/README.md -------------------------------------------------------------------------------- /examples/clip/clip_finetune_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/clip/clip_finetune_model.py -------------------------------------------------------------------------------- /examples/clip/config/clip_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/clip/config/clip_config.py -------------------------------------------------------------------------------- /examples/clip/config/model_clip.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/clip/config/model_clip.sh -------------------------------------------------------------------------------- /examples/clip/config/model_clip_ftwp.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/clip/config/model_clip_ftwp.sh -------------------------------------------------------------------------------- /examples/clip/finetune_clip_cifar10.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/clip/finetune_clip_cifar10.py -------------------------------------------------------------------------------- /examples/clip/finetune_clip_cifar10wp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/clip/finetune_clip_cifar10wp.py -------------------------------------------------------------------------------- /examples/clip/inference_clip.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/clip/inference_clip.py -------------------------------------------------------------------------------- /examples/clip/scripts/finetune_cifar10.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/clip/scripts/finetune_cifar10.sh -------------------------------------------------------------------------------- /examples/clip/scripts/finetune_cifar10wp.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/clip/scripts/finetune_cifar10wp.sh -------------------------------------------------------------------------------- /examples/clip/scripts/inference_clip.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/clip/scripts/inference_clip.sh -------------------------------------------------------------------------------- /examples/clip/transform_param.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/clip/transform_param.py -------------------------------------------------------------------------------- /examples/clip/transform_param_new.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/clip/transform_param_new.py -------------------------------------------------------------------------------- /examples/cogview/inference_cogview.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/cogview/inference_cogview.py -------------------------------------------------------------------------------- /examples/cogview/inference_cogview_caps.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/cogview/inference_cogview_caps.py -------------------------------------------------------------------------------- /examples/cogview/pretrain_gpt2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/cogview/pretrain_gpt2.py -------------------------------------------------------------------------------- /examples/cogview/scripts/ds_config.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/cogview/scripts/ds_config.json -------------------------------------------------------------------------------- /examples/cogview/scripts/ds_config_zero.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/cogview/scripts/ds_config_zero.json -------------------------------------------------------------------------------- /examples/cogview/scripts/large_scale_select.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/cogview/scripts/large_scale_select.sh -------------------------------------------------------------------------------- /examples/cogview/scripts/pretrain_multiple_nodes.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/cogview/scripts/pretrain_multiple_nodes.sh -------------------------------------------------------------------------------- /examples/cogview/scripts/text2image_cogview.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/cogview/scripts/text2image_cogview.sh -------------------------------------------------------------------------------- /examples/deit/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/deit/README.md -------------------------------------------------------------------------------- /examples/deit/config/deit_base_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/deit/config/deit_base_config.py -------------------------------------------------------------------------------- /examples/deit/config/deit_small_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/deit/config/deit_small_config.py -------------------------------------------------------------------------------- /examples/deit/config/deit_tiny_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/deit/config/deit_tiny_config.py -------------------------------------------------------------------------------- /examples/deit/finetune_vit_cifar10.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/deit/finetune_vit_cifar10.py -------------------------------------------------------------------------------- /examples/deit/scripts/finetune_cifar10.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/deit/scripts/finetune_cifar10.sh -------------------------------------------------------------------------------- /examples/deit/transform_param.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/deit/transform_param.py -------------------------------------------------------------------------------- /examples/deit/vit_ft_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/deit/vit_ft_model.py -------------------------------------------------------------------------------- /examples/dpr/inference_dpr_ctx.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/dpr/inference_dpr_ctx.py -------------------------------------------------------------------------------- /examples/dpr/inference_dpr_question_encoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/dpr/inference_dpr_question_encoder.py -------------------------------------------------------------------------------- /examples/dpr/inference_dpr_reader.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/dpr/inference_dpr_reader.py -------------------------------------------------------------------------------- /examples/dpr/scripts/inference_dpr_ctx.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/dpr/scripts/inference_dpr_ctx.sh -------------------------------------------------------------------------------- /examples/dpr/scripts/inference_dpr_question_encoder.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/dpr/scripts/inference_dpr_question_encoder.sh -------------------------------------------------------------------------------- /examples/dpr/scripts/inference_dpr_reader.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/dpr/scripts/inference_dpr_reader.sh -------------------------------------------------------------------------------- /examples/eva2/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/eva2/README.md -------------------------------------------------------------------------------- /examples/eva2/inference_eva2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/eva2/inference_eva2.py -------------------------------------------------------------------------------- /examples/eva2/transform_param.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/eva2/transform_param.py -------------------------------------------------------------------------------- /examples/eva2clip/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/eva2clip/README.md -------------------------------------------------------------------------------- /examples/eva2clip/transform_param.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/eva2clip/transform_param.py -------------------------------------------------------------------------------- /examples/eva2clip/transforms.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/eva2clip/transforms.py -------------------------------------------------------------------------------- /examples/glm-130b/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/glm-130b/README.md -------------------------------------------------------------------------------- /examples/glm/config/model_glm_10B.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/glm/config/model_glm_10B.sh -------------------------------------------------------------------------------- /examples/glm/config/model_glm_10B_chinese.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/glm/config/model_glm_10B_chinese.sh -------------------------------------------------------------------------------- /examples/glm/config/model_glm_large_chinese.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/glm/config/model_glm_large_chinese.sh -------------------------------------------------------------------------------- /examples/glm/config/model_glm_roberta_large.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/glm/config/model_glm_roberta_large.sh -------------------------------------------------------------------------------- /examples/glm/finetune_glm_boolq.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/glm/finetune_glm_boolq.py -------------------------------------------------------------------------------- /examples/glm/finetune_glm_sst2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/glm/finetune_glm_sst2.py -------------------------------------------------------------------------------- /examples/glm/inference_glm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/glm/inference_glm.py -------------------------------------------------------------------------------- /examples/glm/scripts/ds_config_ft.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/glm/scripts/ds_config_ft.json -------------------------------------------------------------------------------- /examples/glm/scripts/finetune_boolq.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/glm/scripts/finetune_boolq.sh -------------------------------------------------------------------------------- /examples/glm/scripts/finetune_sst2.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/glm/scripts/finetune_sst2.sh -------------------------------------------------------------------------------- /examples/glm/scripts/generate_glm.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/glm/scripts/generate_glm.sh -------------------------------------------------------------------------------- /examples/glm4v/blip2_img_processor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/glm4v/blip2_img_processor.py -------------------------------------------------------------------------------- /examples/glm4v/chat_sat.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/glm4v/chat_sat.py -------------------------------------------------------------------------------- /examples/glm4v/chat_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/glm4v/chat_utils.py -------------------------------------------------------------------------------- /examples/glm4v/chatglm4_chat.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/glm4v/chatglm4_chat.py -------------------------------------------------------------------------------- /examples/glm4v/cli_demo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/glm4v/cli_demo.py -------------------------------------------------------------------------------- /examples/gpt2/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/gpt2/README.md -------------------------------------------------------------------------------- /examples/gpt2/inference_gpt2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/gpt2/inference_gpt2.py -------------------------------------------------------------------------------- /examples/gpt2/transform_param.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/gpt2/transform_param.py -------------------------------------------------------------------------------- /examples/gptneo/inference_gptneo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/gptneo/inference_gptneo.py -------------------------------------------------------------------------------- /examples/gptneo/scripts/inference_gptneo.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/gptneo/scripts/inference_gptneo.sh -------------------------------------------------------------------------------- /examples/llama/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/llama/README.md -------------------------------------------------------------------------------- /examples/llama/chat_sat.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/llama/chat_sat.py -------------------------------------------------------------------------------- /examples/llama/split_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/llama/split_model.py -------------------------------------------------------------------------------- /examples/llama/transform_param.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/llama/transform_param.py -------------------------------------------------------------------------------- /examples/mae/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/mae/README.md -------------------------------------------------------------------------------- /examples/mae/config/mae_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/mae/config/mae_config.py -------------------------------------------------------------------------------- /examples/mae/finetune_mae_cifar10.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/mae/finetune_mae_cifar10.py -------------------------------------------------------------------------------- /examples/mae/inference_mae.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/mae/inference_mae.ipynb -------------------------------------------------------------------------------- /examples/mae/mae_finetune_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/mae/mae_finetune_model.py -------------------------------------------------------------------------------- /examples/mae/models_mae.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/mae/models_mae.py -------------------------------------------------------------------------------- /examples/mae/scripts/finetune_cifar10.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/mae/scripts/finetune_cifar10.sh -------------------------------------------------------------------------------- /examples/mae/transform_param.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/mae/transform_param.py -------------------------------------------------------------------------------- /examples/mae/util/crop.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/mae/util/crop.py -------------------------------------------------------------------------------- /examples/mae/util/datasets.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/mae/util/datasets.py -------------------------------------------------------------------------------- /examples/mae/util/lars.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/mae/util/lars.py -------------------------------------------------------------------------------- /examples/mae/util/lr_decay.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/mae/util/lr_decay.py -------------------------------------------------------------------------------- /examples/mae/util/lr_sched.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/mae/util/lr_sched.py -------------------------------------------------------------------------------- /examples/mae/util/misc.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/mae/util/misc.py -------------------------------------------------------------------------------- /examples/mae/util/pos_embed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/mae/util/pos_embed.py -------------------------------------------------------------------------------- /examples/mixtral-8x7b/chat_sat.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/mixtral-8x7b/chat_sat.py -------------------------------------------------------------------------------- /examples/mixtral-8x7b/transform_param.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/mixtral-8x7b/transform_param.py -------------------------------------------------------------------------------- /examples/roberta/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/roberta/README.md -------------------------------------------------------------------------------- /examples/roberta/config/model_roberta_base.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/roberta/config/model_roberta_base.sh -------------------------------------------------------------------------------- /examples/roberta/config/model_roberta_large.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/roberta/config/model_roberta_large.sh -------------------------------------------------------------------------------- /examples/roberta/finetune_roberta_boolq.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/roberta/finetune_roberta_boolq.py -------------------------------------------------------------------------------- /examples/roberta/finetune_roberta_cb.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/roberta/finetune_roberta_cb.py -------------------------------------------------------------------------------- /examples/roberta/finetune_roberta_copa.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/roberta/finetune_roberta_copa.py -------------------------------------------------------------------------------- /examples/roberta/finetune_roberta_multirc.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/roberta/finetune_roberta_multirc.py -------------------------------------------------------------------------------- /examples/roberta/finetune_roberta_record.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/roberta/finetune_roberta_record.py -------------------------------------------------------------------------------- /examples/roberta/finetune_roberta_rte.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/roberta/finetune_roberta_rte.py -------------------------------------------------------------------------------- /examples/roberta/finetune_roberta_wic.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/roberta/finetune_roberta_wic.py -------------------------------------------------------------------------------- /examples/roberta/finetune_roberta_wsc.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/roberta/finetune_roberta_wsc.py -------------------------------------------------------------------------------- /examples/roberta/finetune_roberta_wsc_concat.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/roberta/finetune_roberta_wsc_concat.py -------------------------------------------------------------------------------- /examples/roberta/inference_roberta.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/roberta/inference_roberta.py -------------------------------------------------------------------------------- /examples/roberta/scripts/ds_config_ft.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/roberta/scripts/ds_config_ft.json -------------------------------------------------------------------------------- /examples/roberta/scripts/finetune_superglue.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/roberta/scripts/finetune_superglue.sh -------------------------------------------------------------------------------- /examples/roberta/scripts/finetune_superglue_multidevice.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/roberta/scripts/finetune_superglue_multidevice.sh -------------------------------------------------------------------------------- /examples/roberta/scripts/inference_roberta.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/roberta/scripts/inference_roberta.sh -------------------------------------------------------------------------------- /examples/roberta/transform_param.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/roberta/transform_param.py -------------------------------------------------------------------------------- /examples/t5/config/config_t5_large.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/t5/config/config_t5_large.json -------------------------------------------------------------------------------- /examples/t5/config/model_t5_large.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/t5/config/model_t5_large.sh -------------------------------------------------------------------------------- /examples/t5/finetune_t5.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/t5/finetune_t5.py -------------------------------------------------------------------------------- /examples/t5/inference_t5.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/t5/inference_t5.py -------------------------------------------------------------------------------- /examples/t5/scripts/config_t5_tmp.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/t5/scripts/config_t5_tmp.json -------------------------------------------------------------------------------- /examples/t5/scripts/finetune_t5.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/t5/scripts/finetune_t5.sh -------------------------------------------------------------------------------- /examples/t5/scripts/generate_t5.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/t5/scripts/generate_t5.sh -------------------------------------------------------------------------------- /examples/t5/test_t5.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/t5/test_t5.py -------------------------------------------------------------------------------- /examples/vit/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/vit/README.md -------------------------------------------------------------------------------- /examples/vit/config/vit_base_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/vit/config/vit_base_config.py -------------------------------------------------------------------------------- /examples/vit/finetune_vit_cifar10.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/vit/finetune_vit_cifar10.py -------------------------------------------------------------------------------- /examples/vit/scripts/finetune_cifar10.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/vit/scripts/finetune_cifar10.sh -------------------------------------------------------------------------------- /examples/vit/transform_param.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/vit/transform_param.py -------------------------------------------------------------------------------- /examples/vit/vit_ft_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/vit/vit_ft_model.py -------------------------------------------------------------------------------- /examples/yolos/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/README.md -------------------------------------------------------------------------------- /examples/yolos/config/yolos_tiny_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/config/yolos_tiny_config.py -------------------------------------------------------------------------------- /examples/yolos/datasets_/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/datasets_/__init__.py -------------------------------------------------------------------------------- /examples/yolos/datasets_/coco.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/datasets_/coco.py -------------------------------------------------------------------------------- /examples/yolos/datasets_/coco_eval.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/datasets_/coco_eval.py -------------------------------------------------------------------------------- /examples/yolos/datasets_/transforms.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/datasets_/transforms.py -------------------------------------------------------------------------------- /examples/yolos/datasets_/voc.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/datasets_/voc.py -------------------------------------------------------------------------------- /examples/yolos/infer_util.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/infer_util.py -------------------------------------------------------------------------------- /examples/yolos/inference_yolos.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/inference_yolos.py -------------------------------------------------------------------------------- /examples/yolos/models/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/models/__init__.py -------------------------------------------------------------------------------- /examples/yolos/models/backbone.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/models/backbone.py -------------------------------------------------------------------------------- /examples/yolos/models/detector.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/models/detector.py -------------------------------------------------------------------------------- /examples/yolos/models/layers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/models/layers/__init__.py -------------------------------------------------------------------------------- /examples/yolos/models/layers/drop.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/models/layers/drop.py -------------------------------------------------------------------------------- /examples/yolos/models/layers/helper.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/models/layers/helper.py -------------------------------------------------------------------------------- /examples/yolos/models/layers/weight_init.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/models/layers/weight_init.py -------------------------------------------------------------------------------- /examples/yolos/models/matcher.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/models/matcher.py -------------------------------------------------------------------------------- /examples/yolos/scripts/finetune_coco.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/scripts/finetune_coco.sh -------------------------------------------------------------------------------- /examples/yolos/train_yolos_coco.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/train_yolos_coco.py -------------------------------------------------------------------------------- /examples/yolos/transform_param.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/transform_param.py -------------------------------------------------------------------------------- /examples/yolos/util/__init__.py: -------------------------------------------------------------------------------- 1 | # Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved 2 | -------------------------------------------------------------------------------- /examples/yolos/util/box_ops.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/util/box_ops.py -------------------------------------------------------------------------------- /examples/yolos/util/debug.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/util/debug.py -------------------------------------------------------------------------------- /examples/yolos/util/misc.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/util/misc.py -------------------------------------------------------------------------------- /examples/yolos/util/plot_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/util/plot_utils.py -------------------------------------------------------------------------------- /examples/yolos/util/scheduler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/util/scheduler.py -------------------------------------------------------------------------------- /examples/yolos/visualization_new/exp/gt_img.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/visualization_new/exp/gt_img.png -------------------------------------------------------------------------------- /examples/yolos/visualization_new/exp/img.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/visualization_new/exp/img.png -------------------------------------------------------------------------------- /examples/yolos/visualization_new/exp/pred_img.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/examples/yolos/visualization_new/exp/pred_img.png -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/requirements.txt -------------------------------------------------------------------------------- /sat/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/__init__.py -------------------------------------------------------------------------------- /sat/arguments.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/arguments.py -------------------------------------------------------------------------------- /sat/data_utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/data_utils/__init__.py -------------------------------------------------------------------------------- /sat/data_utils/configure_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/data_utils/configure_data.py -------------------------------------------------------------------------------- /sat/data_utils/datasets.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/data_utils/datasets.py -------------------------------------------------------------------------------- /sat/data_utils/hf_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/data_utils/hf_dataset.py -------------------------------------------------------------------------------- /sat/data_utils/jsonlds.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/data_utils/jsonlds.py -------------------------------------------------------------------------------- /sat/data_utils/samplers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/data_utils/samplers.py -------------------------------------------------------------------------------- /sat/data_utils/webds.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/data_utils/webds.py -------------------------------------------------------------------------------- /sat/generation/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /sat/generation/autoregressive_sampling.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/generation/autoregressive_sampling.py -------------------------------------------------------------------------------- /sat/generation/cuda2d_sampling.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/generation/cuda2d_sampling.py -------------------------------------------------------------------------------- /sat/generation/magnify.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/generation/magnify.py -------------------------------------------------------------------------------- /sat/generation/sampling_strategies/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/generation/sampling_strategies/__init__.py -------------------------------------------------------------------------------- /sat/generation/sampling_strategies/base_strategy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/generation/sampling_strategies/base_strategy.py -------------------------------------------------------------------------------- /sat/generation/sampling_strategies/beam_search_strategy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/generation/sampling_strategies/beam_search_strategy.py -------------------------------------------------------------------------------- /sat/generation/sampling_strategies/iterative_entfilter_strategy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/generation/sampling_strategies/iterative_entfilter_strategy.py -------------------------------------------------------------------------------- /sat/generation/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/generation/utils.py -------------------------------------------------------------------------------- /sat/helpers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/helpers.py -------------------------------------------------------------------------------- /sat/model/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/__init__.py -------------------------------------------------------------------------------- /sat/model/attention/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/attention/__init__.py -------------------------------------------------------------------------------- /sat/model/attention/memory_efficient_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/attention/memory_efficient_attention.py -------------------------------------------------------------------------------- /sat/model/base_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/base_model.py -------------------------------------------------------------------------------- /sat/model/cached_autoregressive_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/cached_autoregressive_model.py -------------------------------------------------------------------------------- /sat/model/encoder_decoder_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/encoder_decoder_model.py -------------------------------------------------------------------------------- /sat/model/finetune/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/finetune/__init__.py -------------------------------------------------------------------------------- /sat/model/finetune/adapter.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/finetune/adapter.py -------------------------------------------------------------------------------- /sat/model/finetune/ffadd.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/finetune/ffadd.py -------------------------------------------------------------------------------- /sat/model/finetune/lora.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/finetune/lora.py -------------------------------------------------------------------------------- /sat/model/finetune/lora2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/finetune/lora2.py -------------------------------------------------------------------------------- /sat/model/finetune/mlp_head.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/finetune/mlp_head.py -------------------------------------------------------------------------------- /sat/model/finetune/prompt_tuning.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/finetune/prompt_tuning.py -------------------------------------------------------------------------------- /sat/model/mixins.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/mixins.py -------------------------------------------------------------------------------- /sat/model/official/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/__init__.py -------------------------------------------------------------------------------- /sat/model/official/bert_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/bert_model.py -------------------------------------------------------------------------------- /sat/model/official/cait_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/cait_model.py -------------------------------------------------------------------------------- /sat/model/official/chatglm2_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/chatglm2_model.py -------------------------------------------------------------------------------- /sat/model/official/chatglm3_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/chatglm3_model.py -------------------------------------------------------------------------------- /sat/model/official/chatglm4_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/chatglm4_model.py -------------------------------------------------------------------------------- /sat/model/official/chatglm_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/chatglm_model.py -------------------------------------------------------------------------------- /sat/model/official/clip_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/clip_model.py -------------------------------------------------------------------------------- /sat/model/official/cuda2d_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/cuda2d_model.py -------------------------------------------------------------------------------- /sat/model/official/distill_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/distill_model.py -------------------------------------------------------------------------------- /sat/model/official/dpr_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/dpr_model.py -------------------------------------------------------------------------------- /sat/model/official/eva2_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/eva2_model.py -------------------------------------------------------------------------------- /sat/model/official/eva_clip_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/eva_clip_model.py -------------------------------------------------------------------------------- /sat/model/official/glm130B_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/glm130B_model.py -------------------------------------------------------------------------------- /sat/model/official/glm4v_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/glm4v_model.py -------------------------------------------------------------------------------- /sat/model/official/glm_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/glm_model.py -------------------------------------------------------------------------------- /sat/model/official/gpt2_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/gpt2_model.py -------------------------------------------------------------------------------- /sat/model/official/gptneo_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/gptneo_model.py -------------------------------------------------------------------------------- /sat/model/official/llama_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/llama_model.py -------------------------------------------------------------------------------- /sat/model/official/mae_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/mae_model.py -------------------------------------------------------------------------------- /sat/model/official/mixtral_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/mixtral_model.py -------------------------------------------------------------------------------- /sat/model/official/roberta_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/roberta_model.py -------------------------------------------------------------------------------- /sat/model/official/t5_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/t5_model.py -------------------------------------------------------------------------------- /sat/model/official/vit_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/vit_model.py -------------------------------------------------------------------------------- /sat/model/official/yolos_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/official/yolos_model.py -------------------------------------------------------------------------------- /sat/model/position_embedding/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/position_embedding/__init__.py -------------------------------------------------------------------------------- /sat/model/position_embedding/rotary_embeddings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/position_embedding/rotary_embeddings.py -------------------------------------------------------------------------------- /sat/model/position_embedding/rotary_embeddings_original.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/position_embedding/rotary_embeddings_original.py -------------------------------------------------------------------------------- /sat/model/position_embedding/sincos2d.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/position_embedding/sincos2d.py -------------------------------------------------------------------------------- /sat/model/position_embedding/triton_rotary.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/position_embedding/triton_rotary.py -------------------------------------------------------------------------------- /sat/model/position_embedding/triton_rotary_embeddings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/position_embedding/triton_rotary_embeddings.py -------------------------------------------------------------------------------- /sat/model/position_embedding/vision_rotary_embeddings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/position_embedding/vision_rotary_embeddings.py -------------------------------------------------------------------------------- /sat/model/registry.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/registry.py -------------------------------------------------------------------------------- /sat/model/transformer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/model/transformer.py -------------------------------------------------------------------------------- /sat/mpu/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/mpu/__init__.py -------------------------------------------------------------------------------- /sat/mpu/cross_entropy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/mpu/cross_entropy.py -------------------------------------------------------------------------------- /sat/mpu/data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/mpu/data.py -------------------------------------------------------------------------------- /sat/mpu/initialize.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/mpu/initialize.py -------------------------------------------------------------------------------- /sat/mpu/layers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/mpu/layers.py -------------------------------------------------------------------------------- /sat/mpu/mappings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/mpu/mappings.py -------------------------------------------------------------------------------- /sat/mpu/operation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/mpu/operation.py -------------------------------------------------------------------------------- /sat/mpu/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/mpu/utils.py -------------------------------------------------------------------------------- /sat/ops/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/ops/__init__.py -------------------------------------------------------------------------------- /sat/ops/csrc/adam/fused_ema_adam_frontend.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/ops/csrc/adam/fused_ema_adam_frontend.cpp -------------------------------------------------------------------------------- /sat/ops/csrc/adam/multi_tensor_apply.cuh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/ops/csrc/adam/multi_tensor_apply.cuh -------------------------------------------------------------------------------- /sat/ops/csrc/adam/multi_tensor_ema_adam.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/ops/csrc/adam/multi_tensor_ema_adam.cu -------------------------------------------------------------------------------- /sat/ops/csrc/includes/compat.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/ops/csrc/includes/compat.h -------------------------------------------------------------------------------- /sat/ops/csrc/includes/type_shim.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/ops/csrc/includes/type_shim.h -------------------------------------------------------------------------------- /sat/ops/fused_ema_adam.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/ops/fused_ema_adam.py -------------------------------------------------------------------------------- /sat/ops/layernorm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/ops/layernorm.py -------------------------------------------------------------------------------- /sat/ops/local_attention_function.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/ops/local_attention_function.py -------------------------------------------------------------------------------- /sat/ops/memory_efficient_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/ops/memory_efficient_attention.py -------------------------------------------------------------------------------- /sat/ops/ops_builder/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/ops/ops_builder/__init__.py -------------------------------------------------------------------------------- /sat/ops/ops_builder/builder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/ops/ops_builder/builder.py -------------------------------------------------------------------------------- /sat/ops/ops_builder/fused_ema_adam.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/ops/ops_builder/fused_ema_adam.py -------------------------------------------------------------------------------- /sat/ops/scaled_mask_softmax.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/ops/scaled_mask_softmax.py -------------------------------------------------------------------------------- /sat/quantization/__init__.py: -------------------------------------------------------------------------------- 1 | from .kernels import quantize -------------------------------------------------------------------------------- /sat/quantization/kernels.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/quantization/kernels.py -------------------------------------------------------------------------------- /sat/resources/__init__.py: -------------------------------------------------------------------------------- 1 | from .download import auto_create -------------------------------------------------------------------------------- /sat/resources/download.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/resources/download.py -------------------------------------------------------------------------------- /sat/resources/urls.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/resources/urls.py -------------------------------------------------------------------------------- /sat/tokenization/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/__init__.py -------------------------------------------------------------------------------- /sat/tokenization/cogview/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/cogview/__init__.py -------------------------------------------------------------------------------- /sat/tokenization/cogview/sp_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/cogview/sp_tokenizer.py -------------------------------------------------------------------------------- /sat/tokenization/cogview/templates.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/cogview/templates.py -------------------------------------------------------------------------------- /sat/tokenization/cogview/unified_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/cogview/unified_tokenizer.py -------------------------------------------------------------------------------- /sat/tokenization/cogview/vqvae/LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/cogview/vqvae/LICENSE -------------------------------------------------------------------------------- /sat/tokenization/cogview/vqvae/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/cogview/vqvae/README.md -------------------------------------------------------------------------------- /sat/tokenization/cogview/vqvae/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/cogview/vqvae/__init__.py -------------------------------------------------------------------------------- /sat/tokenization/cogview/vqvae/api.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/cogview/vqvae/api.py -------------------------------------------------------------------------------- /sat/tokenization/cogview/vqvae/vqvae_diffusion.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/cogview/vqvae/vqvae_diffusion.py -------------------------------------------------------------------------------- /sat/tokenization/cogview/vqvae/vqvae_zc.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/cogview/vqvae/vqvae_zc.py -------------------------------------------------------------------------------- /sat/tokenization/cogview/vqvae_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/cogview/vqvae_tokenizer.py -------------------------------------------------------------------------------- /sat/tokenization/embed_assets/chinese_sentencepiece/cog-pretrain.model: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/embed_assets/chinese_sentencepiece/cog-pretrain.model -------------------------------------------------------------------------------- /sat/tokenization/embed_assets/chinese_sentencepiece/cog-pretrain.vocab: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/embed_assets/chinese_sentencepiece/cog-pretrain.vocab -------------------------------------------------------------------------------- /sat/tokenization/embed_assets/english_tokenizer/bert-base-uncased-vocab.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/embed_assets/english_tokenizer/bert-base-uncased-vocab.txt -------------------------------------------------------------------------------- /sat/tokenization/embed_assets/english_tokenizer/bert-large-uncased-vocab.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/embed_assets/english_tokenizer/bert-large-uncased-vocab.txt -------------------------------------------------------------------------------- /sat/tokenization/embed_assets/english_tokenizer/gpt2-merges.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/embed_assets/english_tokenizer/gpt2-merges.txt -------------------------------------------------------------------------------- /sat/tokenization/embed_assets/english_tokenizer/gpt2-vocab.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/embed_assets/english_tokenizer/gpt2-vocab.json -------------------------------------------------------------------------------- /sat/tokenization/embed_assets/english_tokenizer/roberta-merges.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/embed_assets/english_tokenizer/roberta-merges.txt -------------------------------------------------------------------------------- /sat/tokenization/embed_assets/english_tokenizer/roberta-vocab.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/embed_assets/english_tokenizer/roberta-vocab.json -------------------------------------------------------------------------------- /sat/tokenization/glm/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/glm/__init__.py -------------------------------------------------------------------------------- /sat/tokenization/glm/sp_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/glm/sp_tokenizer.py -------------------------------------------------------------------------------- /sat/tokenization/glm/tokenization.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/glm/tokenization.py -------------------------------------------------------------------------------- /sat/tokenization/glm/tokenization_gpt2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/glm/tokenization_gpt2.py -------------------------------------------------------------------------------- /sat/tokenization/glm/tokenization_wordpiece.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/glm/tokenization_wordpiece.py -------------------------------------------------------------------------------- /sat/tokenization/hf_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/hf_tokenizer.py -------------------------------------------------------------------------------- /sat/tokenization/icetk_glm_130B/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/icetk_glm_130B/__init__.py -------------------------------------------------------------------------------- /sat/tokenization/icetk_glm_130B/ice_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/icetk_glm_130B/ice_tokenizer.py -------------------------------------------------------------------------------- /sat/tokenization/icetk_glm_130B/tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/tokenization/icetk_glm_130B/tokenizer.py -------------------------------------------------------------------------------- /sat/training/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /sat/training/deepspeed_training.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/training/deepspeed_training.py -------------------------------------------------------------------------------- /sat/training/deepspeed_zero0.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/training/deepspeed_zero0.json -------------------------------------------------------------------------------- /sat/training/deepspeed_zero1.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/training/deepspeed_zero1.json -------------------------------------------------------------------------------- /sat/training/deepspeed_zero2.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/training/deepspeed_zero2.json -------------------------------------------------------------------------------- /sat/training/learning_rates.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/training/learning_rates.py -------------------------------------------------------------------------------- /sat/training/model_io.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/training/model_io.py -------------------------------------------------------------------------------- /sat/training/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/training/utils.py -------------------------------------------------------------------------------- /sat/transformer_defaults.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/sat/transformer_defaults.py -------------------------------------------------------------------------------- /setup.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/setup.py -------------------------------------------------------------------------------- /tests/deepspeed_test.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tests/deepspeed_test.json -------------------------------------------------------------------------------- /tests/sbatch_launch.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tests/sbatch_launch.sh -------------------------------------------------------------------------------- /tests/single_launch.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tests/single_launch.sh -------------------------------------------------------------------------------- /tests/test_base_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tests/test_base_model.py -------------------------------------------------------------------------------- /tests/test_inference.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tests/test_inference.py -------------------------------------------------------------------------------- /tests/test_jsonlds.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tests/test_jsonlds.py -------------------------------------------------------------------------------- /tests/test_list_info.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tests/test_list_info.py -------------------------------------------------------------------------------- /tests/test_mea.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tests/test_mea.py -------------------------------------------------------------------------------- /tests/test_model_parallel.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tests/test_model_parallel.py -------------------------------------------------------------------------------- /tests/test_multi_webds.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tests/test_multi_webds.py -------------------------------------------------------------------------------- /tests/test_nested_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tests/test_nested_model.py -------------------------------------------------------------------------------- /tests/test_read_img.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tests/test_read_img.py -------------------------------------------------------------------------------- /tests/test_remote_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tests/test_remote_data.py -------------------------------------------------------------------------------- /tests/test_speed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tests/test_speed.py -------------------------------------------------------------------------------- /tests/test_train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tests/test_train.py -------------------------------------------------------------------------------- /tests/test_train_dp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tests/test_train_dp.py -------------------------------------------------------------------------------- /tests/test_train_nested.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tests/test_train_nested.py -------------------------------------------------------------------------------- /tests/test_triton_rotary_embedding.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tests/test_triton_rotary_embedding.py -------------------------------------------------------------------------------- /tutorials/inference.md: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tutorials/model_usage.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tutorials/model_usage.md -------------------------------------------------------------------------------- /tutorials/speed_self_check.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tutorials/speed_self_check.md -------------------------------------------------------------------------------- /tutorials/training.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/THUDM/SwissArmyTransformer/HEAD/tutorials/training.md --------------------------------------------------------------------------------