Datasets:

---

DanielPFlorian

/

Transformers-Github-Issues

like 0

[ { "id": 1843244711, "node_id": "MDU6TGFiZWwxODQzMjQ0NzEx", "url": "https://api.github.com/repos/huggingface/transformers/labels/New%20model", "name": "New model", "color": "fbca04", "default": false, "description": "" } ]

[]

[ "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored.", "+1", "Does dall-e mini currently added to transformers?\r\nCurrently, it doesn't `eBart` is not recognized in transformers library.", "cc @patil-suraj who's currently working on making it easier to use from transformers", "There are dozens of DALL-E models currently listed on the Hugging Face site. Unless this is a specific variant/implementation that has yet to be added, it seems this issue can be closed." ]

[]

[]

[ "Tested on one GPU, two GPUs and TPUs, this runs fine everywhere. So just waiting for the small adjustments and it should be good to be merged :-) ", "Thanks a lot!" ]

[]

[]

[ "@mayanb I don't think this model has a TF version till now, check:- https://huggingface.co/facebook/bart-large-mnli/tree/main", "@frankhart2018 ah you're right! thanks!" ]

[]

[]

[]

[]

[]

[ "Hello! Unfortunately, recent transformers versions only work with torch 1.4.0+. The README is incorrect respective to that, and I'll update it in the coming days.", "> Hello! Unfortunately, recent transformers versions only work with torch 1.4.0+. The README is incorrect respective to that, and I'll update it in the coming days.\r\n\r\nThank you for the response.\r\n\r\nOkay. Does it mean that there is no another way for expressing `masked_index = torch.nonzero(input_ids == self.tokenizer.mask_token_id, as_tuple=False)` in pytorch 1.1.0? Is it still possible to use `torch.where(...)`? if yes, how to express it with `torch.where()`?\r\n\r\nThis is because the available machine we used is in an old driver version of CUDA.\r\n\r\nSincerely,\r\nMY", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored." ]

[]

[]

[ "@saurabhhssaurabh `trainer.predict()` expects an instance of `torch.utils.data.Dataset` to be passed and not a single sentence. I think it will be easier to use the trained model at self.model to predict rather than trying to use trainer object, as it does not have any single prediction method yet.", "@frankhart2018 \r\nThank you for replying. I will implement code using prediction from self.model." ]

[ { "id": 2659267025, "node_id": "MDU6TGFiZWwyNjU5MjY3MDI1", "url": "https://api.github.com/repos/huggingface/transformers/labels/DeepSpeed", "name": "DeepSpeed", "color": "4D34F7", "default": false, "description": "" } ]

[ { "login": "stas00", "id": 10676103, "node_id": "MDQ6VXNlcjEwNjc2MTAz", "avatar_url": "https://avatars.githubusercontent.com/u/10676103?v=4", "gravatar_id": "", "url": "https://api.github.com/users/stas00", "html_url": "https://github.com/stas00", "followers_url": "https://api.github.com/users/stas00/followers", "following_url": "https://api.github.com/users/stas00/following{/other_user}", "gists_url": "https://api.github.com/users/stas00/gists{/gist_id}", "starred_url": "https://api.github.com/users/stas00/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/stas00/subscriptions", "organizations_url": "https://api.github.com/users/stas00/orgs", "repos_url": "https://api.github.com/users/stas00/repos", "events_url": "https://api.github.com/users/stas00/events{/privacy}", "received_events_url": "https://api.github.com/users/stas00/received_events", "type": "User", "site_admin": false } ]

[ "Also adding the logs from the beginning of training with deepspeed:\r\n\r\ndeepspeed examples/seq2seq/run_summarization.py --model_name_or_path t5-small --do_train --do_eval --dataset_name cnn_dailymail --dataset_config \"3.0.0\" --source_prefix \"summarize: \" --output_dir /tmp/tst-summarization --per_device_train_batch_size=4 --per_device_eval_batch_size=4 --overwrite_output_dir --predict_with_generate --deepspeed examples/tests/deepspeed/ds_config.json\r\n[2021-03-27 17:02:34,357] [WARNING] [runner.py:117:fetch_hostfile] Unable to find hostfile, will proceed with training with local resources only.\r\n[2021-03-27 17:02:34,381] [INFO] [runner.py:358:main] cmd = /media/disk1/oriyor/hf_venv_3.6/bin/python -u -m deepspeed.launcher.launch --world_info=eyJsb2NhbGhvc3QiOiBbMF19 --master_addr=127.0.0.1 --master_port=29500 examples/seq2seq/run_summarization.py --model_name_or_path t5-small --do_train --do_eval --dataset_name cnn_dailymail --dataset_config 3.0.0 --source_prefix summarize: --output_dir /tmp/tst-summarization --per_device_train_batch_size=4 --per_device_eval_batch_size=4 --overwrite_output_dir --predict_with_generate --deepspeed examples/tests/deepspeed/ds_config.json\r\n[2021-03-27 17:02:34,981] [INFO] [launch.py:80:main] WORLD INFO DICT: {'localhost': [0]}\r\n[2021-03-27 17:02:34,981] [INFO] [launch.py:89:main] nnodes=1, num_local_procs=1, node_rank=0\r\n[2021-03-27 17:02:34,981] [INFO] [launch.py:101:main] global_rank_mapping=defaultdict(<class 'list'>, {'localhost': [0]})\r\n[2021-03-27 17:02:34,981] [INFO] [launch.py:102:main] dist_world_size=1\r\n[2021-03-27 17:02:34,981] [INFO] [launch.py:105:main] Setting CUDA_VISIBLE_DEVICES=0\r\n[2021-03-27 17:02:36,820] [INFO] [distributed.py:47:init_distributed] Initializing torch distributed with backend: nccl\r\nWARNING:__main__:Process rank: 0, device: cuda:0, n_gpu: 1distributed training: True, 16-bits training: False\r\nINFO:__main__:Training/evaluation parameters Seq2SeqTrainingArguments(output_dir='/tmp/tst-summarization', overwrite_output_dir=True, do_train=True, do_eval=True, do_predict=False, evaluation_strategy=<IntervalStrategy.NO: 'no'>, prediction_loss_only=False, per_device_train_batch_size=4, per_device_eval_batch_size=4, per_gpu_train_batch_size=None, per_gpu_eval_batch_size=None, gradient_accumulation_steps=1, eval_accumulation_steps=None, learning_rate=5e-05, weight_decay=0.0, adam_beta1=0.9, adam_beta2=0.999, adam_epsilon=1e-08, max_grad_norm=1.0, num_train_epochs=3.0, max_steps=-1, lr_scheduler_type=<SchedulerType.LINEAR: 'linear'>, warmup_ratio=0.0, warmup_steps=0, logging_dir='runs/Mar27_17-02-36_rack-jonathan-g04', logging_strategy=<IntervalStrategy.STEPS: 'steps'>, logging_first_step=False, logging_steps=500, save_strategy=<IntervalStrategy.STEPS: 'steps'>, save_steps=500, save_total_limit=None, no_cuda=False, seed=42, fp16=False, fp16_opt_level='O1', fp16_backend='auto', fp16_full_eval=False, local_rank=0, tpu_num_cores=None, tpu_metrics_debug=False, debug=False, dataloader_drop_last=False, eval_steps=500, dataloader_num_workers=0, past_index=-1, run_name='/tmp/tst-summarization', disable_tqdm=False, remove_unused_columns=True, label_names=None, load_best_model_at_end=False, metric_for_best_model=None, greater_is_better=None, ignore_data_skip=False, sharded_ddp=[], deepspeed='examples/tests/deepspeed/ds_config.json', label_smoothing_factor=0.0, adafactor=False, group_by_length=False, report_to=['tensorboard'], ddp_find_unused_parameters=None, dataloader_pin_memory=True, skip_memory_metrics=False, sortish_sampler=False, predict_with_generate=True)\r\nWARNING:datasets.builder:Reusing dataset cnn_dailymail (/home/oriy/.cache/huggingface/datasets/cnn_dailymail/3.0.0/3.0.0/0a01b1abede4f646130574f203de57a293ded8a7a11e3406a539453afdfeb2c0)\r\nloading configuration file https://huggingface.co/t5-small/resolve/main/config.json from cache at /home/oriy/.cache/huggingface/transformers/fe501e8fd6425b8ec93df37767fcce78ce626e34cc5edc859c662350cf712e41.406701565c0afd9899544c1cb8b93185a76f00b31e5ce7f6e18bbaef02241985\r\nModel config T5Config {\r\n \"architectures\": [\r\n \"T5WithLMHeadModel\"\r\n ],\r\n \"d_ff\": 2048,\r\n \"d_kv\": 64,\r\n \"d_model\": 512,\r\n \"decoder_start_token_id\": 0,\r\n \"dropout_rate\": 0.1,\r\n \"eos_token_id\": 1,\r\n \"feed_forward_proj\": \"relu\",\r\n \"initializer_factor\": 1.0,\r\n \"is_encoder_decoder\": true,\r\n \"layer_norm_epsilon\": 1e-06,\r\n \"model_type\": \"t5\",\r\n \"n_positions\": 512,\r\n \"num_decoder_layers\": 6,\r\n \"num_heads\": 8,\r\n \"num_layers\": 6,\r\n \"output_past\": true,\r\n \"pad_token_id\": 0,\r\n \"relative_attention_num_buckets\": 32,\r\n \"task_specific_params\": {\r\n \"summarization\": {\r\n \"early_stopping\": true,\r\n \"length_penalty\": 2.0,\r\n \"max_length\": 200,\r\n \"min_length\": 30,\r\n \"no_repeat_ngram_size\": 3,\r\n \"num_beams\": 4,\r\n \"prefix\": \"summarize: \"\r\n },\r\n \"translation_en_to_de\": {\r\n \"early_stopping\": true,\r\n \"max_length\": 300,\r\n \"num_beams\": 4,\r\n \"prefix\": \"translate English to German: \"\r\n },\r\n \"translation_en_to_fr\": {\r\n \"early_stopping\": true,\r\n \"max_length\": 300,\r\n \"num_beams\": 4,\r\n \"prefix\": \"translate English to French: \"\r\n },\r\n \"translation_en_to_ro\": {\r\n \"early_stopping\": true,\r\n \"max_length\": 300,\r\n \"num_beams\": 4,\r\n \"prefix\": \"translate English to Romanian: \"\r\n }\r\n },\r\n \"transformers_version\": \"4.5.0.dev0\",\r\n \"use_cache\": true,\r\n \"vocab_size\": 32128\r\n}\r\n\r\nloading configuration file https://huggingface.co/t5-small/resolve/main/config.json from cache at /home/oriy/.cache/huggingface/transformers/fe501e8fd6425b8ec93df37767fcce78ce626e34cc5edc859c662350cf712e41.406701565c0afd9899544c1cb8b93185a76f00b31e5ce7f6e18bbaef02241985\r\nModel config T5Config {\r\n \"architectures\": [\r\n \"T5WithLMHeadModel\"\r\n ],\r\n \"d_ff\": 2048,\r\n \"d_kv\": 64,\r\n \"d_model\": 512,\r\n \"decoder_start_token_id\": 0,\r\n \"dropout_rate\": 0.1,\r\n \"eos_token_id\": 1,\r\n \"feed_forward_proj\": \"relu\",\r\n \"initializer_factor\": 1.0,\r\n \"is_encoder_decoder\": true,\r\n \"layer_norm_epsilon\": 1e-06,\r\n \"model_type\": \"t5\",\r\n \"n_positions\": 512,\r\n \"num_decoder_layers\": 6,\r\n \"num_heads\": 8,\r\n \"num_layers\": 6,\r\n \"output_past\": true,\r\n \"pad_token_id\": 0,\r\n \"relative_attention_num_buckets\": 32,\r\n \"task_specific_params\": {\r\n \"summarization\": {\r\n \"early_stopping\": true,\r\n \"length_penalty\": 2.0,\r\n \"max_length\": 200,\r\n \"min_length\": 30,\r\n \"no_repeat_ngram_size\": 3,\r\n \"num_beams\": 4,\r\n \"prefix\": \"summarize: \"\r\n },\r\n \"translation_en_to_de\": {\r\n \"early_stopping\": true,\r\n \"max_length\": 300,\r\n \"num_beams\": 4,\r\n \"prefix\": \"translate English to German: \"\r\n },\r\n \"translation_en_to_fr\": {\r\n \"early_stopping\": true,\r\n \"max_length\": 300,\r\n \"num_beams\": 4,\r\n \"prefix\": \"translate English to French: \"\r\n },\r\n \"translation_en_to_ro\": {\r\n \"early_stopping\": true,\r\n \"max_length\": 300,\r\n \"num_beams\": 4,\r\n \"prefix\": \"translate English to Romanian: \"\r\n }\r\n },\r\n \"transformers_version\": \"4.5.0.dev0\",\r\n \"use_cache\": true,\r\n \"vocab_size\": 32128\r\n}\r\n\r\nloading file https://huggingface.co/t5-small/resolve/main/spiece.model from cache at /home/oriy/.cache/huggingface/transformers/65fc04e21f45f61430aea0c4fedffac16a4d20d78b8e6601d8d996ebefefecd2.3b69006860e7b5d0a63ffdddc01ddcd6b7c318a6f4fd793596552c741734c62d\r\nloading file https://huggingface.co/t5-small/resolve/main/tokenizer.json from cache at /home/oriy/.cache/huggingface/transformers/06779097c78e12f47ef67ecb728810c2ae757ee0a9efe9390c6419783d99382d.8627f1bd5d270a9fd2e5a51c8bec3223896587cc3cfe13edeabb0992ab43c529\r\nloading file https://huggingface.co/t5-small/resolve/main/added_tokens.json from cache at None\r\nloading file https://huggingface.co/t5-small/resolve/main/special_tokens_map.json from cache at None\r\nloading file https://huggingface.co/t5-small/resolve/main/tokenizer_config.json from cache at None\r\nloading weights file https://huggingface.co/t5-small/resolve/main/pytorch_model.bin from cache at /home/oriy/.cache/huggingface/transformers/fee5a3a0ae379232608b6eed45d2d7a0d2966b9683728838412caccc41b4b0ed.ddacdc89ec88482db20c676f0861a336f3d0409f94748c209847b49529d73885\r\nAll model checkpoint weights were used when initializing T5ForConditionalGeneration.\r\n\r\nAll the weights of T5ForConditionalGeneration were initialized from the model checkpoint at t5-small.\r\nIf your task is similar to the task the model of the checkpoint was trained on, you can already use T5ForConditionalGeneration for predictions without further training.\r\nWARNING:datasets.arrow_dataset:Loading cached processed dataset at /home/oriy/.cache/huggingface/datasets/cnn_dailymail/3.0.0/3.0.0/0a01b1abede4f646130574f203de57a293ded8a7a11e3406a539453afdfeb2c0/cache-3c2d8ad9af1d1a3e.arrow\r\nWARNING:datasets.arrow_dataset:Loading cached processed dataset at /home/oriy/.cache/huggingface/datasets/cnn_dailymail/3.0.0/3.0.0/0a01b1abede4f646130574f203de57a293ded8a7a11e3406a539453afdfeb2c0/cache-2e7e82c8de410d07.arrow\r\nUpdating the `scheduler` config from examples/tests/deepspeed/ds_config.json with other command line arguments\r\nsetting optimizer.params.lr to 5e-05\r\nsetting optimizer.params.betas to [0.9, 0.999]\r\nsetting optimizer.params.eps to 1e-08\r\nsetting optimizer.params.weight_decay to 0.0\r\nUpdating the `scheduler` config from examples/tests/deepspeed/ds_config.json with other command line arguments\r\nsetting scheduler.params.warmup_max_lr to 5e-05\r\nsetting scheduler.params.warmup_num_steps to 0\r\n[2021-03-27 17:02:46,871] [INFO] [logging.py:60:log_dist] [Rank 0] DeepSpeed info: version=0.3.13, git-hash=unknown, git-branch=unknown\r\n[2021-03-27 17:02:48,970] [INFO] [engine.py:77:_initialize_parameter_parallel_groups] data_parallel_size: 1, parameter_parallel_size: 1\r\nhuggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\r\nTo disable this warning, you can either:\r\n\t- Avoid using `tokenizers` before the fork if possible\r\n\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\r\nhuggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\r\nTo disable this warning, you can either:\r\n\t- Avoid using `tokenizers` before the fork if possible\r\n\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\r\nhuggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\r\nTo disable this warning, you can either:\r\n\t- Avoid using `tokenizers` before the fork if possible\r\n\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\r\nhuggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\r\nTo disable this warning, you can either:\r\n\t- Avoid using `tokenizers` before the fork if possible\r\n\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\r\nhuggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\r\nTo disable this warning, you can either:\r\n\t- Avoid using `tokenizers` before the fork if possible\r\n\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\r\nUsing /home/oriy/.cache/torch_extensions as PyTorch extensions root...\r\nhuggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\r\nTo disable this warning, you can either:\r\n\t- Avoid using `tokenizers` before the fork if possible\r\n\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\r\nhuggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\r\nTo disable this warning, you can either:\r\n\t- Avoid using `tokenizers` before the fork if possible\r\n\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\r\nhuggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\r\nTo disable this warning, you can either:\r\n\t- Avoid using `tokenizers` before the fork if possible\r\n\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\r\nDetected CUDA files, patching ldflags\r\nEmitting ninja build file /home/oriy/.cache/torch_extensions/cpu_adam/build.ninja...\r\nBuilding extension module cpu_adam...\r\nAllowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)\r\nhuggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\r\nTo disable this warning, you can either:\r\n\t- Avoid using `tokenizers` before the fork if possible\r\n\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\r\nninja: no work to do.\r\nLoading extension module cpu_adam...\r\nTime to load cpu_adam op: 0.43370747566223145 seconds\r\nAdam Optimizer #0 is created with AVX2 arithmetic capability.\r\nConfig: alpha=0.000050, betas=(0.900000, 0.999000), weight_decay=0.000000, adam_w=1\r\n[2021-03-27 17:02:52,144] [INFO] [engine.py:602:_configure_optimizer] Using DeepSpeed Optimizer param name adam as basic optimizer\r\n[2021-03-27 17:02:52,145] [INFO] [engine.py:606:_configure_optimizer] DeepSpeed Basic Optimizer = DeepSpeedCPUAdam\r\nChecking ZeRO support for optimizer=DeepSpeedCPUAdam type=<class 'deepspeed.ops.adam.cpu_adam.DeepSpeedCPUAdam'>\r\n[2021-03-27 17:02:52,145] [INFO] [logging.py:60:log_dist] [Rank 0] Creating fp16 ZeRO stage 2 optimizer\r\nUsing /home/oriy/.cache/torch_extensions as PyTorch extensions root...\r\nhuggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\r\nTo disable this warning, you can either:\r\n\t- Avoid using `tokenizers` before the fork if possible\r\n\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\r\nhuggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\r\nTo disable this warning, you can either:\r\n\t- Avoid using `tokenizers` before the fork if possible\r\n\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\r\nhuggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\r\nTo disable this warning, you can either:\r\n\t- Avoid using `tokenizers` before the fork if possible\r\n\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\r\nEmitting ninja build file /home/oriy/.cache/torch_extensions/utils/build.ninja...\r\nBuilding extension module utils...\r\nAllowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)\r\nhuggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\r\nTo disable this warning, you can either:\r\n\t- Avoid using `tokenizers` before the fork if possible\r\n\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\r\nninja: no work to do.\r\nLoading extension module utils...\r\nTime to load utils op: 0.29197263717651367 seconds\r\n[2021-03-27 17:02:52,437] [INFO] [stage2.py:130:__init__] Reduce bucket size 200000000.0\r\n[2021-03-27 17:02:52,438] [INFO] [stage2.py:131:__init__] Allgather bucket size 200000000.0\r\n[2021-03-27 17:02:52,438] [INFO] [stage2.py:132:__init__] CPU Offload: True\r\n[2021-03-27 17:02:52,846] [INFO] [stage2.py:399:__init__] optimizer state initialized\r\n[2021-03-27 17:02:52,846] [INFO] [logging.py:60:log_dist] [Rank 0] DeepSpeed Final Optimizer = adam\r\n[2021-03-27 17:02:52,847] [INFO] [engine.py:439:_configure_lr_scheduler] DeepSpeed using configured LR scheduler = WarmupLR\r\n[2021-03-27 17:02:52,847] [INFO] [logging.py:60:log_dist] [Rank 0] DeepSpeed LR Scheduler = <deepspeed.runtime.lr_schedules.WarmupLR object at 0x7fea742ef2b0>\r\n[2021-03-27 17:02:52,847] [INFO] [logging.py:60:log_dist] [Rank 0] step=0, skipped=0, lr=[5e-05], mom=[[0.9, 0.999]]\r\n[2021-03-27 17:02:52,847] [INFO] [config.py:737:print] DeepSpeedEngine configuration:\r\n[2021-03-27 17:02:52,847] [INFO] [config.py:741:print] activation_checkpointing_config {\r\n \"contiguous_memory_optimization\": false,\r\n \"cpu_checkpointing\": false,\r\n \"number_checkpoints\": null,\r\n \"partition_activations\": false,\r\n \"profile\": false,\r\n \"synchronize_checkpoint_boundary\": false\r\n}\r\n[2021-03-27 17:02:52,847] [INFO] [config.py:741:print] allreduce_always_fp32 ........ False\r\n[2021-03-27 17:02:52,847] [INFO] [config.py:741:print] amp_enabled .................. False\r\n[2021-03-27 17:02:52,847] [INFO] [config.py:741:print] amp_params ................... False\r\n[2021-03-27 17:02:52,847] [INFO] [config.py:741:print] checkpoint_tag_validation_enabled True\r\n[2021-03-27 17:02:52,847] [INFO] [config.py:741:print] checkpoint_tag_validation_fail False\r\n[2021-03-27 17:02:52,847] [INFO] [config.py:741:print] disable_allgather ............ False\r\n[2021-03-27 17:02:52,848] [INFO] [config.py:741:print] dump_state ................... False\r\n[2021-03-27 17:02:52,848] [INFO] [config.py:741:print] dynamic_loss_scale_args ...... {'init_scale': 4294967296, 'scale_window': 1000, 'delayed_shift': 2, 'min_scale': 1}\r\n[2021-03-27 17:02:52,848] [INFO] [config.py:741:print] elasticity_enabled ........... False\r\n[2021-03-27 17:02:52,848] [INFO] [config.py:741:print] flops_profiler_config ........ {\r\n \"detailed\": true,\r\n \"enabled\": false,\r\n \"module_depth\": -1,\r\n \"profile_step\": 1,\r\n \"top_modules\": 3\r\n}\r\n[2021-03-27 17:02:52,848] [INFO] [config.py:741:print] fp16_enabled ................. True\r\n[2021-03-27 17:02:52,848] [INFO] [config.py:741:print] global_rank .................. 0\r\n[2021-03-27 17:02:52,848] [INFO] [config.py:741:print] gradient_accumulation_steps .. 1\r\n[2021-03-27 17:02:52,848] [INFO] [config.py:741:print] gradient_clipping ............ 1.0\r\n[2021-03-27 17:02:52,848] [INFO] [config.py:741:print] gradient_predivide_factor .... 1.0\r\n[2021-03-27 17:02:52,848] [INFO] [config.py:741:print] initial_dynamic_scale ........ 4294967296\r\n[2021-03-27 17:02:52,848] [INFO] [config.py:741:print] loss_scale ................... 0\r\n[2021-03-27 17:02:52,848] [INFO] [config.py:741:print] memory_breakdown ............. False\r\n[2021-03-27 17:02:52,848] [INFO] [config.py:741:print] optimizer_legacy_fusion ...... False\r\n[2021-03-27 17:02:52,848] [INFO] [config.py:741:print] optimizer_name ............... adam\r\n[2021-03-27 17:02:52,848] [INFO] [config.py:741:print] optimizer_params ............. {'lr': 5e-05, 'betas': [0.9, 0.999], 'eps': 1e-08, 'weight_decay': 0.0}\r\n[2021-03-27 17:02:52,848] [INFO] [config.py:741:print] pipeline ..................... {'stages': 'auto', 'partition': 'best', 'seed_layers': False, 'activation_checkpoint_interval': 0}\r\n[2021-03-27 17:02:52,848] [INFO] [config.py:741:print] pld_enabled .................. False\r\n[2021-03-27 17:02:52,848] [INFO] [config.py:741:print] pld_params ................... False\r\n[2021-03-27 17:02:52,848] [INFO] [config.py:741:print] prescale_gradients ........... False\r\n[2021-03-27 17:02:52,849] [INFO] [config.py:741:print] scheduler_name ............... WarmupLR\r\n[2021-03-27 17:02:52,849] [INFO] [config.py:741:print] scheduler_params ............. {'warmup_min_lr': 0, 'warmup_max_lr': 5e-05, 'warmup_num_steps': 0}\r\n[2021-03-27 17:02:52,849] [INFO] [config.py:741:print] sparse_attention ............. None\r\n[2021-03-27 17:02:52,849] [INFO] [config.py:741:print] sparse_gradients_enabled ..... False\r\n[2021-03-27 17:02:52,849] [INFO] [config.py:741:print] steps_per_print .............. 10\r\n[2021-03-27 17:02:52,849] [INFO] [config.py:741:print] tensorboard_enabled .......... False\r\n[2021-03-27 17:02:52,849] [INFO] [config.py:741:print] tensorboard_job_name ......... DeepSpeedJobName\r\n[2021-03-27 17:02:52,849] [INFO] [config.py:741:print] tensorboard_output_path ...... \r\n[2021-03-27 17:02:52,849] [INFO] [config.py:741:print] train_batch_size ............. 4\r\n[2021-03-27 17:02:52,849] [INFO] [config.py:741:print] train_micro_batch_size_per_gpu 4\r\n[2021-03-27 17:02:52,849] [INFO] [config.py:741:print] wall_clock_breakdown ......... False\r\n[2021-03-27 17:02:52,849] [INFO] [config.py:741:print] world_size ................... 1\r\n[2021-03-27 17:02:52,849] [INFO] [config.py:741:print] zero_allow_untested_optimizer False\r\n[2021-03-27 17:02:52,849] [INFO] [config.py:741:print] zero_config .................. {\r\n \"allgather_bucket_size\": 200000000.0,\r\n \"allgather_partitions\": true,\r\n \"contiguous_gradients\": true,\r\n \"cpu_offload\": true,\r\n \"cpu_offload_params\": false,\r\n \"cpu_offload_use_pin_memory\": false,\r\n \"elastic_checkpoint\": true,\r\n \"load_from_fp32_weights\": true,\r\n \"max_live_parameters\": 1000000000,\r\n \"max_reuse_distance\": 1000000000,\r\n \"overlap_comm\": true,\r\n \"param_persistence_threshold\": 100000,\r\n \"prefetch_bucket_size\": 50000000,\r\n \"reduce_bucket_size\": 200000000.0,\r\n \"reduce_scatter\": true,\r\n \"stage\": 2,\r\n \"sub_group_size\": 1000000000000\r\n}\r\n[2021-03-27 17:02:52,849] [INFO] [config.py:741:print] zero_enabled ................. True\r\n[2021-03-27 17:02:52,849] [INFO] [config.py:741:print] zero_optimization_stage ...... 2\r\n[2021-03-27 17:02:52,850] [INFO] [config.py:748:print] json = {\r\n \"fp16\":{\r\n \"enabled\":true,\r\n \"hysteresis\":2,\r\n \"loss_scale\":0,\r\n \"loss_scale_window\":1000,\r\n \"min_loss_scale\":1\r\n },\r\n \"gradient_accumulation_steps\":1,\r\n \"gradient_clipping\":1.0,\r\n \"optimizer\":{\r\n \"params\":{\r\n \"betas\":[\r\n 0.9,\r\n 0.999\r\n ],\r\n \"eps\":1e-08,\r\n \"lr\":5e-05,\r\n \"weight_decay\":0.0\r\n },\r\n \"type\":\"Adam\"\r\n },\r\n \"scheduler\":{\r\n \"params\":{\r\n \"warmup_max_lr\":5e-05,\r\n \"warmup_min_lr\":0,\r\n \"warmup_num_steps\":0\r\n },\r\n \"type\":\"WarmupLR\"\r\n },\r\n \"train_micro_batch_size_per_gpu\":4,\r\n \"zero_optimization\":{\r\n \"allgather_bucket_size\":200000000.0,\r\n \"allgather_partitions\":true,\r\n \"contiguous_gradients\":true,\r\n \"cpu_offload\":true,\r\n \"overlap_comm\":true,\r\n \"reduce_bucket_size\":200000000.0,\r\n \"reduce_scatter\":true,\r\n \"stage\":2\r\n }\r\n}\r\nUsing /home/oriy/.cache/torch_extensions as PyTorch extensions root...\r\nNo modifications detected for re-loaded extension module utils, skipping build step...\r\nLoading extension module utils...\r\nTime to load utils op: 0.0005881786346435547 seconds\r\n***** Running training *****\r\n Num examples = 287113\r\n Num Epochs = 3\r\n Instantaneous batch size per device = 4\r\n Total train batch size (w. parallel, distributed & accumulation) = 4\r\n Gradient Accumulation steps = 1\r\n Total optimization steps = 215337\r\n 0%| | 0/215337 [00:00<?, ?it/s][2021-03-27 17:02:53,333] [INFO] [stage2.py:1391:step] [deepspeed] fp16 dynamic loss scale overflow! Rank 0 Skipping step. Attempted loss scale: 4294967296, reducing to 4294967296\r\n 0%| | 1/215337 [00:00<26:38:16, 2.25it/s][2021-03-27 17:02:53,687] [INFO] [stage2.py:1391:step] [deepspeed] fp16 dynamic loss scale overflow! Rank 0 Skipping step. Attempted loss scale: 4294967296, reducing to 2147483648.0\r\n", "Next week I hope https://github.com/huggingface/transformers/pull/10753 will be finished, but for now here are the results on rtx-3090 24GB card with the unfinished zero-3 PR.\r\n\r\nAs you can see Deepspeed zero3's cpu offload is a way way more memory-efficient:\r\n\r\n```\r\n# baseline\r\n\r\nBS=4; CUDA_VISIBLE_DEVICES=0 PYTHONPATH=src USE_TF=0 python examples/seq2seq/run_translation.py --model_name_or_path t5-small --output_dir /tmp/zero3 --overwrite_output_dir --max_train_samples 64 --max_val_samples 64 --max_source_length 128 --max_target_length 128 --val_max_target_length 128 --do_train --num_train_epochs 1 --per_device_train_batch_size $BS --per_device_eval_batch_size $BS --learning_rate 3e-3 --warmup_steps 500 --predict_with_generate --logging_steps 0 --save_steps 0 --eval_steps 1 --group_by_length --adafactor --dataset_name wmt16 --dataset_config ro-en --source_lang en --target_lang ro --source_prefix \"translate English to Romanian: \" \r\n\r\n***** train metrics *****\r\n epoch = 1.0\r\n init_mem_cpu_alloc_delta = 3MB\r\n init_mem_cpu_peaked_delta = 0MB\r\n init_mem_gpu_alloc_delta = 230MB\r\n init_mem_gpu_peaked_delta = 0MB\r\n train_mem_cpu_alloc_delta = 60MB\r\n train_mem_cpu_peaked_delta = 0MB\r\n train_mem_gpu_alloc_delta = 231MB\r\n train_mem_gpu_peaked_delta = 226MB\r\n train_runtime = 3.619\r\n train_samples = 64\r\n train_samples_per_second = 4.421\r\n\r\n# zero2\r\n\r\n\r\nBS=4; PYTHONPATH=src USE_TF=0 deepspeed --num_gpus 1 examples/seq2seq/run_translation.py --model_name_or_path t5-small --output_dir /tmp/zero3 --overwrite_output_dir --max_train_samples 64 --max_val_samples 64 --max_source_length 128 --max_target_length 128 --val_max_target_length 128 --do_train --num_train_epochs 1 --per_device_train_batch_size $BS --per_device_eval_batch_size $BS --learning_rate 3e-3 --warmup_steps 500 --predict_with_generate --logging_steps 0 --save_steps 0 --eval_steps 1 --group_by_length --adafactor --dataset_name wmt16 --dataset_config ro-en --source_lang en --target_lang ro --source_prefix \"translate English to Romanian: \" --deepspeed examples/tests/deepspeed/ds_config_zero2.json\r\n\r\n\r\n***** train metrics *****\r\n epoch = 1.0\r\n init_mem_cpu_alloc_delta = 7MB\r\n init_mem_cpu_peaked_delta = 0MB\r\n init_mem_gpu_alloc_delta = 0MB\r\n init_mem_gpu_peaked_delta = 0MB\r\n train_mem_cpu_alloc_delta = 70MB\r\n train_mem_cpu_peaked_delta = 0MB\r\n train_mem_gpu_alloc_delta = 148MB\r\n train_mem_gpu_peaked_delta = 3559MB\r\n train_runtime = 5.0669\r\n train_samples = 64\r\n train_samples_per_second = 3.158\r\n\r\n\r\n# zero3\r\n\r\n\r\nBS=4; PYTHONPATH=src USE_TF=0 deepspeed --num_gpus 1 examples/seq2seq/run_translation.py --model_name_or_path t5-small --output_dir /tmp/zero3 --overwrite_output_dir --max_train_samples 64 --max_val_samples 64 --max_source_length 128 --max_target_length 128 --val_max_target_length 128 --do_train --num_train_epochs 1 --per_device_train_batch_size $BS --per_device_eval_batch_size $BS --learning_rate 3e-3 --warmup_steps 500 --predict_with_generate --logging_steps 0 --save_steps 0 --eval_steps 1 --group_by_length --adafactor --dataset_name wmt16 --dataset_config ro-en --source_lang en --target_lang ro --source_prefix \"translate English to Romanian: \" --deepspeed examples/tests/deepspeed/ds_config_zero3.json\r\n\r\n***** train metrics *****\r\n epoch = 1.0\r\n init_mem_cpu_alloc_delta = 7MB\r\n init_mem_cpu_peaked_delta = 0MB\r\n init_mem_gpu_alloc_delta = 0MB\r\n init_mem_gpu_peaked_delta = 0MB\r\n train_mem_cpu_alloc_delta = 71MB\r\n train_mem_cpu_peaked_delta = 0MB\r\n train_mem_gpu_alloc_delta = -52MB\r\n train_mem_gpu_peaked_delta = 244MB\r\n train_runtime = 7.6324\r\n train_samples = 64\r\n train_samples_per_second = 2.096\r\n```\r\n\r\nThe config files are from the PR I linked to in the first para.\r\n\r\nSo please give us a few more days - this is also depending on deepspeed merging several PRs and making a new release.\r\n", "I suspect my cpu memory profiling functions are missing some allocations, which is odd. Surely, there must be more cpu memory used with cpu_offload. I will investigate this. \r\n\r\nSuspecting that `tracemalloc` doesn't tracks c++ allocations, which is what deepspeed does. might have to switch to sampling, but python threads's GIL is a big problem to get correct results.\r\n\r\n**edit:** this should fix it: https://github.com/huggingface/transformers/pull/10937", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored." ]

[]

[]

[ "Thanks for updating, this looks great!" ]

[]

[]

[ "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored." ]

[]

[]

[ "Indeed, do you want to open a PR with the fix since you found it?\r\n\r\nPS: I didn't know that \\`\\`\\`diff feature, it's soooo pretty 🤩 !", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored." ]

[]

[]

[ "> LGTM! Should the images be added before merge?\r\n\r\nAre you referring to container images? they will be added after the release of `transformers`. If you are referring to the `TODO: Add a screenshot of PR + Text template to make it easy to open.` nope I would add them as soon as we went through the process. To make screenshots while doing it. " ]

[]

[]

[ "Hello! Do you get the same error when not installing PyInstaller, instead using `haystack` in a virtual environment?", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored.", "Any update on this one?" ]

[]

[]

[ "I don't think Longformer is supported on TPU, @patrickvonplaten will confirm.", "@sgugger Thanks!\r\nLooking forward to @patrickvonplaten confirmation. ", "Hey @mabdullah1994, yeah `Longformer` is sadly not yet supported on TPU. We just merged Big Bird: https://huggingface.co/transformers/master/model_doc/bigbird.html though, which should work on TPU. It would be amazing if you could try it out :-)", "@patrickvonplaten Thanks for the update Patrick!\r\nJust a quick query: I have a dataset with large sequences and I don't want to truncate the text. What options do I have? Will XLNet be able to handle large sequences with pre-trained models? Could you point me towards an example of using stride for this use case? Thanks!", "Well, tried `BigBird` and getting a similar error on Google Colab\r\n\r\n```\r\nRuntimeError: torch_xla/csrc/tensor_methods.cpp:880 : Check failed: xla::ShapeUtil::Compatible(shapes.back(), tensor_shape) \r\n*** Begin stack trace ***\r\n\ttensorflow::CurrentStackTrace()\r\n\ttorch_xla::XLATensor::cat(absl::lts_2020_02_25::Span<torch_xla::XLATensor const>, long)\r\n\ttorch_xla::AtenXlaType::cat(c10::ArrayRef<at::Tensor>, long)\r\n\tc10::impl::wrap_kernel_functor_unboxed_<c10::impl::detail::WrapFunctionIntoRuntimeFunctor_<at::Tensor (*)(c10::ArrayRef<at::Tensor>, long), at::Tensor, c10::guts::typelist::typelist<c10::ArrayRef<at::Tensor>, long> >, at::Tensor (c10::ArrayRef<at::Tensor>, long)>::call(c10::OperatorKernel*, c10::ArrayRef<at::Tensor>, long)\r\n\t\r\n\tat::cat(c10::ArrayRef<at::Tensor>, long)\r\n\t\r\n\t\r\n\t\r\n\tat::cat(c10::ArrayRef<at::Tensor>, long)\r\n\t\r\n\t_PyMethodDef_RawFastCallKeywords\r\n\t_PyCFunction_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyFunction_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyObject_Call_Prepend\r\n\tPyObject_Call\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyObject_Call_Prepend\r\n\t\r\n\t_PyObject_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyObject_Call_Prepend\r\n\tPyObject_Call\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyObject_Call_Prepend\r\n\t\r\n\t_PyObject_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyObject_Call_Prepend\r\n\tPyObject_Call\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyObject_Call_Prepend\r\n\t\r\n\t_PyObject_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyObject_Call_Prepend\r\n\tPyObject_Call\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyObject_Call_Prepend\r\n\t\r\n\t_PyObject_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyObject_Call_Prepend\r\n\tPyObject_Call\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyObject_Call_Prepend\r\n\t\r\n\t_PyObject_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyObject_Call_Prepend\r\n\tPyObject_Call\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyObject_Call_Prepend\r\n\t\r\n\tPyObject_Call\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyFunction_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyFunction_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyFunction_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyFunction_FastCallDict\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyFunction_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyFunction_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\tPyEval_EvalCode\r\n\t\r\n\t_PyMethodDef_RawFastCallKeywords\r\n\t_PyCFunction_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyFunction_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyFunction_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyObject_Call_Prepend\r\n\tPyObject_Call\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyFunction_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyFunction_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyFunction_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyFunction_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyFunction_FastCallDict\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyFunction_FastCallDict\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyFunction_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyFunction_FastCallKeywords\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyObject_Call_Prepend\r\n\tPyObject_Call\r\n\t_PyEval_EvalFrameDefault\r\n\t_PyEval_EvalCodeWithName\r\n\t_PyFunction_FastCallKeywords\r\n*** End stack trace ***\r\ns64[1,1,1]{2,1,0} vs. f32[1,1,1]{2,1,0}\r\n```", "Hey @mabdullah1994, \r\n\r\nCould you maybe open a new issue showcasing that big bird doesn't work on PyTorch/XLA? :-)", "Hey @patrickvonplaten \r\n\r\nJust created a new issue #11363 with the details of the BigBird issue. Please advice. Thanks!", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored.", "Any updates on this?", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored.", "hey @patrickvonplaten, with the release of the new trainer should this issue be resolved. I'm using the latest version of transformers and still getting this for models like [allenai/led-base-16384](https://huggingface.co/allenai/led-base-16384) running on TPU." ]

[]

[ { "login": "patrickvonplaten", "id": 23423619, "node_id": "MDQ6VXNlcjIzNDIzNjE5", "avatar_url": "https://avatars.githubusercontent.com/u/23423619?v=4", "gravatar_id": "", "url": "https://api.github.com/users/patrickvonplaten", "html_url": "https://github.com/patrickvonplaten", "followers_url": "https://api.github.com/users/patrickvonplaten/followers", "following_url": "https://api.github.com/users/patrickvonplaten/following{/other_user}", "gists_url": "https://api.github.com/users/patrickvonplaten/gists{/gist_id}", "starred_url": "https://api.github.com/users/patrickvonplaten/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/patrickvonplaten/subscriptions", "organizations_url": "https://api.github.com/users/patrickvonplaten/orgs", "repos_url": "https://api.github.com/users/patrickvonplaten/repos", "events_url": "https://api.github.com/users/patrickvonplaten/events{/privacy}", "received_events_url": "https://api.github.com/users/patrickvonplaten/received_events", "type": "User", "site_admin": false } ]

[ "If I train the model on the crime-and-punishment\r\n```shell\r\npython examples/language-modeling/run_clm.py --model_name_or_path google/reformer-crime-and-punishment --dataset_name crime_and_punish --do_train --do_eval --output_dir /home2/zhenggo1/checkpoint/reformer_clm\r\n```\r\nthe bug is below\r\n```python\r\nAssertionError: If training, make sure that config.axial_pos_shape factors: (512, 1024) multiply to sequence length. Got prod((512, 1024)) != sequence_length: 1024. You might want to consider padding your sequence length to 524288 or changing config.axial_pos_shape.\r\n```", "Hi,\r\n\r\nI have been playing around with Reformer these few days so I hope I can give some insights. Axial positional encoding in Reformer requires that sequence length must be fixed to the product of `axial_pos_embds_dim`. See the documentation here\r\n\r\nhttps://huggingface.co/transformers/model_doc/reformer.html#axial-positional-encodings\r\n\r\nSo you have to either pad the sequence length to that fixed size, or change the value for `axial_pos_embds_dim` to a smaller value. Due to this reason, I believe example scripts won't work with Reformer out of the box.\r\n\r\nThe Reformer examples from Google's Trax actually don't use axial positional encoding, just normal positional encoding (see [here](https://github.com/google/trax/blob/master/trax/examples/NER_using_Reformer.ipynb)). So I actually disable axial positional encoding (passing `axial_pos_embds=False` to Reformer config) and it works fine. By disabling this, I can also use dynamic padding (pad to max length within a batch) and saves even more memory.\r\n\r\nI haven't tested the accuracy difference between with and without axial positional encoding. But axial positional encoding is so slow for a dataset with varying sequence lengths that I find it impractical.", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored." ]

[]

[]

[ "Maybe @polm or @singletongue have an idea!", "I didn't implement BertTokenizer so I'm a little out of my depth here, but the code below in a clean environment worked fine for me with no weird hashes.\r\n\r\n```\r\nfrom transformers import BertJapaneseTokenizer\r\n\r\nname = \"cl-tohoku/bert-base-japanese-whole-word-masking\"\r\nname = \"cl-tohoku/bert-base-japanese\"\r\ntokenizer = BertJapaneseTokenizer.from_pretrained(name)\r\n\r\ntext = \"テレビでサッカーの試合を見る。\"\r\n\r\nout = tokenizer.tokenize(text)\r\nprint(out)\r\n```\r\n\r\nI will note it is especially weird that the last word in your list (`。`) doesn't have the hashes.", "Thank you for you reply.\r\nHere is the result.\r\n['テレビ', '##で', '##サッカー', '##の', '##試', '##合', '##を', '##見', '##る', '。']\r\n", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored.", "@leoxu1007\r\nCould it be possible that you set `word_tokenizer_type` to `basic` ?\r\nI reproduced the same result by this configuration.\r\nI mean I got ['テレビ', '##で', '##サッカー', '##の', '##試', '##合', '##を', '##見', '##る', '。'].\r\n\r\nNow, `BertJapaneseTokenizer` pretrained tokenizer's default configuration is `word_tokenizer_type='mecab'`.\r\nSo we don't usually get this unexpected result.\r\nI tried the example with `mecab` I got ['テレビ', 'で', 'サッカー', 'の', '試合', 'を', '見る', '。'].\r\n" ]

[]

[]

[ "@sgugger Please review" ]

[]

[]

[ "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored." ]

[]

[]

[ "I'm also facing the same issue. Did you find any fix yet . ??", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored.", "I'm also facing the same issue. Did you guys find any fix yet . ??", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored.", "Same problem here, please write if someone found a valid solution.", "Facing same error", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored.", "Hi, I've had the same error but with `roberta-base`. It appeared that I had an empty folder named `roberta-base` in my working directory. Removing it solved the issue.", "I found this issue is caused by setting cache directory using checkpoint name\r\nTrainingArguments(checkpoint,evaluation_strategy='steps')\r\n\r\nchange checkpoint to something else resolve the issue\r\n", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored.", "> #353\r\n\r\nGot the same issue, thanks for reporting it here. Was able to fix it following after going through your comment.", "> Hi, I've had the same error but with `roberta-base`. It appeared that I had an empty folder named `roberta-base` in my working directory. Removing it solved the issue.\r\n\r\nYou are literally an angel." ]

[]

[]

[ "Hi,\r\n\r\nIs it possible to ask questions related to training on the [forum](https://discuss.huggingface.co/) rather than here? For example, all questions related to training LongFormer can be found [here](https://discuss.huggingface.co/search?q=longformer).\r\n\r\nThe authors of Transformers like to keep Github issues for bugs/feature requests.\r\n\r\nThank you. ", "sure. thank you for the quick response" ]

[]

[]

[ "You should install Transformers from source. See #10446.", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored." ]

[ { "id": 1905493434, "node_id": "MDU6TGFiZWwxOTA1NDkzNDM0", "url": "https://api.github.com/repos/huggingface/transformers/labels/dependencies", "name": "dependencies", "color": "0366d6", "default": false, "description": "Pull requests that update a dependency file" } ]

[]

[ "Looks like pyyaml is up-to-date now, so this is no longer needed." ]

[]

[]

[]

[]

[]

[ "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored." ]

[ { "id": 2796628563, "node_id": "MDU6TGFiZWwyNzk2NjI4NTYz", "url": "https://api.github.com/repos/huggingface/transformers/labels/WIP", "name": "WIP", "color": "234C99", "default": false, "description": "Label your PR/Issue with WIP for some long outstanding Issues/PRs that are work in progress" } ]

[]

[ "The `max_length` and `min_length` are in terms of tokens, not words. As some words consist of multiple tokens, this results in fewer words to be generated than you might expect. ", "@NielsRogge\r\nThanks for the answer. It makes sense. But when are words consist of multiple tokens, can you give me some examples?\r\n\r\nAlso, would it be better for arguments (max_length, min_length) refer to number of words instead of tokens as to better control the outputs, which are natural language for human?", "Running into a similar issue when using `generator = pipeline('text-generation', model='EleutherAI/gpt-neo-2.7B')` . I can get better control when using `min_length=..,max_length=..` but I have no ultimate control when e.g. querying for `Below is the code for a react app with a blue button that says 'click me'`\r\n\r\n\r\n```\r\n{'generated_text': \"Below is the code for a react app with a blue button that says 'click me' that is to be used by react-router. \\nimport React, { Component } from 'react';\\n\\nimport { Link } from 'react\"}]\r\n```\r\n\r\n\r\nMy result is \"cut off\" and I would be very happy to set a desired length of resulting words.", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored.", "Stalebots are so much an anti-quality thing :-/", "> Running into a similar issue when using `generator = pipeline('text-generation', model='EleutherAI/gpt-neo-2.7B')` . I can get better control when using `min_length=..,max_length=..` but I have no ultimate control when e.g. querying for `Below is the code for a react app with a blue button that says 'click me'`\r\n> \r\n> ```\r\n> {'generated_text': \"Below is the code for a react app with a blue button that says 'click me' that is to be used by react-router. \\nimport React, { Component } from 'react';\\n\\nimport { Link } from 'react\"}]\r\n> ```\r\n> \r\n> My result is \"cut off\" and I would be very happy to set a desired length of resulting words.\r\n\r\nSame issue for me, anyone found a solution regarding this? ", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored.", "***Stalebots are so much an anti-quality measure and have not been fixed***", "cc @patil-suraj @patrickvonplaten ", "@chris-aeviator - do you want to have exactly `max_length` words? In this case you have to disable the eos_token_id => you should be able to just do `model.generate(...., eos_token_id=None)`" ]

[]

[ { "login": "LysandreJik", "id": 30755778, "node_id": "MDQ6VXNlcjMwNzU1Nzc4", "avatar_url": "https://avatars.githubusercontent.com/u/30755778?v=4", "gravatar_id": "", "url": "https://api.github.com/users/LysandreJik", "html_url": "https://github.com/LysandreJik", "followers_url": "https://api.github.com/users/LysandreJik/followers", "following_url": "https://api.github.com/users/LysandreJik/following{/other_user}", "gists_url": "https://api.github.com/users/LysandreJik/gists{/gist_id}", "starred_url": "https://api.github.com/users/LysandreJik/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/LysandreJik/subscriptions", "organizations_url": "https://api.github.com/users/LysandreJik/orgs", "repos_url": "https://api.github.com/users/LysandreJik/repos", "events_url": "https://api.github.com/users/LysandreJik/events{/privacy}", "received_events_url": "https://api.github.com/users/LysandreJik/received_events", "type": "User", "site_admin": false } ]

[ "@LysandreJik - you'll see that the last test (marked 'slow') in tests/test_modeling_megatron_bert.py points to a checkpoint in examples/megatron-models (downloaded following the instructions described in examples/megatron-models/README.md). I was not sure how to deal with that so suggestions are welcome (for other items too ;)).", "We have a few failing tests, let me break them down for you:\r\n\r\n### build_doc\r\n\r\nThe build_doc is failing because of the following errors:\r\n```\r\n/home/circleci/transformers/docs/source/model_doc/megatron_bert.rst:document isn't included in any toctree\r\n```\r\n\r\nThe `megatron_bert.rst` should be defined in the index of the docs :)\r\n\r\n### check_code_quality\r\n\r\nThe error is:\r\n```\r\n2 files would be reformatted, 783 files would be left unchanged.\r\n```\r\n\r\nFor this, you should install the quality tools: `pip install -e .[quality]` (from the root of the repo)\r\nand run the following:\r\n```\r\nmake fixup\r\n```\r\n\r\nThis is going to fix some files, and tell you if there are errors it cannot resolve. If there are some, it should tell you how to fix them.\r\n\r\n### run_test_flax, run_tests_tf, and run_tests_pipelines_tf\r\n\r\nThis is due to the following error:\r\n\r\n```\r\n____________ ERROR collecting tests/test_modeling_megatron_bert.py _____________\r\ntests/test_modeling_megatron_bert.py:256: in <module>\r\n class MegatronBertModelTest(ModelTesterMixin, unittest.TestCase):\r\ntests/test_modeling_megatron_bert.py:259: in MegatronBertModelTest\r\n MegatronBertModel,\r\nE NameError: name 'MegatronBertModel' is not defined\r\n```\r\n\r\nI think this comes from a missing `@require_torch` decorator on one of your tests. This decorator tells the suite that this test requires torch, and to not run this test if torch is not found as a dependency inside the environment.\r\n\r\nIf that's not it, then it may be that it's missing a dummy object. Running `make fix-copies` should fix this, but you should already have run this if you have done the fix mentioned above relative to the style/code quality." ]

[]

[]

[]

[]

[ { "login": "sgugger", "id": 35901082, "node_id": "MDQ6VXNlcjM1OTAxMDgy", "avatar_url": "https://avatars.githubusercontent.com/u/35901082?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sgugger", "html_url": "https://github.com/sgugger", "followers_url": "https://api.github.com/users/sgugger/followers", "following_url": "https://api.github.com/users/sgugger/following{/other_user}", "gists_url": "https://api.github.com/users/sgugger/gists{/gist_id}", "starred_url": "https://api.github.com/users/sgugger/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sgugger/subscriptions", "organizations_url": "https://api.github.com/users/sgugger/orgs", "repos_url": "https://api.github.com/users/sgugger/repos", "events_url": "https://api.github.com/users/sgugger/events{/privacy}", "received_events_url": "https://api.github.com/users/sgugger/received_events", "type": "User", "site_admin": false } ]

[ "I think the solution you suggested on the forums (leaving the length computation up to the user in the dataset) is a bit better. Will investigate a bit more.", "> I think the solution you suggested on the forums (leaving the length computation up to the user in the dataset) is a bit better. Will investigate a bit more.\r\n\r\nI agree, that surely is a good solution if the user is aware of it. It just took us a while to find the source of that delay before training. Maybe we can have some sort of user hint? ", "Fixed by #10953" ]

[]

[]

[ "### Update:\r\n\r\n**I was wrong, my error is originally coming from badly assigned TrainingArguments.label_names. However, I strongly recommend fixes.**\r\n\r\nContinued investigating and realised my error is appearing because I don't understand the attribute `TrainingArguments.label_names`. I thought that it should be a list of strings, where each string is my specified name of a class.\r\n\r\n**Some background on my code/data structure:**\r\n\r\nI'm doing Multi-Class classification on sentences and my training data is an ordered list of a string sentences, and an associated list of classes in string form (i.e. the class's name). I.e. `sentences[i]` is a sentence sample, and `classes[i]` is the class of that sentence, more specifically the _name_ of that class, a string.\r\n\r\nIf I understand the HuggingFace - Transformers documentation correctly, I should be passing these classes to the Model as the `indices` to a One-Hot encoded list of classes instead. Or in another interpretation, class[i] should be a class number. So in my custom Dataset class, I send my sentences through a Transformers Tokenizer, and I use `MultiLabelBinarizer` from scikit-learn to One-Hot-Encode all my classes, convert it to a tensor, and then call argmax(dim=-1) on the `classes` tensor.\r\n\r\nOf course, I don't want the metrics report to just say \"class 0 has the f1-score X\", so I thought I could pass original class names to Trainer to be used when printing this metrics report. This is what I thought `TrainingArgument.label_names` was for, so I set `TrainingArguments.label_names = MultiLabelBinarizer.classes_`.\r\n\r\n**How my error appears**\r\n\r\nDuring the evaluation stage in `prediction_step()`, I saw with pdb that `inputs` indeed has a `labels` item, which makes sense as I'm trying to evaluate how well the model is performing here.\r\n\r\nI now understand that `prediction_step()` is used both when **predicting** _(i.e. we have no target labels in `inputs` and thus expect to not obtain a loss value)_, and for **evaluating** _(i.e. we **do** have target labels and should be able to get a loss value)_.\r\n\r\nAnd of course, this is what the `has_labels` variable is used for - indicating the presence of target labels in `input`, and thus that `prediction_step()` should be able to get a loss value using `Trainer.compute_loss`. However if `has_labels=False`, `prediction_step()` assumes that `outputs` variable **will not** have the item `loss`, and so we do not need to ignore this key when converting the `outputs` dict/SequenceClassificationOutput.\r\n\r\n\r\nHowever, since I apparently specified `TrainingArguments.label_names` incorrectly, `has_labels` becomes False _when it shouldn't be_, and everything gets messed up. `prediction_step()` thus assumes that we're predicting and doesn't filter out `loss` in outputs, which leads to the error described in my first post.\r\n\r\nI still don't understand what `TrainingArguments.label_names` is or should be.\r\nI recommend that two things should be done:\r\n\r\n- Improve the documentation regarding `TrainingArguments.label_names`, specifying the expected format, etc. \r\n- Earlier in the `Trainer` class, check that `TrainingArguments.label_names` is reasonably formatted and raise a specific error if it isn't, so the user doesn't recieve the above extremely confusing rabbit hole of an error that I did.. ", "Ping @sgugger ", "I'm unsure what you want to improve, the [documentation](https://huggingface.co/transformers/main_classes/trainer.html#trainingarguments) states the following:\r\n\r\n\"\"\"\r\nlabel_names (List[str], optional) –\r\n\r\nThe list of keys in your dictionary of inputs that correspond to the labels.\r\n\r\nWill eventually default to [\"labels\"] except if the model used is one of the XxxForQuestionAnswering in which case it will default to [\"start_positions\", \"end_positions\"].\r\n\"\"\"\r\n\r\nIt's clearly indicated it needs to be a list of strings, that have to be the names of the keys for the labels in your input dictionary. I have no idea what your `MultiLabelBinarizer.classes_` contains since you did not share the code of this class.\r\n\r\nMore generally, please look at the [warning in the Trainer documentation](https://huggingface.co/transformers/main_classes/trainer.html#trainer) since you appear to be using the Trainer with a model that is not a model of the Transformers library.\r\n", "Oh alright, I didn't see that warning. Thank you!\r\n\r\nThe `MultiLabelBinarizer` from scikit-learn transforms list of class/label strings into a matrix, where each row is a one-hot-encoded version of the label. `MultiLabelBinarizer.classes_` returns the list of all class/label names detected in the original class list, with same ordering as the one-hot-encoded version.\r\n\r\nIt sounds like I understood `TrainingArguments.label_names` correctly then, but that my usage of a custom model is messing up the behaviour somehow. Are there any tips/strategies to fix these strange behaviours? Should I just override `prediction_step` and try to fix how has_labels is being assigned?", "The easiest way to have your custom model work with `Trainer` with no strange behavior is to subclass `PreTrainedModel` (for instance by copying the `XXXForSequenceClassification` and tweaking it to your needs). Otherwise, subclassing and overriding the `prediction_step` method is the most straightforward path.", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored.", "I ran into exactly the same issue today.\r\nI was also thinking that the parameter `label_names` in `TrainingArguments` refers to `data[\"train\"].features[\"label\"].names`.\r\nThe error message `IndexError: tuple index out of range` was not helpful at all and I only found the problem by trial and error.\r\n\r\nActually, I was not able to find the description for `label_names` in the [documentation](https://huggingface.co/docs/transformers/v4.14.1/en/main_classes/trainer#transformers.TrainingArguments) but only in the linked source code.\r\n\r\nBesides, I don't even understand what \"The list of keys in your dictionary of inputs that correspond to the labels.\" should mean.\r\n\r\nWhat \"dictionary of inputs\" and what \"list of keys\"?\r\n\r\nMy dataset looks like this\r\n```\r\nDatasetDict({\r\n train: Dataset({\r\n features: ['text', 'label'],\r\n num_rows: 9245\r\n })\r\n test: Dataset({\r\n features: ['text', 'label'],\r\n num_rows: 1028\r\n })\r\n})\r\n```\r\nThe only dictionaries I see is `DatasetDict` with keys \"train\" and \"test\" and each `Dataset` with keys \"features\" and \"num_rows\".\r\n\r\nIt would be really helpful if the description of the parameter `label_names` and the error message could be improved.", "YES I completely agree. This was very confusing in the documentation. I also interpreted it to mean the list of keys in my label2id dictionary, but it turns out that all it wanted for `label_names` in my case was `['labels']`. That is the name of the column in my input that holds the labels. I hope this helps anyone who was still struggling to understand what \"a list of the names of the keys for the labels in your input dictionary\" means :)" ]

[]

[]

[ "same problem.\r\n\r\nEdit: I updated the package \"tokenizers\" to the latest version and it works fine.", "Hi! Do you have a colab so we can reproduce the issue? Or some commands we can run to obtain the same environment you have and test it out?", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored." ]

[]

[]

[]

[]

[]

[]

[]

[]

[ "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored." ]

[]

[]

[ "Hey @lexhuismans,\r\n\r\nThanks a lot for your PR! Could you also add a test to verify that T5 can be used with a 3D mask?\r\n", "Hey @patrickvonplaten,\r\n\r\nThanks for your message. I had a shot in adding a test for the 3D attention mask. The test passed on my device. \r\n\r\nI based the test on a similar test for a default attention mask in test_modeling_bert.py. (Not sure if bert already tests for 3D attention mask?) \r\n\r\nAlso, I did a rebase before pushing which is why there are so many other commits in-between. \r\n\r\nLet me know if something is still missing or incorrect so I can have a look at it. ", "I made a new PR with just the two commits #11197. " ]

[]

[]

[ "Ok, added the missing part in Accelerate. In your code, before gathering the labels and predictions, you should pad them across the processes running if `pad_to_max_length` is False:\r\n```python\r\nif not args.pad_to_max_length:\r\n predictions = accelerator.pad_across_processes(predictions, dim=1, pad_index=-100)\r\n labels = accelerator.pad_across_processes(batch[\"labels\"], dim=1, pad_index=-100)\r\n```\r\n\r\nThis should solve the issue when `pad_to_max_length` is left at `False`!", "Thanks for your comment, @sgugger. I fix the bugs and add documentation to README.", "Thank you for spotting the mistake with unintentionally deleted `--label_all_tokens` from argarser. I added that argument back.", "Just checked on TPU for completeness and it runs perfectly fine there, so we're all good!\r\n\r\nThanks for your contribution!!!" ]

[]

[]

[ "Related issue: https://github.com/huggingface/transformers/issues/9147, with proposed fix in https://github.com/huggingface/transformers/pull/9807", "Why do we need this condition?\r\nhttps://github.com/huggingface/transformers/blob/1a3e0c4fe6868b4eb1105dfe601a79d7e5d11a0f/src/transformers/tokenization_utils_base.py#L1669-L1672\r\nWas introduced here: https://github.com/huggingface/transformers/commit/863e553f75daeaf09aea9cd521ac3a3b3f09e29f Is it needed in any other scenario?\r\nWould it be better to do `unresolved_files.append(file_id)` unconditionally?", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored." ]

[]

[]

[ "Yes, if you place all files which you find on the model page on the hub in a directory, then it will work.", "You can even just `git clone` the model repo", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored." ]

[]

[]

[]

[]

[]

[ "Creating a quick PR now to alert to some issues I'm encountering, will backfill with proper GitHub issues after I knock off from work.\r\n\r\nOther issues:\r\n- [ ] If task_name is `None`, `metrics` is undefined, and will throw `... metrics used before assignment`" ]

[]

[]

[ "Yes, the name should indeed be updated! If you want to do a PR with this, please go ahead!" ]

[]

[ { "login": "sgugger", "id": 35901082, "node_id": "MDQ6VXNlcjM1OTAxMDgy", "avatar_url": "https://avatars.githubusercontent.com/u/35901082?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sgugger", "html_url": "https://github.com/sgugger", "followers_url": "https://api.github.com/users/sgugger/followers", "following_url": "https://api.github.com/users/sgugger/following{/other_user}", "gists_url": "https://api.github.com/users/sgugger/gists{/gist_id}", "starred_url": "https://api.github.com/users/sgugger/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sgugger/subscriptions", "organizations_url": "https://api.github.com/users/sgugger/orgs", "repos_url": "https://api.github.com/users/sgugger/repos", "events_url": "https://api.github.com/users/sgugger/events{/privacy}", "received_events_url": "https://api.github.com/users/sgugger/received_events", "type": "User", "site_admin": false } ]

[ "The checkpoints are saved to resume training in case you are interrupted, so saving only the best checkpoints wouldn't work with this.\r\n\r\nThe `load_best_model_at_end` functionality already keeps track of the best checkpoint during training and reloads it at the end, I think it should cover what you need.", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored.", "So，when I finish training，how can I load the best performing checkpoint ？ @sgugger ", " When I check the `trainer_state.json` file I found these mssage:\r\n```\r\n \"best_metric\": null,\r\n \"best_model_checkpoint\": null,\r\n \"epoch\": 100.0,\r\n \"global_step\": 559300,\r\n \"is_hyper_param_search\": false,\r\n \"is_local_process_zero\": true,\r\n \"is_world_process_zero\": true,\r\n```\r\n\r\nas shown above，\"best_model_checkpoint\" is null.\r\n\r\n", "If you did not specify `--load_best_model_at_end` for your script, you won't get it automatically." ]

[]

[]

[ "Hey @joe32140,\r\n\r\nSorry, I think this PR already fixes the problem: https://github.com/huggingface/transformers/pull/10906/files" ]

[]

[]

[ "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored.", "I have the same error.", "Me too" ]

[]

[]

[ "I'm not sure you are aware but the `Trainer` can take a `model_init` parameter that... well... creates the model ;-) Have you explored how it could help with this particular problem?\r\n\r\nThe changes in the other parts of the lib look reasonable to me at first glance.", "Thanks for the very detailed summary @stas00! All of the changes you propose make sense. The changes to `from_pretrained` look inevitable, and the approach you propose looks like it does the job without being invasive in other parts of the library that we want to keep readable like the model files.\r\n\r\nI know the API isn't final and prone to changes, but could we imagine a flag like `deepspeed_aware_instantiation` or `deepspeed_partitioning` in the `from_pretrained` method, rather than a `deepspeed_is_zero3_enabled(True)`?\r\nI think this would be more in line with how we manage things in the library from the user's perspective (which is principally through kwargs). I know none of this is final, but thinking of the API beforehand doesn't sound like a bad idea before everything is implemented :)", "> I'm not sure you are aware but the `Trainer` can take a `model_init` parameter that... well... creates the model\r\n\r\nI need trainer init to complete before `model_init` is called then - i.e. I need the fully initialized Trainer object inside `model_init`. \r\n\r\nthe model init will depends on how the training is expected to run, so I guess we need a sort of trainer - pre-init :)", "> I know the API isn't final and prone to changes, but could we imagine a flag like `deepspeed_aware_instantiation` or `deepspeed_partitioning` in the `from_pretrained` method, rather than a `deepspeed_is_zero3_enabled(True)`?\r\n> I think this would be more in line with how we manage things in the library from the user's perspective (which is principally through kwargs). I know none of this is final, but thinking of the API beforehand doesn't sound like a bad idea before everything is implemented :)\r\n\r\nAbsolutely, this would be ideal. \r\n\r\nthe problem with this approach is that for example we will have to change all examples to support new arguments to `.from_pretrained`, that's why I am trying to make it work transparently.\r\n\r\nBut the examples will still have to call `deepspeed_is_zero3_enabled()` and just pass the result to `.from_pretrained`...\r\n\r\nBut we could support both - if the new argument is there we use it, if it's not there as a fallback we check the global state helper function. I'm trying to solve this on the global level and not just on the use-case where the user calls each function explicitly (which is not the case in examples, but of course we could change them too.)\r\n\r\nI'm totally not attached to the proposed way, we can choose what resonates the most.\r\n\r\nThank you for your feedback, @LysandreJik and @sgugger ", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored." ]

[ { "id": 1897896961, "node_id": "MDU6TGFiZWwxODk3ODk2OTYx", "url": "https://api.github.com/repos/huggingface/transformers/labels/Migration", "name": "Migration", "color": "e99695", "default": false, "description": "" } ]

[]

[ "There is no `BertLayerNorm` anymore since all it was adding has been ported to main PyTorch. The BERT model is now jusing `torch.nn.LayerNorm`.\r\n\r\nSo to make your code working, instead of trying to import it from transformers, just define it as:\r\n```\r\nBertLayerNorm = torch.nn.LayerNorm\r\n```", "@sgugger your suggestion resolved the issue. Thanks!", "Closing then :-)" ]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[]

[ "Checked all the slow tests run. I have no idea on how to implement a test that checks the model is only loaded once, so I'm going merge this and if anyone wants to tackle that, it can be done in a separate PR." ]

[]

[]

[ "Please note: This checkpoint uses BertModel with XLMRobertaTokenizer so AutoTokenizer won't work with this checkpoint!" ]

[]

[]

[]

[]

[]

[ "I tried adding\r\n```python\r\ndel training_args\r\ndel trainer\r\ndel model_config\r\ndel run\r\ngc.collect()\r\nth.cuda.empty_cache()\r\n```\r\nto the end of each loop, but it does not seem to change anything. ", "I think the memory problem comes from the wandb integration. I do not see the problem without it: memory resets at 0 at each new step of the loop and goes back to the same max value.", "Use torc with no grad inside for loop", "Seems like the same problem occurs with wandb's sweeps, so it looks like a wandb problem more than a huggingface one. I can't use wandb then, sucks :/", "cc @borisdayma so you are aware." ]

[]

[]

[ "It seems that your training epochs is set to 1500. Set it up to 5 for a quick trial", "It is giving training loss of more than a 100 in that case!", "Is your training finished or is still running? If it is the second, just try again with less number of epochs for example", "This is the output for 5 epochs:\r\n\r\n\r\nTrainOutput(global_step=10, training_loss=93.45169677734376, metrics={'train_runtime': 48.9011, 'train_samples_per_second': 0.204, 'total_flos': 2.6027104384512e+16, 'epoch': 5.0, 'init_mem_cpu_alloc_delta': 348007, 'init_mem_gpu_alloc_delta': 377847808, 'init_mem_cpu_peaked_delta': 18306, 'init_mem_gpu_peaked_delta': 0, 'train_mem_cpu_alloc_delta': 706705, 'train_mem_gpu_alloc_delta': 1120621568, 'train_mem_cpu_peaked_delta': 161498645, 'train_mem_gpu_peaked_delta': 7221921792})", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored." ]

[]

[]

[]

[]

[]

[ "Found solution from [#10446](https://github.com/huggingface/transformers/issues/10446).\r\n\r\nShould follow this step instead:\r\n`git clone https://github.com/huggingface/transformers`\r\n\r\n`cd transformers`\r\n\r\n`pip install .`\r\n\r\n" ]

[]

[]

[ "I have no written the `MLFlowCallback` (external integrations are maintained by contributors or the authors of the external libraries themselves) but I can confirm the command will indeed not log the model weights. The callback does get the model in the kwargs, so it's completely possible to get it from there and upload it, like it's done in the `WandbCallback`.", "Thanks @sgugger for your fast reply! \r\n\r\nI tested your suggestion, but it doesn't quite work. I don't know about the `WandbCallback`, but in `MLFlowCallback` in the `on_train_end`, when the model is saved and logged to `mlflow`, the `mlflow` run is ended, and the later loggin of metrics, like evaluation and testing are logged in separate run, which is not what we want.\r\nI don't know if this happens when you create `fake_trainer` with `fake_trainer = Trainer(args=args, model=model, tokenizer=tokenizer)` or when the artifacts are logged with `self._ml_flow.log_artifacts(temp_dir)`.\r\n\r\nAlso the current `MLFlowCallback` doesn't log testing metrics. I know this is not a fault of the callback, it is in the `trainer.predict()` method that doesn't have call to `log()` internally. The workaround is to call `trainer.log(metrics)` after \r\n```\r\ntrainer.log_metrics(\"test\", metrics)\r\ntrainer.save_metrics(\"test\", metrics)\r\n```\r\nin the example. ", "As I said, I haven't written that callback: integrations with reporting platforms are entirely maintained by the developers of those integrations or the community. You can open a PR with your fixes!", "I understand. However, not having a callback hook on the `save_model` would be difficult. \r\n\r\nIf somebody is interested, a dirty workaround I did is,\r\n1. Register own `MLflowCallback`\r\n\r\n```\r\ntrainer = Trainer(\r\n ...\r\n callbacks=[MLflowCallback]\r\n)\r\ntrainer.remove_callback(transformers.integrations.MLflowCallback)\r\n```\r\n2. Add method in the class:\r\n```\r\n def log_artifact(self, output_dir):\r\n if self._initialized:\r\n logger.info(\"Logging artifacts. This may take time.\")\r\n self._ml_flow.log_artifacts(output_dir)\r\n```\r\n\r\n3. In the `run_ner.py` file, at the very end (or after ` trainer.save_model()`) added\r\n```\r\nml_flow_callback = trainer.pop_callback(MLflowCallback)\r\nml_flow_callback.log_artifact(training_args.output_dir)\r\n```\r\nWhich removes the `MLflowCallback` and tells to log the model.\r\n\r\nI know it is dirty, but if I come up with better solution I will open PR.\r\n\r\nThanks!", "> However, not having a callback hook on the save_model would be difficult.\r\n\r\nNot that this hook would be called when each checkpoint is saved, not just at the end of training. So you would not only save the last model.", "You are right, even with my hack of logging the saved model from the `output_dir`, transfers the checkpoint models as well, which is not what we need.\r\n\r\nI think modifying `MLflowCallback.on_train_end` with the code from `Trainer._save` should save only the model in temp directory and log it to mlflow. This way, we don't lose the current mlflow run and we dont save everything from the `output_dir`.\r\n\r\n```\r\n def on_train_end(self, args, state, control, model=None, tokenizer=None, **kwargs):\r\n if self._initialized and state.is_world_process_zero and self._log_artifacts:\r\n logger.info(\"Logging artifacts. This may take time.\")\r\n with tempfile.TemporaryDirectory() as temp_dir:\r\n\r\n if not isinstance(model, PreTrainedModel):\r\n if isinstance(unwrap_model(model), PreTrainedModel):\r\n state_dict = model.state_dict()\r\n unwrap_model(model).save_pretrained(temp_dir, state_dict=state_dict)\r\n else:\r\n logger.info(\"Trainer.model is not a `PreTrainedModel`, only saving its state dict.\")\r\n state_dict = model.state_dict()\r\n torch.save(state_dict, os.path.join(temp_dir, WEIGHTS_NAME))\r\n else:\r\n state_dict = model.state_dict()\r\n model.save_pretrained(temp_dir, state_dict=state_dict)\r\n if tokenizer is not None:\r\n tokenizer.save_pretrained(temp_dir)\r\n\r\n # Good practice: save your training arguments together with the trained model\r\n torch.save(args, os.path.join(temp_dir, \"training_args.bin\"))\r\n self._ml_flow.log_artifacts(temp_dir)\r\n```\r\n\r\nIf you think this is a good idea, maybe it can be added in the `MLflowCallback` integration.\r\nThanks!", "That sounds like the best compromise yes.", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored.", "Hello, any update on this problem? I am trying to log a model with mlflow but the artifacts aren't registered.\r\n\r\nCould you please help me with this ?\r\n\r\nBest Regards,", "> Hello, any update on this problem? I am trying to log a model with mlflow but the artifacts aren't registered.\r\n> \r\n> Could you please help me with this ?\r\n> \r\n> Best Regards,\r\n\r\nDid you export `HF_MLFLOW_LOG_ARTIFACTS` environment variable and set it to `True`?", "I was just trying with `HF_MLFLOW_LOG_ARTIFACTS` set and nothing was appearing in the mlflow artifacts" ]

[]

[]

[ "Succeed to reproduce with another model and dataset.\r\n\r\nhttps://gist.github.com/iamNCJ/a30afcbac392f6036bed65198ce5295e\r\n\r\n[gist](https://gist.github.com/iamNCJ/a30afcbac392f6036bed65198ce5295e)\r\n\r\nThis gist is derived from [an example provided by the pytorch lightening team](https://colab.research.google.com/github/PytorchLightning/pytorch-lightning/blob/master/notebooks/04-transformers-text-classification.ipynb), but it also causes this problem with multiple gpus.\r\n\r\nOutput:\r\n```text\r\nTraceback (most recent call last): File \"glue.py\", line 272, in <module>\r\n trainer.fit(model, dm)\r\n File \"/home/<user>/.conda/envs/RACE/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py\", line 498, in fit\r\n self.dispatch()\r\n File \"/home/<user>/.conda/envs/RACE/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py\", line 545, in dispatch\r\n self.accelerator.start_training(self)\r\n File \"/home/<user>/.conda/envs/RACE/lib/python3.8/site-packages/pytorch_lightning/accelerators/accelerator.py\", line 73, in start_training\r\n self.training_type_plugin.start_training(trainer)\r\n File \"/home/<user>/.conda/envs/RACE/lib/python3.8/site-packages/pytorch_lightning/plugins/training_type/ddp_spawn.py\", line 106, in start_training\r\n mp.spawn(self.new_process, **self.mp_spawn_kwargs)\r\n File \"/home/<user>/.conda/envs/RACE/lib/python3.8/site-packages/torch/multiprocessing/spawn.py\", line 230, in spawn\r\n return start_processes(fn, args, nprocs, join, daemon, start_method='spawn')\r\n File \"/home/<user>/.conda/envs/RACE/lib/python3.8/site-packages/torch/multiprocessing/spawn.py\", line 179, in start_processes\r\n process.start()\r\n File \"/home/<user>/.conda/envs/RACE/lib/python3.8/multiprocessing/process.py\", line 121, in start\r\n self._popen = self._Popen(self)\r\n File \"/home/<user>/.conda/envs/RACE/lib/python3.8/multiprocessing/context.py\", line 284, in _Popen\r\n return Popen(process_obj)\r\n File \"/home/<user>/.conda/envs/RACE/lib/python3.8/multiprocessing/popen_spawn_posix.py\", line 32, in __init__\r\n super().__init__(process_obj)\r\n File \"/home/<user>/.conda/envs/RACE/lib/python3.8/multiprocessing/popen_fork.py\", line 19, in __init__\r\n self._launch(process_obj)\r\n File \"/home/<user>/.conda/envs/RACE/lib/python3.8/multiprocessing/popen_spawn_posix.py\", line 47, in _launch\r\n reduction.dump(process_obj, fp)\r\n File \"/home/<user>/.conda/envs/RACE/lib/python3.8/multiprocessing/reduction.py\", line 60, in dump\r\n ForkingPickler(file, protocol).dump(obj)\r\nAttributeError: Can't pickle local object 'get_linear_schedule_with_warmup.<locals>.lr_lambda'\r\n```", "I succeed to start training after using DDP instead of DDP Spawn, since DDP Spawn forces the model to be pickleable but DDP doesn't, but I still wonder if it's possible to make the scheduler pickleable.", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored." ]

[]

[]

[]

[]

[]

[ "I am experiencing the same error. I have been all the day working around without solving it. I have tried trough Docker Cuda versions 9.2, 10.0 and 10.1 and different versions of pytorch including 1.3, 1.5 and 1.6. I have tried with different combinations of GTX1080 and RTX2080\r\n\r\nAdding \"--ddp_find_unused_parameters=true\" to the python command does not fix the error.\r\n\r\nAny help is really appreciated as I am working on the fine-tuning week @patrickvonplaten ", "I am experience this error too.\r\nCUDA 11.2\r\n4xT4 - 16Gb\r\n`--dataset_config_name=\"ru\"`", "@raja1196 I think I have found the bug. Could you try modifying in run_common_voice.py the gradient_checkpointing to False, as it is written below:\r\n\r\n```\r\ngradient_checkpointing: Optional[bool] = field(\r\n default=False,\r\n metadata={\r\n \"help\": \"If True, use gradient checkpointing to save memory at the expense of slower backward pass.\"\r\n },\r\n )\r\n```\r\n\r\nAnd then running the script without gradient_checkpointing as follows:\r\n\r\n`python -m torch.distributed.launch \\ --nproc_per_node 4 run_common_voice.py \\ --model_name_or_path=\"facebook/wav2vec2-large-xlsr-53\" \\ --dataset_config_name=\"tr\" \\ # use this argument to specify the language code --output_dir=./wav2vec2-large-xlsr-turkish-demo \\ --overwrite_output_dir \\ --num_train_epochs=\"5\" \\ --per_device_train_batch_size=\"16\" \\ --learning_rate=\"3e-4\" \\ --warmup_steps=\"500\" \\ --evaluation_strategy=\"steps\" \\ --save_steps=\"400\" \\ --eval_steps=\"400\" \\ --logging_steps=\"400\" \\ --save_total_limit=\"3\" \\ --freeze_feature_extractor \\ --feat_proj_dropout=\"0.0\" \\ --layerdrop=\"0.1\" \\ --fp16 \\ --group_by_length \\ --do_train --do_eval`\r\n\r\nThis solves the problem in my case and now I am able to run it with two GPUs. If it works to you, I will do PR", "@ivangtorre s solution works. unfortunately, I have to reduce the batchsize quite a lot.\r\n\r\nUpdate: I stopped using distributed training for now, as I did not get any performance gains somehow. Does anyone know whether the CTC loss of this model is computed in a distributed way, or are the outputs gathered on a single gpu before computing loss?", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored." ]

[]

[]

[ "Hi, thanks for opening an issue! Seen with @n1to, and this comes from the Unigram-based tokenizers: Unigram's tokenize cuts the string into tokens, and then converts them to IDs. Unknown tokens are detected during the token to IDs conversion, rather than when the string is cut into tokens.\r\n\r\nThis is different to BPE, where the string is cut in a lot of independant characters, converted to IDs, then merged to gether.\r\n\r\nThis is also different to WordPiece, where we start from the word and cut it until we find a token representation for each word piece; if we don't, then that's unknown.", "hi @LysandreJik . Thanks for looking into this, and sharing the info\r\n\r\nbased on your response it seems that for the `XLMRobertaTokenizer` tokenizer, we **cannot** guarantee that the following holds:\r\n\r\n```Python\r\nassert tokenizer.decode(tokenizer.convert_tokens_to_ids(tokenizer.tokenize(text))) == text\r\n```\r\n\r\nam i right ?", "I believe that's true for any tokenizer. If the tokenizer cannot tokenize one part of your text as it is not part of your vocabulary, then some information is lost.", "Hey guys,\r\n\r\nfor the sake of completeness, here's the double check with the reference implementation/tokenizer:\r\n\r\n```python\r\nimport torch\r\nxlmr = torch.hub.load('pytorch/fairseq', 'xlmr.base')\r\nxlmr.eval()\r\n\r\ntokens = xlmr.encode('请在黄鹂餐厅预订今晚7点半的位置。')\r\n```\r\n\r\nIt outputs:\r\n\r\n```bash\r\ntensor([ 0, 6, 9736, 213, 19390, 3, 113638, 209093, 155755,\r\n 966, 2391, 6193, 57486, 30, 2])\r\n```\r\n\r\n3 is the id for the unknown token, but you \"reverse\" tokenization with:\r\n\r\n```python\r\nxlmr.decode(tokens)\r\n```\r\n\r\nThis outputs:\r\n\r\n```bash\r\n'请在黄<unk>餐厅预订今晚7点半的位置。'\r\n```\r\n\r\nSo the `<unk>` token also appears :)", "@LysandreJik . agree that for any tokenizer, some information loss might happen, if the token is not part of the vocab.\r\n\r\nI guess, `SentencePiece` tokenizer is unique in a way : in the sense that \r\n\r\n- `SentencePieceProcessor provides a lossless data conversion that allows the original raw sentence to be perfectly reconstructed from the encoded data, i.e., Decode(Encode(input)) == input.`\r\n- where, Encode and Decode correspond to tokenization and de-tokenization respectively.\r\n- https://github.com/google/sentencepiece/blob/bc53923a9147dc8ffa54034c8ed774de78cc4d39/src/sentencepiece_processor.h#L118\r\n\r\nBecause of this, in the `tokenize` api for `XLMRobertaTokenizer`, there is no `<unk>` when the string is being cut into tokens \r\n\r\nBut, in the `encode` api when the tokens are converted to ids, `<unk>` are permitted as @stefan-it confirmed.\r\n\r\nhttps://github.com/google/sentencepiece/blob/9cf136582d9cce492ba5a0cfb775f9e777fe07ea/src/unigram_model.cc#L433 \r\n", "Thanks folks for the discussion and insight into the behaviour of tokenizers in HF. \r\n\r\nClosing this issue, since its not a bug per se.", "\r\nhi guys.\r\n\r\nI try to reproduce the code that is at the beginning of the topic and I get the following:\r\n\r\n" ]

[]

[]

[ "Oh actually, an additional comment: the title of the summarization notebook is currently \"Text classification on GLUE\" (and same for the translation)", "Fixed the titles and the sentence, so merging. Thanks for the review!" ]

[]

[]

[]

[]

[]

[ "Could you reproduce this in a colab so that we can take a look? Thanks!", "I reworked my code so I didn't get the error anymore I'm honestly not sure how it got fixed", "I think it maybe because I didn't restart the runtime\r\n", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored." ]

[]

[ { "login": "patrickvonplaten", "id": 23423619, "node_id": "MDQ6VXNlcjIzNDIzNjE5", "avatar_url": "https://avatars.githubusercontent.com/u/23423619?v=4", "gravatar_id": "", "url": "https://api.github.com/users/patrickvonplaten", "html_url": "https://github.com/patrickvonplaten", "followers_url": "https://api.github.com/users/patrickvonplaten/followers", "following_url": "https://api.github.com/users/patrickvonplaten/following{/other_user}", "gists_url": "https://api.github.com/users/patrickvonplaten/gists{/gist_id}", "starred_url": "https://api.github.com/users/patrickvonplaten/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/patrickvonplaten/subscriptions", "organizations_url": "https://api.github.com/users/patrickvonplaten/orgs", "repos_url": "https://api.github.com/users/patrickvonplaten/repos", "events_url": "https://api.github.com/users/patrickvonplaten/events{/privacy}", "received_events_url": "https://api.github.com/users/patrickvonplaten/received_events", "type": "User", "site_admin": false }, { "login": "patil-suraj", "id": 27137566, "node_id": "MDQ6VXNlcjI3MTM3NTY2", "avatar_url": "https://avatars.githubusercontent.com/u/27137566?v=4", "gravatar_id": "", "url": "https://api.github.com/users/patil-suraj", "html_url": "https://github.com/patil-suraj", "followers_url": "https://api.github.com/users/patil-suraj/followers", "following_url": "https://api.github.com/users/patil-suraj/following{/other_user}", "gists_url": "https://api.github.com/users/patil-suraj/gists{/gist_id}", "starred_url": "https://api.github.com/users/patil-suraj/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/patil-suraj/subscriptions", "organizations_url": "https://api.github.com/users/patil-suraj/orgs", "repos_url": "https://api.github.com/users/patil-suraj/repos", "events_url": "https://api.github.com/users/patil-suraj/events{/privacy}", "received_events_url": "https://api.github.com/users/patil-suraj/received_events", "type": "User", "site_admin": false } ]

[ "Wav2Vec2 Pre-Training is more important.", "I would also like to be able to pre-train a Wav2Vec2 model using my own raw audio files in a self-supervised way. It would be even better if I could use a pre-trained model as a starting point. Is there any way to do this currently?", "@czonios Yes, you can fine-tune a Wav2Vec2 model! Please check [this blogpost](https://huggingface.co/blog/fine-tune-xlsr-wav2vec2) by @patrickvonplaten. \r\n\r\nPre-training is not available as of now.", "Hey, \r\n\r\nWe should have Wav2Vec2 Pretraining added in ~2 weeks.", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored.", "Unstale", "#11306 is under way I think", "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored." ]

[]

[]

[ "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored." ]

[]

[]

[ "This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.\n\nPlease note that issues that do not follow the [contributing guidelines](https://github.com/huggingface/transformers/blob/master/CONTRIBUTING.md) are likely to be ignored." ]

[]

[]

[ "LGTM! I ran one training job successfully with these changes. ", "I tested it with `pytorch1.7.1`\r\n`564829616587.dkr.ecr.us-east-1.amazonaws.com/huggingface-pytorch-training:1.7.1-transformers4.4.0-py36-gpu-cu110-ubuntu18.04` \r\nand `pytorch1.6` \r\n`763104351884.dkr.ecr.us-east-1.amazonaws.com/huggingface-pytorch-training:1.6.0-transformers4.4.2-gpu-py36-cu110-ubuntu18.04`\r\n" ]