pip error error: unknown file type '.pyx' (from 'fairseq/data/data_utils_fast.pyx')

[luoqishuai]

❓ Questions and Help

What is your question?

When i pip fairseq in build dockerfile.Raise error. dockerfile FROM pytorch/pytorch:1.3-cuda10.1-cudnn7-devel RUN pip install redis fairseq setuptools flask I see fairseq need torch>=1.4. Is it a version problem? It seems that this installation does not work properly with the GPU.

Code

Step 10/13 : RUN pip install redis fairseq setuptools flask ---> Running in 44f0cf032e31 Looking in indexes: https://pypi.douban.com/simple/ Collecting redis Downloading https://pypi.doubanio.com/packages/a7/7c/24fb0511df653cf1a5d938d8f5d19802a88cef255706fdda242ff97e91b7/redis-3.5.3-py2.py3-none-any.whl (72kB) Collecting fairseq Downloading https://pypi.doubanio.com/packages/67/bf/de299e082e7af010d35162cb9a185dc6c17db71624590f2f379aeb2519ff/fairseq-0.9.0.tar.gz (306kB) Requirement already satisfied: setuptools in /opt/conda/lib/python3.6/site-packages (41.4.0) Collecting flask Downloading https://pypi.doubanio.com/packages/f2/28/2a03252dfb9ebf377f40fba6a7841b47083260bf8bd8e737b0c6952df83f/Flask-1.1.2-py2.py3-none-any.whl (94kB) Requirement already satisfied: cffi in /opt/conda/lib/python3.6/site-packages (from fairseq) (1.12.3) Collecting cython (from fairseq) Downloading https://pypi.doubanio.com/packages/19/49/91ebe4a00bf894d08dd9680cd9dfb05936eb2848eebd9402b43885aa74cf/Cython-0.29.21-cp36-cp36m-manylinux1_x86_64.whl (2.0MB) Requirement already satisfied: numpy in /opt/conda/lib/python3.6/site-packages (from fairseq) (1.17.2) Collecting regex (from fairseq) Downloading https://pypi.doubanio.com/packages/e4/90/0dae9bdebd8f8f8a39f1b80fdef240bec36ff64359f5fc584034ce4633cf/regex-2020.9.27-cp36-cp36m-manylinux2010_x86_64.whl (662kB) Collecting sacrebleu (from fairseq) Downloading https://pypi.doubanio.com/packages/a3/c4/8e948f601a4f9609e8b2b58f31966cb13cf17b940b82aa3e767f01c42c52/sacrebleu-1.4.14-py3-none-any.whl (64kB) Requirement already satisfied: torch in /opt/conda/lib/python3.6/site-packages (from fairseq) (1.3.0) Requirement already satisfied: tqdm in /opt/conda/lib/python3.6/site-packages (from fairseq) (4.32.1) Collecting click>=5.1 (from flask) Downloading https://pypi.doubanio.com/packages/d2/3d/fa76db83bf75c4f8d338c2fd15c8d33fdd7ad23a9b5e57eb6c5de26b430e/click-7.1.2-py2.py3-none-any.whl (82kB) Requirement already satisfied: Jinja2>=2.10.1 in /opt/conda/lib/python3.6/site-packages (from flask) (2.10.3) Collecting Werkzeug>=0.15 (from flask) Downloading https://pypi.doubanio.com/packages/cc/94/5f7079a0e00bd6863ef8f1da638721e9da21e5bacee597595b318f71d62e/Werkzeug-1.0.1-py2.py3-none-any.whl (298kB) Collecting itsdangerous>=0.24 (from flask) Downloading https://pypi.doubanio.com/packages/76/ae/44b03b253d6fade317f32c24d100b3b35c2239807046a4c953c7b89fa49e/itsdangerous-1.1.0-py2.py3-none-any.whl Requirement already satisfied: pycparser in /opt/conda/lib/python3.6/site-packages (from cffi->fairseq) (2.19) Collecting portalocker (from sacrebleu->fairseq) Downloading https://pypi.doubanio.com/packages/89/a6/3814b7107e0788040870e8825eebf214d72166adf656ba7d4bf14759a06a/portalocker-2.0.0-py2.py3-none-any.whl Requirement already satisfied: MarkupSafe>=0.23 in /opt/conda/lib/python3.6/site-packages (from Jinja2>=2.10.1->flask) (1.1.1) Building wheels for collected packages: fairseq Building wheel for fairseq (setup.py): started Building wheel for fairseq (setup.py): finished with status 'error' ERROR: Command errored out with exit status 1: command: /opt/conda/bin/python -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-install-vo_fqsrw/fairseq/setup.py'"'"'; file='"'"'/tmp/pip-install-vo_fqsrw/fairseq/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(file);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, file, '"'"'exec'"'"'))' bdist_wheel -d /tmp/pip-wheel-9qy3xz6y --python-tag cp36 cwd: /tmp/pip-install-vo_fqsrw/fairseq/ Complete output (276 lines): No CUDA runtime is found, using CUDA_HOME='/usr/local/cuda' running bdist_wheel running build running build_py creating build creating build/lib.linux-x86_64-3.6 creating build/lib.linux-x86_64-3.6/examples copying examples/init.py -> build/lib.linux-x86_64-3.6/examples creating build/lib.linux-x86_64-3.6/fairseq copying fairseq/sequence_scorer.py -> build/lib.linux-x86_64-3.6/fairseq copying fairseq/options.py -> build/lib.linux-x86_64-3.6/fairseq copying fairseq/tokenizer.py -> build/lib.linux-x86_64-3.6/fairseq copying fairseq/distributed_utils.py -> build/lib.linux-x86_64-3.6/fairseq copying fairseq/binarizer.py -> build/lib.linux-x86_64-3.6/fairseq copying fairseq/trainer.py -> build/lib.linux-x86_64-3.6/fairseq copying fairseq/init.py -> build/lib.linux-x86_64-3.6/fairseq copying fairseq/file_utils.py -> build/lib.linux-x86_64-3.6/fairseq copying fairseq/sequence_generator.py -> build/lib.linux-x86_64-3.6/fairseq copying fairseq/progress_bar.py -> build/lib.linux-x86_64-3.6/fairseq copying fairseq/legacy_distributed_data_parallel.py -> build/lib.linux-x86_64-3.6/fairseq copying fairseq/meters.py -> build/lib.linux-x86_64-3.6/fairseq copying fairseq/hub_utils.py -> build/lib.linux-x86_64-3.6/fairseq copying fairseq/utils.py -> build/lib.linux-x86_64-3.6/fairseq copying fairseq/search.py -> build/lib.linux-x86_64-3.6/fairseq copying fairseq/registry.py -> build/lib.linux-x86_64-3.6/fairseq copying fairseq/iterative_refinement_generator.py -> build/lib.linux-x86_64-3.6/fairseq copying fairseq/bleu.py -> build/lib.linux-x86_64-3.6/fairseq copying fairseq/checkpoint_utils.py -> build/lib.linux-x86_64-3.6/fairseq copying fairseq/pdb.py -> build/lib.linux-x86_64-3.6/fairseq creating build/lib.linux-x86_64-3.6/fairseq_cli copying fairseq_cli/interactive.py -> build/lib.linux-x86_64-3.6/fairseq_cli copying fairseq_cli/score.py -> build/lib.linux-x86_64-3.6/fairseq_cli copying fairseq_cli/init.py -> build/lib.linux-x86_64-3.6/fairseq_cli copying fairseq_cli/preprocess.py -> build/lib.linux-x86_64-3.6/fairseq_cli copying fairseq_cli/train.py -> build/lib.linux-x86_64-3.6/fairseq_cli copying fairseq_cli/setup.py -> build/lib.linux-x86_64-3.6/fairseq_cli copying fairseq_cli/generate.py -> build/lib.linux-x86_64-3.6/fairseq_cli copying fairseq_cli/eval_lm.py -> build/lib.linux-x86_64-3.6/fairseq_cli creating build/lib.linux-x86_64-3.6/examples/speech_recognition copying examples/speech_recognition/w2l_decoder.py -> build/lib.linux-x86_64-3.6/examples/speech_recognition copying examples/speech_recognition/infer.py -> build/lib.linux-x86_64-3.6/examples/speech_recognition copying examples/speech_recognition/init.py -> build/lib.linux-x86_64-3.6/examples/speech_recognition creating build/lib.linux-x86_64-3.6/examples/noisychannel copying examples/noisychannel/rerank_score_lm.py -> build/lib.linux-x86_64-3.6/examples/noisychannel copying examples/noisychannel/rerank_score_bw.py -> build/lib.linux-x86_64-3.6/examples/noisychannel copying examples/noisychannel/init.py -> build/lib.linux-x86_64-3.6/examples/noisychannel copying examples/noisychannel/rerank_generate.py -> build/lib.linux-x86_64-3.6/examples/noisychannel copying examples/noisychannel/rerank_tune.py -> build/lib.linux-x86_64-3.6/examples/noisychannel copying examples/noisychannel/rerank_options.py -> build/lib.linux-x86_64-3.6/examples/noisychannel copying examples/noisychannel/rerank.py -> build/lib.linux-x86_64-3.6/examples/noisychannel copying examples/noisychannel/rerank_utils.py -> build/lib.linux-x86_64-3.6/examples/noisychannel creating build/lib.linux-x86_64-3.6/examples/speech_recognition/criterions copying examples/speech_recognition/criterions/cross_entropy_acc.py -> build/lib.linux-x86_64-3.6/examples/speech_recognition/criterions copying examples/speech_recognition/criterions/init.py -> build/lib.linux-x86_64-3.6/examples/speech_recognition/criterions copying examples/speech_recognition/criterions/ASG_loss.py -> build/lib.linux-x86_64-3.6/examples/speech_recognition/criterions copying examples/speech_recognition/criterions/CTC_loss.py -> build/lib.linux-x86_64-3.6/examples/speech_recognition/criterions creating build/lib.linux-x86_64-3.6/examples/speech_recognition/models copying examples/speech_recognition/models/vggtransformer.py -> build/lib.linux-x86_64-3.6/examples/speech_recognition/models copying examples/speech_recognition/models/init.py -> build/lib.linux-x86_64-3.6/examples/speech_recognition/models copying examples/speech_recognition/models/w2l_conv_glu_enc.py -> build/lib.linux-x86_64-3.6/examples/speech_recognition/models creating build/lib.linux-x86_64-3.6/examples/speech_recognition/data copying examples/speech_recognition/data/asr_dataset.py -> build/lib.linux-x86_64-3.6/examples/speech_recognition/data copying examples/speech_recognition/data/data_utils.py -> build/lib.linux-x86_64-3.6/examples/speech_recognition/data copying examples/speech_recognition/data/init.py -> build/lib.linux-x86_64-3.6/examples/speech_recognition/data copying examples/speech_recognition/data/collaters.py -> build/lib.linux-x86_64-3.6/examples/speech_recognition/data copying examples/speech_recognition/data/replabels.py -> build/lib.linux-x86_64-3.6/examples/speech_recognition/data creating build/lib.linux-x86_64-3.6/examples/speech_recognition/tasks copying examples/speech_recognition/tasks/init.py -> build/lib.linux-x86_64-3.6/examples/speech_recognition/tasks copying examples/speech_recognition/tasks/speech_recognition.py -> build/lib.linux-x86_64-3.6/examples/speech_recognition/tasks creating build/lib.linux-x86_64-3.6/fairseq/criterions copying fairseq/criterions/sentence_ranking.py -> build/lib.linux-x86_64-3.6/fairseq/criterions copying fairseq/criterions/label_smoothed_cross_entropy.py -> build/lib.linux-x86_64-3.6/fairseq/criterions copying fairseq/criterions/adaptive_loss.py -> build/lib.linux-x86_64-3.6/fairseq/criterions copying fairseq/criterions/sentence_prediction.py -> build/lib.linux-x86_64-3.6/fairseq/criterions copying fairseq/criterions/nat_loss.py -> build/lib.linux-x86_64-3.6/fairseq/criterions copying fairseq/criterions/init.py -> build/lib.linux-x86_64-3.6/fairseq/criterions copying fairseq/criterions/masked_lm.py -> build/lib.linux-x86_64-3.6/fairseq/criterions copying fairseq/criterions/legacy_masked_lm.py -> build/lib.linux-x86_64-3.6/fairseq/criterions copying fairseq/criterions/cross_entropy.py -> build/lib.linux-x86_64-3.6/fairseq/criterions copying fairseq/criterions/label_smoothed_cross_entropy_with_alignment.py -> build/lib.linux-x86_64-3.6/fairseq/criterions copying fairseq/criterions/binary_cross_entropy.py -> build/lib.linux-x86_64-3.6/fairseq/criterions copying fairseq/criterions/composite_loss.py -> build/lib.linux-x86_64-3.6/fairseq/criterions copying fairseq/criterions/fairseq_criterion.py -> build/lib.linux-x86_64-3.6/fairseq/criterions creating build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/nonautoregressive_transformer.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/fconv_self_att.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/multilingual_transformer.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/insertion_transformer.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/transformer_from_pretrained_xlm.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/fairseq_incremental_decoder.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/nonautoregressive_ensembles.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/model_utils.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/transformer_lm.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/lightconv.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/init.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/masked_lm.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/wav2vec.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/fairseq_decoder.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/transformer.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/cmlm_transformer.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/composite_encoder.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/fconv.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/fairseq_model.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/lstm.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/fconv_lm.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/lightconv_lm.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/fairseq_encoder.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/levenshtein_transformer.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/distributed_fairseq_model.py -> build/lib.linux-x86_64-3.6/fairseq/models copying fairseq/models/iterative_nonautoregressive_transformer.py -> build/lib.linux-x86_64-3.6/fairseq/models creating build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/backtranslation_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/mask_tokens_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/plasma_utils.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/nested_dictionary_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/subsample_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/raw_label_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/prepend_token_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/monolingual_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/data_utils.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/concat_sentences_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/init.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/id_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/token_block_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/transform_eos_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/truncate_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/concat_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/transform_eos_lang_pair_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/append_token_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/colorize_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/num_samples_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/round_robin_zip_datasets.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/fairseq_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/pad_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/roll_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/sort_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/multi_corpus_sampled_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/list_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/resampling_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/denoising_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/dictionary.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/replace_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/offset_tokens_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/base_wrapper_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/numel_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/noising.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/strip_token_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/indexed_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/lm_context_window_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/sharded_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/prepend_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/language_pair_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/iterators.py -> build/lib.linux-x86_64-3.6/fairseq/data copying fairseq/data/lru_cache_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data creating build/lib.linux-x86_64-3.6/fairseq/optim copying fairseq/optim/sgd.py -> build/lib.linux-x86_64-3.6/fairseq/optim copying fairseq/optim/adamax.py -> build/lib.linux-x86_64-3.6/fairseq/optim copying fairseq/optim/bmuf.py -> build/lib.linux-x86_64-3.6/fairseq/optim copying fairseq/optim/adam.py -> build/lib.linux-x86_64-3.6/fairseq/optim copying fairseq/optim/init.py -> build/lib.linux-x86_64-3.6/fairseq/optim copying fairseq/optim/adagrad.py -> build/lib.linux-x86_64-3.6/fairseq/optim copying fairseq/optim/adadelta.py -> build/lib.linux-x86_64-3.6/fairseq/optim copying fairseq/optim/adafactor.py -> build/lib.linux-x86_64-3.6/fairseq/optim copying fairseq/optim/fairseq_optimizer.py -> build/lib.linux-x86_64-3.6/fairseq/optim copying fairseq/optim/nag.py -> build/lib.linux-x86_64-3.6/fairseq/optim copying fairseq/optim/fp16_optimizer.py -> build/lib.linux-x86_64-3.6/fairseq/optim creating build/lib.linux-x86_64-3.6/fairseq/tasks copying fairseq/tasks/sentence_ranking.py -> build/lib.linux-x86_64-3.6/fairseq/tasks copying fairseq/tasks/audio_pretraining.py -> build/lib.linux-x86_64-3.6/fairseq/tasks copying fairseq/tasks/sentence_prediction.py -> build/lib.linux-x86_64-3.6/fairseq/tasks copying fairseq/tasks/denoising.py -> build/lib.linux-x86_64-3.6/fairseq/tasks copying fairseq/tasks/translation_moe.py -> build/lib.linux-x86_64-3.6/fairseq/tasks copying fairseq/tasks/init.py -> build/lib.linux-x86_64-3.6/fairseq/tasks copying fairseq/tasks/masked_lm.py -> build/lib.linux-x86_64-3.6/fairseq/tasks copying fairseq/tasks/translation_lev.py -> build/lib.linux-x86_64-3.6/fairseq/tasks copying fairseq/tasks/legacy_masked_lm.py -> build/lib.linux-x86_64-3.6/fairseq/tasks copying fairseq/tasks/fairseq_task.py -> build/lib.linux-x86_64-3.6/fairseq/tasks copying fairseq/tasks/semisupervised_translation.py -> build/lib.linux-x86_64-3.6/fairseq/tasks copying fairseq/tasks/translation_from_pretrained_xlm.py -> build/lib.linux-x86_64-3.6/fairseq/tasks copying fairseq/tasks/cross_lingual_lm.py -> build/lib.linux-x86_64-3.6/fairseq/tasks copying fairseq/tasks/language_modeling.py -> build/lib.linux-x86_64-3.6/fairseq/tasks copying fairseq/tasks/multilingual_masked_lm.py -> build/lib.linux-x86_64-3.6/fairseq/tasks copying fairseq/tasks/multilingual_translation.py -> build/lib.linux-x86_64-3.6/fairseq/tasks copying fairseq/tasks/translation.py -> build/lib.linux-x86_64-3.6/fairseq/tasks creating build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/sinusoidal_positional_embedding.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/highway.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/positional_embedding.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/unfold.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/logsumexp_moe.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/beamable_mm.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/dynamic_convolution.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/init.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/layer_norm.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/learned_positional_embedding.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/vggblock.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/transformer_sentence_encoder_layer.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/linearized_convolution.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/sparse_transformer_sentence_encoder_layer.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/grad_multiply.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/conv_tbc.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/sparse_multihead_attention.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/character_token_embedder.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/scalar_bias.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/mean_pool_gating_network.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/adaptive_softmax.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/transformer_sentence_encoder.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/downsampled_multihead_attention.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/adaptive_input.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/gelu.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/multihead_attention.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/transformer_layer.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/sparse_transformer_sentence_encoder.py -> build/lib.linux-x86_64-3.6/fairseq/modules copying fairseq/modules/lightweight_convolution.py -> build/lib.linux-x86_64-3.6/fairseq/modules creating build/lib.linux-x86_64-3.6/fairseq/models/bart copying fairseq/models/bart/init.py -> build/lib.linux-x86_64-3.6/fairseq/models/bart copying fairseq/models/bart/hub_interface.py -> build/lib.linux-x86_64-3.6/fairseq/models/bart copying fairseq/models/bart/model.py -> build/lib.linux-x86_64-3.6/fairseq/models/bart creating build/lib.linux-x86_64-3.6/fairseq/models/roberta copying fairseq/models/roberta/init.py -> build/lib.linux-x86_64-3.6/fairseq/models/roberta copying fairseq/models/roberta/hub_interface.py -> build/lib.linux-x86_64-3.6/fairseq/models/roberta copying fairseq/models/roberta/alignment_utils.py -> build/lib.linux-x86_64-3.6/fairseq/models/roberta copying fairseq/models/roberta/model.py -> build/lib.linux-x86_64-3.6/fairseq/models/roberta creating build/lib.linux-x86_64-3.6/fairseq/data/audio copying fairseq/data/audio/init.py -> build/lib.linux-x86_64-3.6/fairseq/data/audio copying fairseq/data/audio/raw_audio_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data/audio creating build/lib.linux-x86_64-3.6/fairseq/data/encoders copying fairseq/data/encoders/subword_nmt_bpe.py -> build/lib.linux-x86_64-3.6/fairseq/data/encoders copying fairseq/data/encoders/hf_bert_bpe.py -> build/lib.linux-x86_64-3.6/fairseq/data/encoders copying fairseq/data/encoders/moses_tokenizer.py -> build/lib.linux-x86_64-3.6/fairseq/data/encoders copying fairseq/data/encoders/sentencepiece_bpe.py -> build/lib.linux-x86_64-3.6/fairseq/data/encoders copying fairseq/data/encoders/fastbpe.py -> build/lib.linux-x86_64-3.6/fairseq/data/encoders copying fairseq/data/encoders/gpt2_bpe.py -> build/lib.linux-x86_64-3.6/fairseq/data/encoders copying fairseq/data/encoders/init.py -> build/lib.linux-x86_64-3.6/fairseq/data/encoders copying fairseq/data/encoders/gpt2_bpe_utils.py -> build/lib.linux-x86_64-3.6/fairseq/data/encoders copying fairseq/data/encoders/utils.py -> build/lib.linux-x86_64-3.6/fairseq/data/encoders copying fairseq/data/encoders/space_tokenizer.py -> build/lib.linux-x86_64-3.6/fairseq/data/encoders copying fairseq/data/encoders/nltk_tokenizer.py -> build/lib.linux-x86_64-3.6/fairseq/data/encoders creating build/lib.linux-x86_64-3.6/fairseq/data/legacy copying fairseq/data/legacy/block_pair_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data/legacy copying fairseq/data/legacy/masked_lm_dictionary.py -> build/lib.linux-x86_64-3.6/fairseq/data/legacy copying fairseq/data/legacy/init.py -> build/lib.linux-x86_64-3.6/fairseq/data/legacy copying fairseq/data/legacy/masked_lm_dataset.py -> build/lib.linux-x86_64-3.6/fairseq/data/legacy creating build/lib.linux-x86_64-3.6/fairseq/optim/lr_scheduler copying fairseq/optim/lr_scheduler/triangular_lr_scheduler.py -> build/lib.linux-x86_64-3.6/fairseq/optim/lr_scheduler copying fairseq/optim/lr_scheduler/init.py -> build/lib.linux-x86_64-3.6/fairseq/optim/lr_scheduler copying fairseq/optim/lr_scheduler/reduce_lr_on_plateau.py -> build/lib.linux-x86_64-3.6/fairseq/optim/lr_scheduler copying fairseq/optim/lr_scheduler/polynomial_decay_schedule.py -> build/lib.linux-x86_64-3.6/fairseq/optim/lr_scheduler copying fairseq/optim/lr_scheduler/inverse_square_root_schedule.py -> build/lib.linux-x86_64-3.6/fairseq/optim/lr_scheduler copying fairseq/optim/lr_scheduler/tri_stage_lr_scheduler.py -> build/lib.linux-x86_64-3.6/fairseq/optim/lr_scheduler copying fairseq/optim/lr_scheduler/fixed_schedule.py -> build/lib.linux-x86_64-3.6/fairseq/optim/lr_scheduler copying fairseq/optim/lr_scheduler/cosine_lr_scheduler.py -> build/lib.linux-x86_64-3.6/fairseq/optim/lr_scheduler copying fairseq/optim/lr_scheduler/fairseq_lr_scheduler.py -> build/lib.linux-x86_64-3.6/fairseq/optim/lr_scheduler creating build/lib.linux-x86_64-3.6/fairseq/modules/dynamicconv_layer copying fairseq/modules/dynamicconv_layer/dynamicconv_layer.py -> build/lib.linux-x86_64-3.6/fairseq/modules/dynamicconv_layer copying fairseq/modules/dynamicconv_layer/init.py -> build/lib.linux-x86_64-3.6/fairseq/modules/dynamicconv_layer copying fairseq/modules/dynamicconv_layer/cuda_function_gen.py -> build/lib.linux-x86_64-3.6/fairseq/modules/dynamicconv_layer copying fairseq/modules/dynamicconv_layer/setup.py -> build/lib.linux-x86_64-3.6/fairseq/modules/dynamicconv_layer creating build/lib.linux-x86_64-3.6/fairseq/modules/lightconv_layer copying fairseq/modules/lightconv_layer/lightconv_layer.py -> build/lib.linux-x86_64-3.6/fairseq/modules/lightconv_layer copying fairseq/modules/lightconv_layer/init.py -> build/lib.linux-x86_64-3.6/fairseq/modules/lightconv_layer copying fairseq/modules/lightconv_layer/cuda_function_gen.py -> build/lib.linux-x86_64-3.6/fairseq/modules/lightconv_layer copying fairseq/modules/lightconv_layer/setup.py -> build/lib.linux-x86_64-3.6/fairseq/modules/lightconv_layer running build_ext building 'fairseq.libbleu' extension creating build/temp.linux-x86_64-3.6 creating build/temp.linux-x86_64-3.6/fairseq creating build/temp.linux-x86_64-3.6/fairseq/clib creating build/temp.linux-x86_64-3.6/fairseq/clib/libbleu gcc -pthread -B /opt/conda/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -I/opt/conda/include/python3.6m -c fairseq/clib/libbleu/libbleu.cpp -o build/temp.linux-x86_64-3.6/fairseq/clib/libbleu/libbleu.o -std=c++11 -O3 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=libbleu -D_GLIBCXX_USE_CXX11_ABI=0 cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++ gcc -pthread -B /opt/conda/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -I/opt/conda/include/python3.6m -c fairseq/clib/libbleu/module.cpp -o build/temp.linux-x86_64-3.6/fairseq/clib/libbleu/module.o -std=c++11 -O3 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=libbleu -D_GLIBCXX_USE_CXX11_ABI=0 cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++ g++ -pthread -shared -B /opt/conda/compiler_compat -L/opt/conda/lib -Wl,-rpath=/opt/conda/lib -Wl,--no-as-needed -Wl,--sysroot=/ build/temp.linux-x86_64-3.6/fairseq/clib/libbleu/libbleu.o build/temp.linux-x86_64-3.6/fairseq/clib/libbleu/module.o -o build/lib.linux-x86_64-3.6/fairseq/libbleu.cpython-36m-x86_64-linux-gnu.so building 'fairseq.data.data_utils_fast' extension error: unknown file type '.pyx' (from 'fairseq/data/data_utils_fast.pyx')

ERROR: Failed building wheel for fairseq Running setup.py clean for fairseq

import torch print(torch.rand(3,3).cuda()) tensor([[0.3323, 0.3540, 0.5931], [0.5691, 0.1683, 0.9104], [0.9937, 0.0520, 0.3832]], device='cuda:0') torch.cuda.is_available() True torch.cuda.device_count() 1

but model use cpu , not use gpu and cuda

model=TransformerModel.from_pretrained( '/data/en-ckpt', checkpoint_file='checkpoint_last.pt', data_name_or_path='/data/data-bin/zh-en' ) model.translate('english')

output:

| [en] dictionary: 500888 types | [zh] dictionary: 10736 types Namespace(activation_dropout=0.0, activation_fn='relu', adam_betas='(0.9, 0.98)', adam_eps=1e-08, adaptive_input=False, adaptive_softmax_cutoff=None, adaptive_softmax_dropout=0, arch='transformer', attention_dropout=0.0, best_checkpoint_metric='loss', bpe=None, bucket_cap_mb=25, clip_norm=0.0, cpu=False, criterion='label_smoothed_cross_entropy', cross_self_attention=False, curriculum=0, data='/data/data-bin/zh-en', dataset_impl=None, ddp_backend='c10d', decoder_attention_heads=8, decoder_embed_dim=512, decoder_embed_path=None, decoder_ffn_embed_dim=2048, decoder_input_dim=512, decoder_layerdrop=0, decoder_layers=6, decoder_layers_to_keep=None, decoder_learned_pos=False, decoder_normalize_before=False, decoder_output_dim=512, device_id=0, disable_validation=False, distributed_backend='nccl', distributed_init_method=None, distributed_no_spawn=False, distributed_port=-1, distributed_rank=0, distributed_world_size=1, dropout=0.3, encoder_attention_heads=8, encoder_embed_dim=512, encoder_embed_path=None, encoder_ffn_embed_dim=2048, encoder_layerdrop=0, encoder_layers=6, encoder_layers_to_keep=None, encoder_learned_pos=False, encoder_normalize_before=False, find_unused_parameters=False, fix_batches_to_gpus=False, fp16=True, fp16_init_scale=128, fp16_scale_tolerance=0.0, fp16_scale_window=None, keep_interval_updates=-1, keep_last_epochs=-1, label_smoothing=0.1, layer_wise_attention=False, layernorm_embedding=False, lazy_load=False, left_pad_source=False, left_pad_target=False, log_format='tqdm', log_interval=1000, lr=[0.0005], lr_scheduler='inverse_sqrt', max_epoch=0, max_sentences=None, max_sentences_valid=None, max_source_positions=1024, max_target_positions=1024, max_tokens=4096, max_tokens_valid=4096, max_update=0, maximize_best_checkpoint_metric=False, memory_efficient_fp16=False, min_loss_scale=0.0001, min_lr=-1, no_cross_attention=False, no_epoch_checkpoints=False, no_last_checkpoints=False, no_progress_bar=False, no_save=False, no_save_optimizer_state=False, no_scale_embedding=False, no_token_positional_embeddings=False, num_workers=1, optimizer='adam', optimizer_overrides='{}', raw_text=False, required_batch_size_multiple=8, reset_dataloader=False, reset_lr_scheduler=False, reset_meters=False, reset_optimizer=False, restore_file='checkpoint_last.pt', save_dir='checkpoints', save_interval=1, save_interval_updates=0, seed=1, sentence_avg=False, share_all_embeddings=False, share_decoder_input_output_embed=True, skip_invalid_size_inputs_valid_test=False, source_lang='en', target_lang='zh', task='translation', tbmf_wrapper=False, tensorboard_logdir='', threshold_loss_scale=None, tokenizer=None, train_subset='train', truncate_source=False, update_freq=[1], upsample_primary=1, use_bmuf=False, user_dir=None, valid_subset='valid', validate_interval=1, warmup_init_lr=0, warmup_updates=4000, weight_decay=0.0001) '英 语'

What have you tried?

What's your environment?

• fairseq Version (e.g., 1.0 or master): 0.9.0
• PyTorch Version (e.g., 1.0) 1.3
• OS (e.g., Linux): ubuntu 16.04(dockerfile) ubuntu18.04(computer)
• How you installed fairseq (pip, source):pip
• Build command you used (if compiling from source):
• Python version:3.6.9
• CUDA/cuDNN version:10.1 7

[luoqishuai]

model=model.cuda()

[dendisuhubdy]

had this issue too

[haonan-li]

the same issue

[georgeliu233]

Try sudo pip install --editable ./ ?

[luoqishuai]

sry @dendisuhubdy @haonan-li I solved the problem, but I didn't write out the solution. Your answer, Github didn't send me an email reminding me. It wasn't until Georgeliu tipped me off I ran this command after the normal installation of Fairseq, I do not know whether it is helpful to you

git clone https://github.com/pytorch/fairseq && cd fairseq && pip install --editable .

[alexbprofit]