Project dependencies may have API risk issues
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 5k
- Forks
- 499
- PR merge metrics
- No merged PRs in 30d
Description
Hi, In muzic, inappropriate dependency versioning constraints can cause risks.
Below are the dependencies and version constraints that the project is using
scikit-learn==0.21.3
miditoolkit==0.1.14
fastBPE
dtw-python
nltk
keras
tqdm
matplotlib
pretty_midi
PyYAML
pypianoroll
fairseq==0.10.0
torch==1.7.1
transformers==3.5.1
pypinyin==0.39.1
jieba==0.42.1
tensorboard==2.4.0
scipy==1.3.1
librosa
pyworld
soundfile
pyOpenSSL
secrets
pkuseg==0.0.25
icecream==2.1.3
textrank4zh==0.3
The version constraint == will introduce the risk of dependency conflicts because the scope of dependencies is too strict.
The version constraint No Upper Bound and * will introduce the risk of the missing API Error because the latest version of the dependencies may remove some APIs.
After further analysis, in this project,
The version constraint of dependency scikit-learn can be changed to >=0.15.0,<=0.20.4.
The version constraint of dependency nltk can be changed to >=3.2.2,<=3.7.
The version constraint of dependency tqdm can be changed to >=4.42.0,<=4.64.0.
The version constraint of dependency transformers can be changed to >=3.5.0,<=3.5.1.
The version constraint of dependency pypinyin can be changed to >=0.4.0,<=0.46.0.
The version constraint of dependency scipy can be changed to >=0.12.0,<=1.7.3.
The version constraint of dependency librosa can be changed to >=0.2.0,<=0.3.1.
The version constraint of dependency soundfile can be changed to >=0.8.1,<=0.9.0.post1.
The above modification suggestions can reduce the dependency conflicts as much as possible,
and introduce the latest version as much as possible without calling Error in the projects.
The invocation of the current project includes all the following methods.
The calling methods from the scikit-learn
sklearn.metrics.pairwise.cosine_similarity joblib.load random.choice
The calling methods from the nltk
collections.OrderedDict collections.OrderedDict.keys
The calling methods from the tqdm
tqdm.trange tqdm.tqdm itertools.product
The calling methods from the transformers
transformers.modeling_outputs.BaseModelOutputWithPastAndCrossAttentions transformers.AutoTokenizer.from_pretrained transformers.modeling_gpt2.Block transformers.utils.logging.get_logger transformers.modeling_outputs.CausalLMOutputWithPastAndCrossAttentions
The calling methods from the pypinyin
pypinyin.lazy_pinyin
The calling methods from the scipy
scipy.ndimage.filters.gaussian_filter1d scipy.signal.convolve2d scipy.stats.mode scipy.linalg.block_diag
The calling methods from the librosa
librosa.load
The calling methods from the soundfile
soundfile.write
The calling methods from the all methods
select_notes.append self.apply_sparse_mask remove_prefix local_maxima_rows matrics.append numpy.triu f.read.strip strct.append src_tokens.eq.long.sum letter_pos.append base_tones.items textrank4zh.TextRank4Keyword.analyze step.attn.copy_ MaskedLMDictionary.finalize query.size fairseq.search.DiverseBeamSearch en_word_utils.get_aux_mask min.append BasicTokenizer index.self._sizes.item csv.writer.writerows i.flatten_hypo_samples.np.array.reshape fairseq.models.roberta.RobertaModel.from_pretrained.predict save_demo targets.sum.float score_config.Song self.vocab.dummy_sentence.index_select _get_mass_dataset_key join.strip logging.getLogger.debug optimizer.backward sub_tokens.append lens.max.samples.len.torch.zeros.long torch.sign scipy.signal.convolve2d tick2second loss.mean.mean get_hash collections.defaultdict.keys logits.torch.sigmoid.detach.cpu.numpy key_padding_mask.size.key_padding_mask.size.torch.zeros.type_as OctupleEncoder i.rhyme_words_list.append mask_prob.self.vocab.len.max_instruments.max_bars.generate_mask.reshape beam_size.bsz.torch.arange.view.repeat.view random.sample warnings.warn self.src_dict.pad lyrics.append melody_model.predict device.torch.long.next_pos.poser.convert_tokens_to_ids.torch.tensor.unsqueeze torch.max lrc_len.mld_len.np.zeros.astype src_tokens.size constraintor.TokenConstraintor loss.isnan.any attn_prob.view timer.stop self.replicate_first_beam.cpu notes.append hashlib.md5 self.wre round compute_alignment_weights.float saved_hparams.update create_custom_forward separate avg_attn.div_ self.mask_start n.split i.next_token_ids.unsqueeze fairseq.quantization_utils.quantize_model_scalar.prepare_for_tpu_ numpy.std songmass_cons.get_rhythm_score word_utils.get_in_word_pos align_weights.align.align.attn.log.sum.isinf next fairseq.data.LanguagePairDataset cur_sent.append glob.glob fairseq.search.LengthConstrainedBeamSearch smooth_loss.sum.sum _is_whitespace self.encoder self.src_dict.bos notes_file.readlines get_syllables self.wpe st.readlines numpy.count_nonzero ori_beat.copy pandas.read_csv j.dataset.numpy note.append tmp_string.append bar.strip.split self.mask_interval item.strip.strip timer.start kwargs.items tokens.beam_size.bsz.torch.arange.unsqueeze.type_as.to.resize_ get_structure_score apex.amp.master_params s.readlines.split notes.copy m.readlines.append merged_info.update numpy.arange id.index_select.index_select self.vocab.dummy_sentence SongWrapper.Song.get_noteidx_by_time fairseq.models.roberta.RobertaModel.from_pretrained _key_chord_transition_distribution main multiprocessing.pool.Pool.close new_instrs.append logging_output.get score_config.Song.get_corres_notes src_key.self.encoders.index_select instr_name.instru_order.track_lists_to_merge.append phonemizer.phonemize numpy.mean m.max_decoder_positions alignment.max pkuseg.pkuseg line.strip.split.split get_notes self.tgt.prefetch t.s.torch.eq.float poser.convert_ids_to_tokens self.quant_noise attns.size fairseq.data.indexed_dataset.dataset_exists song_id_test.extend second2tick multiprocessing.Lock torch.ne.unsqueeze self.head_dim.self.num_heads.bsz.k.contiguous.view.transpose librosa.core.load torch.nn.functional.relu beats.split.split numpy.sort e2v enumerate generate Encoder combine_with_sep zipfile.ZipFile.open encoding.sort i.bar_to_timesig.append f.writelines beam_size.reorder_state.view.add_ torch.nn.ModuleDict x.isdigit d.append lyrics.translate.replace self.update_finished dec_vel fairseq.search.BeamSearch s.numel reorder_bar self.model.forward_encoder get_pitch_duration_structure src_tokens.eq s.readlines sys.argv.split ord attn_weights_float.masked_fill.type_as torch.cat.index_select self.lm_model.get_normalized_probs self._clean_text group_by_bar file_line.strip.split midi_utils.enc_ts write_song self.dropout_module zipfile.ZipFile beam_size.i.beam_size.i.finalized.transpose target_word_ids.index_select.index_select self.src_vocab.pad index_to_remove.append NOTES.index position_ids.unsqueeze.view numpy.log labels.contiguous.view denoise chord_truc ret_list.append f.readlines lyrics.copy attention_mask.long.cumsum rest.append torch.cat.tolist miditoolkit.containers.Instrument kwargs.pop.size pretty_midi.PrettyMIDI.write target.detach.cpu bsz.self.src_lang_id.torch.LongTensor.expand.to buildNext.append tmp_mld.append melody.append get_tag feature_vals.append midi_obj.time_signature_changes.append adapt_e.sort copy.deepcopy prev_output_tokens.eq get_keyword_mask numpy.tile.argmax s.split sort_coo mono_lang_pairs.append labels.contiguous seq.extend get_merged_midi_info manager.list.append int.strip n_ctx.generated_finals.unsqueeze fairseq.data.StripTokenDataset chords.append kwargs.get.view subfinals.extend BeamSearchNode file_name.split.split set generate_midi tran.close tm.write torch.cat song_prefix.split new_hparam.split json.load innerpos.strip meloform.meloform_transformer.TransformerMeloFormModel.from_pretrained self.collater source.append self.tgt_dict.index attn_loss_layer.view.view fairseq.data.data_utils.collate_tokens source_sent_ids.index_select.append songmass_cons.get_pitch_score self.head_dim.self.num_heads.bsz.tgt_len.q.contiguous.view.transpose df.set_index.to_csv SongWrapper.Song.get_tone get_pitch_duration_sequence lprobs.view.view XTransformerDecoder device.torch.long.context_finals.torch.tensor.unsqueeze utils.enc_vel item.strip.split beam_size.step.prefix_tokens.unsqueeze.repeat.view.eq df.set_index.set_index i.tokens_clone.eq.long.sum x.rstrip.split functools.lru_cache unicodedata.normalize.split sorted_indices_to_remove.clone midi_tuple.str.encode collections.defaultdict.items sents_seen.keys modes_results.append FileNotFoundError self.head_dim.self.num_heads.bsz.v.contiguous.view.transpose lyric_list.append lines.split.count bbsz_idx.gen_ngrams.get random.shuffle lang_pair.split word_utils.clean.replace newly_finished.append logits.torch.sigmoid.detach.cpu pitch tqdm.auto.tqdm fairseq.data.Dictionary targets.preds.sum.torch.sign.sum write_align_part self._get_num midi_adjust.dump_midi pyworld.cheaptrick align_idxs.append notes.sort src_key.self.encoders.keys kwargs.keys net_output.sample.model.get_targets.view get_transition_score align_weights.align.align.attn.log.sum self.mask_interval_align torch.cat.append transformers.AdamW.zero_grad numpy.argmax.append midiutil.MidiFile.MIDIFile.writeFile all_melody.extend midiutil.MidiFile.MIDIFile pandas.isnull h2.h1.seen.append _select_results x.transpose.view sliding_lyric_train.extend self.args.eval_lang_pair.split zip infer.stage_1_get_l2b word_utils.clean_beat assign_chords_for_sequence_by_template.extend fairseq.models.build_model.train cur_sep.extend src_str.split.split label_dict.pad lprobs.view.sum target_sent_ids.index_select.index_select transformers.EncoderDecoderModel dec_ts model.div_ dict.get secrets.choice extract_topn_from_vector self.pad.src_tokens.ne.self.eos.src_tokens.ne.long.sum output_mappings.append utils.get_sentence_pinyin_finals self.search.step trend2notes.translate.split token_type_ids.view.view sample.values min.split counter.st_idx.sample.split n_folds.StratifiedKFold.split get_fold space.space.str.maketrans.sents.translate.split label_fn pitch_intervals sentencer.convert_ids_to_tokens self._tokenize_chinese_chars preprocess.b2e full_final.extend device.torch.long.next_sentence.sentencer.convert_tokens_to_ids.torch.tensor.unsqueeze self.basic_tokenizer.tokenize maximum_likelihood_scale attn_weights_float.masked_fill.view override_config sents.translate nll_loss.sum.sum self.encoder.get batch_token.torch.Tensor.long get_tracks_to_merge file_name.split loss.detach tokenizer.convert_tokens_to_ids tensor.view.size module.GPT2LMHeadModel.from_pretrained.eval numpy.ones beater.convert_tokens_to_ids range get_sentence build_embedding random.randint x.rstrip dtw.accelerated_dtw infer.stage_2_get_b2t text.self.bpe.split ss.join.split args.root_path.rstrip.split.read_fn x.join.lower get_sample math.log os.environ.split beam_size.step.prefix_tokens.unsqueeze.repeat.view.ne get_intervals.append parse_lyric _chord_frame_log_likelihood line.strip chord_rs.append line.strip.index MaskedLMDictionary.load self._collate en_word_utils.get_keyword_mask block compute_alignment_weights math.inf.torch.tensor.to samples.s.pad_idx.s.ne.long.sum.torch.LongTensor.index_select.sum lyric_reverse.strip torch.nn.ModuleList segments.append mass_mono_datasets.keys numpy.argsort apex.amp.initialize get_tone_id utils.chord_recog.infer_chords_for_sequence merge_sentId target.detach.cpu.numpy midi_utils.encoding_to_midi.dump song.get_pos.append parser.add_argument convert_to_midi_obj model.decoder.forward spacy.load song_id.songid_to_split.append midi_utils.enc_tpo self._set_input_buffer self._run_split_on_punc en_word_utils.get_sents_form_mask self.wse whitespace_tokenize sample_fn instrument.name.rsplit.lower full_sentencer.tokenize collate open dp get_interval torch.nn.utils.clip_grad_norm_ f.write module tmp.append masked_mh_attention.MaskedMHAttention fix utils.encoding_to_midi.dump torch.sort lprobs.lprobs.size.torch.zeros.to.fill_ torch.cat.transpose kwargs.pop align.strip utils.lyrics_match.Lyrics_match sents.replace.replace tokenizer.convert_ids_to_tokens tgt_dict.index tmp_2.index r.get get_song_ids eval_lang_pair.split sen_idx.select_chords.split torch.nn.functional.one_hot XTransformerDecoder.max_positions tgt_strs.append miditoolkit.Lyric net_output.float prev_output_tokens.eq.int sys.exit file_name.lower sys.argv.replace multiprocessing.Lock.acquire self.decoder kwargs.get self.convert.tolist is_sorted self.v_proj beam_size.bsz.torch.arange.unsqueeze.type_as pd_dd.get_pd_dd self.get_all_pos self.model.eval shutil.rmtree torch.add self.strct_buffer.update_pitch get_beat_token self.pitch_ruler._check_duplicate_notes random.choice src_dict.bos get_keyword self.datasets.values.iter.next.datasets.keys batch_idxs.bsz.attn.view.view.view get_pitch.append multiprocessing.pool.Pool.apply_async pad_length self.beam_size.src_lengths.expand.contiguous.view.expand Encoder_SP tuple set.add torch.nn.functional.log_softmax samples_beat.append fairseq.data.data_utils.collate_tokens.to fairseq.models.transformer_lm.TransformerLanguageModel.from_pretrained.cuda.score beam_size.bsz.torch.zeros.to pitch_shift full_sentencer.convert_tokens_to_ids aux_debug.append prev_output_tokens.index_select.index_select syllables.strip.strip sklearn.metrics.pairwise.cosine_similarity r.get.append get_pitch_class_histogram.sum torch.empty.to tgt_dict.bos _sents.append x.strip.split seq.append beats.append self.mask_pitch_duration enc_dur math.ceil score_config.Song.get_noteidx_by_time torch.cat.size agg.get_smoothed_values self.random_pitch_duration transformers.utils.logging.get_logger get_textrank fairseq.models.transformer.TransformerModel.from_pretrained repeat_pair.items numpy.linalg.norm batch_beat.torch.Tensor.long attn_mask.repeat.size str.maketrans torch.nn.LayerNorm torch.set_printoptions numpy.min type _chord_pitch_vectors fairseq.utils.softmax query_list.append src_lengths.sum.item os.path.join.split midi_fn.split.split instru2track.values logits.torch.sigmoid.detach word_utils.get_sents_form int.append en_word_utils.get_in_word_pos self.out_proj.transpose full_finalizer.convert_tokens_to_ids _rescale_rhymes tokenizations.bpe_tokenizer.get_encoder lprobs.size TimeoutError multiprocessing.pool.Pool.join phrase_range.append module.GPT2LMHeadModel.from_pretrained.to fairseq.data.TruncateDataset target.sum.long s.strip get_pitch_score flat_note_seq.copy.extend self.tone_transform args.mt_steps.split v.contiguous.view fairseq.data.Dictionary.finalize sentence.join.split multiprocessing.Manager cut_lyric_valid.extend align_song tgt_dict.eos scipy.signal.argrelextrema kwargs.get.to self.beam_size.src_lengths.expand.contiguous.view.max token.join s.isdigit.append sum_over_languages midi_preprocess.steps.filter_and_merge.filter_and_merge int dec_dur output.Counter.most_common gen_midi tokens.join.replace.strip self.convert prev_output_tokens.size re.sub process_with_catch torch.zeros midi_obj.instruments.notes.sort fairseq.quantization_utils.quantize_model_scalar.eval extract_src notes.extend miditoolkit.midi.containers.Marker midi_utils.number_to_note s.write step.no_repeat_ngram_size.step.bbsz_idx.tokens.tolist os.path.dirname args.langs.dicts.mask self.sp.EncodeAsIds whole_lyric.extend super.__init__ chord_items.append cand_scores.ne self.tgt_key.self.decoders.get_normalized_probs unfin_idx.item transformers.modeling_outputs.CausalLMOutputWithPastAndCrossAttentions get_duration_count.astype song.strip.split fairseq.data.ConcatDataset parallels.append methods.split muli_task loss.mean.item e2t subprocess.check_call tmp_melody.append tabulate.tabulate utils.midi_io.Item reorder_state.view torch.min parser.parse_args.__repr__ numpy.array.tolist w.writelines self.reduce_metrics torch.sigmoid.detach full_poser.convert_tokens_to_ids shift_lyrics_time print self._get_input_buffer music_mt_dataset.MusicMtDataset lprobs.size.torch.zeros.to polish magenta_chord_recognition._key_chord_transition_distribution keep_long_notes ret_lyric.append split_exists lprobs.view.gather item.vel.np.clip.item os.walk chord.strip.split token.split self.model.reorder_encoder_out filter re.match ts.append module.GPT2Config.from_json_file model subsentences.extend src_str.strip.split sample.float midi_obj_1.instruments.append midi.lyrics.append x.split.split self.k_proj logging.basicConfig self.project_in_dim self.__setattr__ self._prepare_batch_for_alignment magenta_chord_recognition._key_chord_distribution preprocess.data_zip.namelist thulac.thulac torch.cuda.is_available fairseq.data.Dictionary.add_file_to_dictionary i.to.index_select melody_list.append cls.prepare os.path.splitext string.strip.split load_midi.dump pyworld.dio preprocess.t2e words.append self.pos_transform track_info.items transformers.BertConfig.from_pretrained self.search.update_constraints fairseq.quantization_utils.quantize_model_scalar KeyError self.search.set_src_lengths aux.append tempo_changes.append get_pitch dc multiprocessing.Manager.list bsz.torch.arange.view source_word_ids.index_select.append merge merge_track_ops.merge_strings self.finalize_hypos dataset.filter_indices_by_size torch.unique preprocess_chord_16.b2e detect_lines_helper.append midi_preprocess.utils.hparams.set_hparams sen_idx.select_chords.split.split bar.split relative_duration get_lyrics midi.markers.append fairseq.data.dictionary.Dictionary.add_file_to_dictionary segmentation chord.split self.lm_head.contiguous numpy.where any j.dataset.size full_final.append miditoolkit.containers.TempoChange join.append fairseq.data.RoundRobinZipDatasets pandas.DataFrame f_label.write self.tgt_lang_id.torch.LongTensor.expand os.system torch.no_grad _init_pinyin_dict score align_m.append tempo.np.mean.item self.bpe_ranks.get targets.sum.detach load_template data_path.midi_fn.lstrip.replace i.lines.strip.split heads_to_prune.items _keyword.get_keyword attn_weights.mean.masked_fill SongWrapper.Song.get_last_note utils.encoding_to_midi i.to full_line.extend tmp.custom_lm.sample.split self.q_proj sentence.join.split.strip self.crnt_unit.pitch.append normalize_to_c_major self.beam_size.src_tokens.expand.contiguous layer strct.split torch.eq tones_to_scales midi_fn.os.path.basename.replace attn.contiguous.view functools.reduce get_similarity midi_obj.tempo_changes.append SequenceTooLongError os.path.normpath fl.write self.get_note self.lm_model batch_idxs.bsz.tokens.view.view.index_select fairseq.data.encoders.build_bpe preprocess.d2e preprocess_chord_16.encoding_to_MIDI samples_pos.append torch.nn.functional.binary_cross_entropy_with_logits d2e midi_fn.os.path.basename.replace.replace convert pandas.isnull.iterrows utils.midi_io.midi2items beats.split.insert source_sent_ids.unsqueeze item.strip self.out_proj.contiguous fairseq.data.encoders.build_tokenizer scipy.linalg.block_diag preprocess.v2e numpy.tile self._no_repeat_ngram.view tokenizations.tokenization_bert.BertTokenizer i.tokens_clone.eq Item songmass_cons.Song temp_features.drop.drop enc_tpo _init_pinyin_dict.items time_signature_reduce wc.write generator.generate ns_id.unsqueeze wv.write pypianoroll.Multitrack sys.path.append period_range.append bsz.self.tgt_lang_id.torch.LongTensor.expand.to encoder_out._replace._replace self.wordpiece_tokenizer.tokenize.append output.detach.cpu.numpy preprocess.encoding_to_str get_tempo_class check_not output.append miditoolkit.midi.parser.MidiFile.get_tick_to_time_mapping md.get_md fairseq.data.PrependTokenDataset key_padding_mask.unsqueeze.to pkuseg.pkuseg.cut src_tokens.max_len.beam_size.bsz.torch.zeros.to.float line.replace.split fairseq.utils.item self.BertTokenizer.super.__init__ pitch.append transformers.AdamW.step get_aux_score src_len.tgt_len.self.num_heads.bsz.attn_weights_float.view.transpose torch.is_tensor sample.get.size remove_empty_track log_probs.append torch.nn.DataParallel sklearn.model_selection.StratifiedKFold self.segment_embeddings sublines.extend self.tokenize eos_mask.tokens.view tar_pitch_midi.append reader.readlines sqlite3.connect.close pyworld.d4c torch.empty self.StructUnit lyric.strip fairseq.models.register_model_architecture torch.cumsum args.root_path.rstrip torch.exp new_e.append lm_score.append torch.tensor.view avg new_sent.append super preprocess_chord_16.t2e df.set_index.iterrows position_ids.unsqueeze.masked_fill_ output.extend cur_song_return.append self.key_constraint midi_preprocess.steps.track_separate.cal_file_features self.wte.size read_lyrics check_length poser.convert_tokens_to_ids self._prefix_tokens utils.midi_chord.infer_chords_for_midi.dump merge_track_ops.merge_lead miditoolkit.MidiFile unicodedata.normalize.strip a.append lyric.split src_tokens.index_select.size torch.nn.DataParallel.parameters writer midi_obj.instruments.notes.append full_finalizer.tokenize nll_loss.sum.size k.hparams_.type fairseq.models.roberta.RobertaModel.from_pretrained.cuda _sample_next_token_ids.unsqueeze cnsenti.Sentiment torch.nn.CrossEntropyLoss not_duplicate transformers.EncoderDecoderModel.generate numpy.linspace symbols_to_strip_from_output.union self.model.reorder_incremental_state attn_loss_each_layer.append max cd.split bar_chords.append load_vocab dataset.set_epoch res.append i.melody_seqs.append check self.downsampling split_sents sents.replace.split f1.write aligned_melody.append note_syllable_mapping.append top1_idx.self.tgt_dict.string.split torch.nn.functional.multi_head_attention_forward finals.join.rstrip lyrics.strip.split self.ln_f loss_fct source_sent_ids.index_select.insert batch_final.torch.Tensor.long.to min sample_lang.get filter_tracks _build_init_nodes new_phone.split agg_logging_output.items full_poser.tokenize self.out_proj.size align_l.append string.strip torch.nn.functional.linear collections.defaultdict note_string.split multiprocessing.Manager.dict torch.tensor.index_select torch.cat.new_zeros ratio.tokens.eq.unsqueeze x.item exit instru.notes.sort f.readlines.split beam_size.bsz.torch.arange.unsqueeze.type_as.resize_ open.readline self.wte src_dict.index self.aux_constraint fairseq.search.PrefixConstrainedBeamSearch results.append numpy.zeros full_tokenizer.tokenize input_ids.unsqueeze.view full_sentence.append file.split n.append output_str_list.append ret.append score_config.Song.get_pitch path.split self.search.init_constraints cnsenti.Sentiment.sentiment_calculate numpy.isin multiprocessing.Pool.map p.append en_constraintor.BeatConstraintor.rhythm_constraint getattr item.max_instruments.item.mask_prob.self.vocab.len.max_instruments.max_bars.generate_mask.reshape.flatten normalize_song prefix_toks.eq.view instrument.name.rsplit _select_results.append self.upsampling torch.masked_select prefix_tokens.size midiutil.MidiFile.MIDIFile.addTrackName tgt_item.tolist argparse.ArgumentParser.parse_args not_mono s.replace.replace sample sent.item input_file.readlines lang.dicts.eos loss.mean.backward score_config.Song.get_pos lyric_str_list.append beam_size.step.prefix_tokens.unsqueeze.repeat.view ImportError is_word numpy.random.rand melody_sent.append midi_utils.encoding_to_midi shift_notes_time new_cands_to_ignore.ge score_vals.append new_instrs.sort targets.sum.long midi_utils.enc_vel re.findall open.write numpy.vstack base_architecture pypinyin.lazy_pinyin fairseq.utils.split_paths attn_weights.mean.size samples_final.append score.score_en.get_score torch.nn.Linear self.tgt_dict.string attn_weights.mean.mean line.strip.strip track_id.track2instru.instru2notes.append midi_file.markers.append x.copy get_vocab log.get numpy.ones_like NotImplementedError fairseq.models.transformer.base_architecture torch.FloatTensor self.embed_positions count_overlapping_lines tokens.beam_size.bsz.torch.arange.unsqueeze.type_as.to miditoolkit.midi.containers.TimeSignature note_start_dict.values logging.getLogger.info get_pitch_count pow get_id pypianoroll.Multitrack.parse_pretty_midi get_score bbsz_idx.banned_tokens.torch.tensor.long get_syllables.insert self.layers.extend generate_mask criterion midi_key_shift utils.enc_tpo utils.swap_value fairseq.models.roberta.RobertaModel.from_pretrained.eval fill_template source_word_ids.unsqueeze.repeat new_meldoy.append ret_melody.append features.file_names.unique src_strs.append input_ids.unsqueeze.unsqueeze n_start.append lyrics.translate scipy.stats.mode samples.get fairseq.models.transformer_lm.TransformerLanguageModel.from_pretrained.cuda args.langs.dicts.unk hparams.update src_tokens.eq.any torch.ones get_intervals criterion.__class__.reduce_metrics prepare_train_data.build_files_separate lang.dicts.mask fairseq.utils.extract_hard_alignment fairseq.options.eval_bool miditoolkit.midi.containers.Lyric prev.append _is_control src_dict.eos signal.alarm sent.append items_.append rk_list.append unicodedata.category.startswith args.langs.dicts.pad word_utils.get_sents_pos all_midi_path.append fairseq.sequence_scorer.SequenceScorer active_scores.view.view mido.MidiFile new_notes.append numpy.random.random self.sp.PieceToId i.i.bar_to_pos.pos_to_tempo.append word.append torch.nn.DataParallel.train torch.sigmoid numpy.max x.transpose.transpose numpy.any net_input.size numpy.array.append full_finalizer.convert_tokens_to_ids.split infer_chords_for_midi song.get_tone midi_tuple.str.encode.hashlib.md5.hexdigest pandas.DataFrame.drop tokens_group_by_bar.append dataset.ordered_indices fairseq.checkpoint_utils.load_checkpoint_to_cpu KMP_search model.encoder.forward_torchscript bpe_tokens.extend bar_chords.split.append optimizer.step tmp_sent.append joblib.load.get_feature_names j.pitch_list.i.pitch_list.SequenceMatcher.ratio os.path.join _prev_value.view t2.append items2events candidates_bars.append samples.s.pad_idx.s.ne.long.sum.torch.LongTensor.index_select alignment.max.item numpy.zeros.read bisect.bisect_left iter OctupleTokenDataset len.append flat_note_seq.copy sentencepiece.SentencePieceProcessor batch_idxs.bsz.tokens.view.view.view utils.get_songs fairseq.data.iterators.EpochBatchIterator miditoolkit.midi.containers.Note numpy.array torch.nn.Embedding sliding_window detect_lines_helper args.valid_lang_pairs.split src_str.strip self.is_finished len EmptySequenceError longestDupSubstring torch.utils.tensorboard.SummaryWriter.add_scalar assign_chords_for_sequence_by_template tgt_dict.pad get_rhythm_score tmp1.split lprobs.view.size l.readlines inner_states.append predict_track_with_model numpy.hstack gt_d.astype.astype i.to.size str.split _generate_addtional_token_ids sent.finalized.append sorted.append model.keys self.replicate_first_beam.size new_src.extend GPT2Model file.endswith self.tgt_dict.eos sorted self.pitch_ruler._accept_notes preprocess.gen_dictionary all item.lower flag.split unicodedata.category finalizer.convert_tokens_to_ids v.songs.extend self.aux_transform label_dict.string attn_mask.repeat.repeat word_utils_en.get_in_word_pos list.append self.vocab.items midiutil.MidiFile.MIDIFile.addNote self.src.prefetch f.close criterion.__class__.aggregate_logging_outputs.values self.aggregate_logging_outputs.items tran.readlines i.flatten_target_samples.np.array.reshape int.item _CHORD_KIND_PITCHES.keys self.decoder.get_normalized_probs torch.nn.Sequential int.split INSTRUMENTS.index subbeats.extend numpy.maximum output.split.append MaskedLMDictionary collections.Counter math.exp wav.astype line.replace.split.join.lower s_midi_path.split.split align_idx.samples.view f_txt.write hasattr rows.append mono_tracks.append aligned_lyric.extend open.flush model.decoder.reorder_incremental_state_scripting self.collate device.torch.long.next_final.finalizer.convert_tokens_to_ids.torch.tensor.unsqueeze split_para_exists get_syllables.append input.items loss.isinf.any clean dict self.drop hparams.clear midi2notes argparse.ArgumentParser.add_argument s.ne dtw_mean.append self.token_constraintor.pitch_constraint i.repeat_pair.append x.eval numpy.fromstring self.strct_buffer.update_crnt_label self.out_proj files.append text.join.replace self.has_encoder numpy.random.choice self.StructBuffer meloform.meloform_transformer.TransformerMeloFormModel.from_pretrained.encode fairseq.data.data_utils.numpy_seed args.model_dir.split flac_dir.split number_of_notes fairseq.models.transformer_lm.TransformerLanguageModel.from_pretrained top_idx.self.tgt_dict.string.split _get_mt_dataset_key zipfile.ZipFile.namelist torch.nn.functional.softmax init batch_idxs.bsz.tokens.view.view new_cands_to_ignore.ge.any torch.cat.contiguous phone.split copy.deepcopy.split full_line.append kwargs.get.long self.beam_size.src_tokens.expand.contiguous.view.expand self.beam_size.src_lengths.expand.contiguous torch.nn.functional.linear.transpose stop_file.readlines self.token_constraintor.pos_constraint batch_final.torch.Tensor.long self.beam_size.src_tokens.expand.contiguous.view.size numpy.argmax segment_text n_ctx.generated_sentences.unsqueeze midi_obj_2.instruments.append convert_to_midi_obj.dump self.layer_norm batch_pos.torch.Tensor.long self.generate torch.gather new_beats.append self.lm_model.eval bbsz_idx.numel extract_melody_and_lyric text.join.replace.strip gen_dictionary self.filter_indices_by_size self.generate_dummy_batch en_word_utils.get_structure_mask tuple.sort x.get align.strip.split input_ids.unsqueeze.size float torch.stack self.strct_buffer.get_last_pitch _struct.split self.vocab.dummy_sentence.insert prepare_lyrics cut_window adapt.append self.convert.numpy targets.preds.sum _control_rhymes new_instr.notes.sort src_words.append collections.OrderedDict.keys src_key.self.encoders torch.from_numpy torch.tensor.to duration_change midi_preprocess.steps.track_separate.remove_file_duplicate_tracks tone.append en_constraintor.BeatConstraintor torch.utils.tensorboard.SummaryWriter SongWrapper.Song.get_pitch e2b self.random_word get_avg_pitch bbsz_idx.size fairseq.models.transformer.TransformerModel.from_pretrained.translate cur_word.append numpy.zeros.write infer.stage_3_get_t2m copy.deepcopy.write numpy.repeat torch.load torch.isnan numpy.sum cand_indices.device.bsz.torch.arange.masked_select transformers.modeling_outputs.BaseModelOutputWithPastAndCrossAttentions write_to_txt self.pitch_debug.append targets.detach.cpu tmp_lyric.append word_list.append batch_beat.torch.Tensor.long.to fairseq.metrics.log_scalar bool score.score_zh.get_score.index _normalize_logits attn.transpose.contiguous.view bbsz_idx.cpu_tokens.tolist tmp_lyr.append prefix_toks.eq.type_as punct.append self.get_sents_num fairseq.models.register_model ratio.tokens.eq.any target_word_ids.unsqueeze fairseq.models.build_model.load_state_dict args.target_langs.split chord.strip.strip src_key.self.encoders.size batch_idxs.bsz.attn.view.view.index_select difflib.SequenceMatcher self._run_strip_accents source_sent_ids.unsqueeze.repeat src_lengths.max.item load_config merge_track_ops.keep_long_notes os.path.exists module.GPT2Config.from_json_file.to_json_string prev_output_tokens.eq.int.cumsum pitches.append attn_weights.mean.transpose args.langs.dicts.eos self.transformer midi_obj.lyrics.append subset.f.json.load.items tokenizer not_too_high dict.keys nlp gt.np.array.reshape meloform.meloform_transformer.TransformerMeloFormModel.from_pretrained.eval batch_idxs.bsz.scores.view.view melody.split generate_sentence lyric2beats.translate.split src_tokens.size.torch.tensor.to.max f.readline.strip padding_mask.unsqueeze.type_as pos_ids.view.view midi_obj.instruments.append traceback.print_exc poly_tracks.append sample.split.split fairseq.models.build_model torch.index_select numpy.zeros_like os.listdir torch.cat.extend target_prefix.first_beam.all pandas.read_csv.iterrows wav_mels.append miditoolkit.midi.containers.TempoChange lead_notes.append select_notes.split torch.tensor.type_as self.pos_debug.append i.to.view mask_length.mask_start.mask_start.s.copy bsz.torch.arange.type_as WordpieceTokenizer pretty_midi.PrettyMIDI batch_idxs.numel.torch.arange.type_as get_inner_pos outs.write all_notes.append len.insert torch.cuda.device_count meloform.meloform_transformer.TransformerMeloFormModel.from_pretrained.generate self.in_word_transform fairseq.quantization_utils.quantize_model_scalar.set_num_updates self.sp.Load collate_fn transformers.BertGenerationEncoder.from_pretrained self.aggregate_logging_outputs isinstance tokenizations.tokenization_bert.BertTokenizer.convert_tokens_to_ids args.metric.metric_func utils.enc_ts sents2words.append fairseq.models.transformer_lm.TransformerLanguageModel.from_pretrained.cuda.sample labels.append get_melody args.word_mask_keep_rand.split scaled_loss.backward fairseq.data.Dictionary.load n.eval dataset.size AP note_start_dict.keys i.hypos.item notes.split subpos.extend torch.topk f.read enumerate.insert cand_indices.device.bsz.torch.arange.masked_select.numel logits.size os.path.isfile net_input.pop joblib.load midi2items generate_raw flatten Event enc_vel en.split pad_idx.s.ne.long.sum roberta.task.load_dataset attn_weights.mean.view numpy.maximum.reduce torch.cat.view transformers.get_linear_schedule_with_warmup self.init_weights m.readlines.split ordered_notes.append target_sent_ids.unsqueeze.repeat n.lower self.PitchRuler boundry.append pickle.dump target_word_ids.unsqueeze.repeat eout_i.encoder_outs._replace EnsembleModelWithAlignment word_utils.get_structure self.pitch_ruler._get_num en_config.PositionDebug logging.getLogger buildNext os.path.splitext.isupper aligns.append numpy.full self.wordpiece_tokenizer.tokenize n_ctx.generated_tokens.unsqueeze Line load_indexed_dataset merged_infos.append avg_attn.add_ sents2words segment.append self.replicate_first_beam.view midi2hz prefix_mask.lprobs.scatter prefix_toks.eq.size self.vocab.get os.path.normpath.startswith fairseq.metrics.aggregate key_debug.append cls f.read.rstrip.split magenta_chord_recognition.infer_chords_for_sequence self._generate TypeError logger.warning yaml.safe_load lyrics_file.readlines cut_melody_test.extend data.append net_input.sum full_pos.extend src_tokens.ne os.makedirs get_sw_score super.build_generator step.prefix_tokens.unsqueeze.repeat miditoolkit.midi.parser.MidiFile.dump get_dur_count events.append datetime.datetime.now target_sent_ids.index_select.append i.lines.strip args.root_path.rstrip.split self.sp.EncodeAsPieces torch.masked_select.numel self.src_dict.eos cum_unfin.append re.compile phonemizer.separator.Separator numpy.random.randint self.bias_k.repeat beam_size.step.prefix_tokens.unsqueeze.repeat.view.unsqueeze self.can_reuse_epoch_itr multiprocessing.pool.Pool instru2program.keys F final_reverse.strip numpy.save hyp_d.astype.astype super._generate self.calculate_banned_tokens remove_suffix res_str_list.split text.replace.lower miditoolkit.containers.Note soundfile.write print_info eval_para_dataset.keys torch.LongTensor tgt_lengths.sum.item self.emb_layer_norm finals.join.rstrip.append new_outs.append map self.adaptor pandas.DataFrame.map song_seq.append SongWrapper.Song.get_note sentence_ids.view.view prev.extend loss.detach.item pyworld.stonemask syl.append en_config.PitchDebug beater.convert_ids_to_tokens hidden_states.view.size adapt self.datasets.values self.src_lang_id.torch.LongTensor.expand select_melody e_segment.append self.beam_size.src_tokens.expand.contiguous.view.ne pattern.append _prepare_init_inputs full_beater.tokenize json.dumps preprocess_chord_16.encoding_to_MIDI.dump src_tokens.max_len.beam_size.bsz.torch.zeros.to.long f.read.strip.split song.get_note self.model.max_decoder_positions os.getenv fairseq.utils.deprecation_warning generate_midi_combine adapt_e lead_notes.sort torch.nn.Dropout word.split src_item.tolist tokenizations.tokenization_bert.BertTokenizer.tokenize strct_mask.append batch_idxs.numel.torch.arange.type_as.unsqueeze cands_to_ignore.any.all hyp.np.array.reshape torch.mean join csv.writer.writerow device.torch.long.context_sentences.torch.tensor.unsqueeze check_alignment music_mass_dataset.MusicMassDataset warnings.filterwarnings sum.write score_config.Song.get_tone fairseq.search.Sampling torch.log load_midi min.extend device.torch.long.next_beat.beater.convert_tokens_to_ids.torch.tensor.unsqueeze hz2midi self.replicate_first_beam.type_as chord.split.index get_pairs sequence_note_pitch_vectors attn_mask.repeat.unsqueeze self.mask_word note_durations self.embed_tokens numpy.round torch.device beats_label.append fairseq.utils.parse_embedding abs i.finals.invalids.add flatten_notes.extend _is_punctuation multiprocessing.Pool.join self.beam_size.src_tokens.expand.contiguous.view.to preprocess_chord_16.v2e args.langs.split batch_sentence.torch.Tensor.long sample_lang.size self.src_dict.index get_tonality torch.LongTensor.sort torch.tensor aligns_l.append transformers.AdamW os.walk.sort hyp_str.strip l.replace self.model.forward_decoder torch.nn.DataParallel.to torch.bmm argparse.ArgumentParser.parse_known_args model.get_targets format.append batch_idxs.bsz.attn.view.view item.start.item.end.np.clip.item cur_song_return.extend transformers.EncoderDecoderModel.eval list.sort lyrics.split cand_indices.eq.type_as textblob.TextBlob SongWrapper.Song fairseq.utils.load_embedding score_config.Song.print_debug self.src_vocab.eos target_sent_ids.unsqueeze self.get_head_mask format song.strip src_tokens.device.new_order.to.long.to cut_lyric_test.extend numpy.load io.BytesIO array_creation_by_count loss.data.item numpy.clip dec_tpo aligned_lyric.append self.project_out_dim attns.append bisect.bisect_right notes_no_drum.append attn_mask.repeat.new_zeros s.isdigit args.source_langs.split program.midi_obj.instruments.notes.append hasattr.save_pretrained item.split criterion.__class__.grad_denom self.wfe self.buffered_future_mask dict.values time.sleep device.torch.long.context_beats.torch.tensor.unsqueeze get_pairs.add src_tokens.max_len.beam_size.bsz.torch.zeros.to.long.fill_ get_hotwords targets.sum.sum extract_target tgt_vocab.dummy_sentence self.beam_size.src_tokens.expand.contiguous.view fairseq.dataclass.utils.gen_parser_from_dataclass utils.top_k_top_p_filtering get_songs dataset_dir.split pandas.to_numeric torch.jit.is_scripting polish_chord c.readlines cand_indices.eq.size miditoolkit.midi.parser.MidiFile self.model.forward_align file_line.strip fairseq.criterions.build_criterion source_word_ids.index_select.index_select rpls.append bar2.split torch.jit.annotate pad_idx.s.ne.long multiprocessing.Pool full_finalizer.convert_tokens_to_ids.strip self.pad.src_tokens.ne.self.eos.src_tokens.ne.long key_rs.append p.apply_async torch.autograd.profiler.record_function position_ids.unsqueeze.unsqueeze patterns.append shutil.copy2 numpy.concatenate source_sent_ids.index_select.extend logging.getLogger.warning midiconvert.number_to_note ValueError label_lengths.append math.log2 torch.arange tqdm.tqdm return_list.append os.path.basename cut_melody_valid.extend song.get_aux_flag isVowel fairseq.search.DiverseSiblingsSearch lang.dicts.pad score.score_zh.get_score in_word.append line.split.split song_seqs.append sents.append pandas.DataFrame.copy pinyin.pinyin_dict.append setattr get_structure str len.endswith beat_ids.view.view cur_sent_durations.append trend.split self.replicate_first_beam.index_select invalid_songs.add torch.topk.size self.wbe SongWrapper.Song.get_first_note ratio.tokens.eq self.transpose_list miditoolkit.containers.TimeSignature batch_sentence.torch.Tensor.long.to get_pitch_histo os.path.isdir textrank4zh.TextRank4Keyword.get_keywords k.contiguous.view numpy.random.permutation fairseq.utils.move_to_cuda timeout aligned_melody.extend items.sort self._is_chinese_char i.strip.replace conn.cursor.execute sen.append dataset.batch_by_size numpy.minimum.reduce criterion.__class__.aggregate_logging_outputs.items self.lm_head e2d drum_idx.append find_backward self.replicate_first_beam.gather torch.ne encoder_hidden_states.size sample.size torch.topk.numpy fairseq.data.data_utils.load_indexed_dataset self.wordpiece_tokenizer.tokenize.extend samples_token.append unicodedata.normalize cand_beams.add lm_logits.contiguous.size align_weights.align.align.attn.log.sum.isnan f.read.rstrip device.torch.long.context_poses.torch.tensor.unsqueeze prepare_train_data.get_shuffled_samples full_beater.convert_tokens_to_ids cand_indices.eq beat.copy list tuple.append hyp_str.strip.split target.index_select.append numpy.histogram self.vocab.dummy_sentence.append get_tick src_vocab.dummy_sentence torch.multinomial elem.item midiutil.MidiFile.MIDIFile.addTempo fairseq.models.build_model.eval meloform_dataset.MeloFormDataset phrase.dicts.append all_lyric.extend src_tokens.device.new_order.to.long text.join.replace.strip.split node.wordid.tolist line.split.strip time_to_pos new_config.items wav.astype.astype full_sentencer.convert_tokens_to_ids.strip full_beat.extend joblib.load.transform line.strip.lstrip w.write separate_sentences temp_res.append self._no_repeat_ngram i.split self.mld_transform average.score.scores.append jieba.lcut max_len.avg_attn_scores.size.beam_size.bsz.torch.empty.to torch.sum bass_model.predict get_duration_count self.model.cuda song_id_valid.extend load_meloform_dataset song_seq.extend notes_path.split of.write fairseq.search.LexicallyConstrainedBeamSearch midi_fn.lstrip lyrics_ali.append thulac.thulac.cut target.index_select.index_select merge.index_select self._run_strip_accents.lower output.detach.cpu module.GPT2LMHeadModel.from_pretrained s.extend find_chorus.Find_Chorus t.write chord_time.append lyric2beats.translate.tolist tf.transform.tocoo find_forward args.mass_steps.split self.has_incremental_states f_id.write sents.strip.split numpy.dot q.contiguous.view source_sent_ids.index_select.index_select MIDI_to_encoding tgt_datasets.append src_tokens.eq.long all_scales.items f.readline.strip.split get_ps_score chord_model.predict tmp_candidates.append fairseq.models.transformer.Embedding bsz.torch.arange.view.repeat numpy.tile.sum syllables.split sqlite3.connect tensor.view.view score.score_en.get_score.index melody.notes.append net_output.sample.model.get_targets.view.ne transformers.BertGenerationDecoder prefix_mask.prefix_toks.unsqueeze score.sklearn.metrics.__dict__ transformers.EncoderDecoderModel.load_state_dict aux_rule mt_para_dataset.keys nf_id.unsqueeze pickle.load yaml.safe_load.update notes_to_str process get_chords sample.get self._run_strip_accents.rstrip phonemizer.separator.Separator.txt.phonemize.replace.replace multiprocessing.Lock.release sqlite3.connect.cursor torch.LongTensor.sum x.split source_word_ids.unsqueeze step.prefix_tokens.unsqueeze midi_preprocess.steps.track_separate.predict_labels multiprocessing.Pool.close torch.nn.DataParallel.forward get_phrase_position_dicts len.startswith instru2track.items replace_phrase_ids.append similar_info.append midi_file.mido.MidiFile.save load_data self.label_dictionary.index sqlite3.connect.commit target_word_ids.index_select.append occupation_rate notes_list.append lyrics.translate.translate m.readlines writer.write target.sum.sum collections.Counter.keys logits.dim join.split tqdm.trange fairseq.utils.resolve_max_positions syllables.clean.split tmp.split textrank4zh.TextRank4Keyword join.index lines_to_sort.sort cal_overlap re.compile.match _prev_key.view full_sentencer.convert_tokens_to_ids.split pandas.DataFrame.reset_index char.isdigit batch_pos.torch.Tensor.long.to librosa.load tokenizations.tokenization_bert.BertTokenizer.convert_ids_to_tokens cand_size.torch.arange.type_as sentencer.convert_tokens_to_ids fsim.write infer.stage_4_get_m2m new_sample.append miditoolkit.Note criterion.logging_outputs_can_be_summed eval self.decoders.values pyworld.synthesize attn_loss.sum.sum tokenized_data_path.rsplit self.strct_transform SongWrapper.Song.print_debug result.append sents.strip.strip lm_logits.contiguous.view model.encoder.reorder_encoder_out fairseq.metrics.log_speed word_utils_en.get_aux_mask PITCH_CLASS_NAMES.index midi_preprocess.steps.track_separate.add_labels _sample_next_token_ids lang.dicts.unk self.bias_v.repeat target.sum.detach line.split.append hidden_states.view.view transformers.modeling_gpt2.Block lines_to_sort.append re.sub.strip x.strip bar1.split full_sentence.extend src_tokens.size.torch.tensor.to layer.self.h.attn.prune_heads merge_str.split tuple.extend itertools.product os.path.split signal.signal aligns_m.append MusicBERTEncoder cur_sent_notes.append song.get_pos tgt_dict.unk criterion.__class__.aggregate_logging_outputs a.sort lyric.append b2e fairseq.tasks.register_task get_tokens beam_size.bsz.torch.arange.unsqueeze line.strip.split XTransformerEncoder tick2step phonemizer.separator.Separator.txt.phonemize.replace numpy.abs word_utils_en.get_keyword_mask self.bpe encoding.append len.extend get_duration_histo lyrics_ali.remove whole_melody.extend select_chords.append _key_chord_distribution tgt_str.split.split numpy.set_printoptions hypo.item x.self.upsampling.view get_aux_mask syllables.strip.split r.readlines src_tokens.index_select.index_select best_segment self.get_noteidx_by_time apex.amp.scale_loss miditoolkit.Instrument numpy.vstack.append notes_str.split infer_pipe torch.cat.dim os.mkdir sum accept_pos_words os.getcwd f.get icecream.ic whitespace_tokenize.extend tgt_key.self.decoders avg_attn_scores.size self.strct_buffer.get_last_pitch.append miditoolkit.midi.containers.Instrument torch.clamp text.replace.replace sentences.append module.GPT2LMHeadModel midi.instruments.append samples_sentence.append final_ids.view.view enc_ts tmp_1.append fairseq.utils.strip_pad dictionary.pad ss.write device.torch.long.context_tokens.torch.tensor.unsqueeze self.syl_transform cls.BertTokenizer.super._from_pretrained targets.detach.cpu.numpy fairseq.criterions.register_criterion src_datasets.append midi_to_encoding duplicates.append detect_lines f.read.split torch.Tensor midi_adjust.split_midi args.root_path.rstrip.split.prepare_fn sample.keys prefix_toks.eq.any numpy.unique self.head_dim.self.num_heads.bsz.tgt_len.q.contiguous.view.transpose.contiguous word_utils.get_aux_mask is_sep.cumsum.unsqueeze splice rest_tmp.append generate_pitch_duration_sequence EnsembleModel chord_file.readlines self.ids_to_tokens.get note_syllable_mapping j.segments.append item.split.split x.count torch.cat.unsqueeze torch.nn.functional.softmax.detach random.random tokens.cand_size.torch.arange.type_as.to meloform.meloform_transformer.TransformerMeloFormModel.from_pretrained.cuda self.vocab.eos pm_new.instruments.append cur_pitch.append full_tokenizer.convert_tokens_to_ids self.encoder.items tokens.join.replace tgt_words.append self.vocab.dummy_sentence.extend note_items.append input_path.os.path.basename.split polyphony_rate self.extract_alignment melody_str_list.append attn_loss_layer.view.pow notes_string.split seperate max_len.beam_size.bsz.torch.zeros.to XTransformerModel whitespace_tokenize.append self.compute_loss model.get_normalized_probs miditoolkit.MidiFile.dump utils.midi_chord.infer_chords_for_midi active_bbsz_idx.view.view self.beam_size.src_lengths.expand.contiguous.view transformers.get_linear_schedule_with_warmup.step self.tgt_dict.bos self.vocab.pad sliding_melody_train.extend beat_reverse.strip syl_sents.split yaml.safe_dump dur.append lyric.extend Args model.generate.tolist attn_weights_float.masked_fill.masked_fill self.invert_attention_mask miditoolkit.TempoChange line.replace p.imap_unordered io.open i.samples.get args.hparams.split collections.Counter.values fairseq.quantization_utils.quantize_model_scalar.train self.token_constraintor.print_debug i.strip read_meta_data lm_score self.add_symbol l.upper torch.mul midi_preprocess.steps.process_midi_file.process_midi_file src_tokens.beam_size.bsz.torch.zeros.to.eq midi_preprocess.steps.merge_tracks_for_train.gen_merge_midi begin.append merge_track_ops.keep_track_with_most_notes torch.logsumexp bar_chords.split numpy.nonzero itertools.count get_chorus v2e utils.temperature_softmax forward_backward bar_to_pos.append seq_gen_cls meloform_dataset_infer.MeloFormDatasetInfer fairseq.data.AppendTokenDataset key_rule yaml.safe_load.items tones.t.t.Counter.most_common open.close full_beat.append batch_idxs.bsz.attn.view.view.size i.tokens_clone.eq.long self.sp.DecodeIds Exception json.loads en_constraintor.BeatConstraintor.print_debug batch_token.torch.Tensor.long.to get_value scipy.ndimage.filters.gaussian_filter1d self.replicate_first_beam x.startswith torch.utils.checkpoint.checkpoint new_patt.append word_utils.clean parameter.numel os.cpu_count config_chains.append collections.OrderedDict masked_attention_decoder_layer.MaskedAttentionDecoderLayer index.node.wordid.tolist argparse.ArgumentParser input csv.writer cur_melody.append get_span attn.transpose.contiguous sys.argv.split.split best_segment.sort full_pos.append innerpos_reverse.strip _key_chord_viterbi get_noteidx_by_time transformers.AutoTokenizer.from_pretrained en_constraintor.TokenConstraintor numpy.pad self.lyr_transform note_items.sort padded align_sent align.align.attn.log get_pitch_class_histogram t2e fairseq.models.transformer.TransformerModel.add_args self.search.prune_sentences
@developer
Could please help me check this issue?
May I pull a request to fix it?
Thank you very much.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by locating the dependency manifest for the Python project and compare its constraints with the package versions listed in the issue. Check the listed scikit-learn, nltk, tqdm, transformers, pypinyin, scipy, librosa, and soundfile API calls, then verify that the project still runs without import or API errors after the constraints are updated.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, scikit-learn
- Domain
- build-system
- Issue type
- Refactor
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 30/100