NVIDIA/NeMo

Issues

• Initialize a Parakeet Cache-Aware Streaming model's encoder from an offline model.

  #11250 opened 10 days ago by gabitza-tech
  0

• Using MSDD model with a different speaker embedding model

  #10681 opened 10 days ago by MahmoudAshraf97
  2

• How to convert downloaded local models into Nemo files？

  #11248 opened 10 days ago by llll111111
  0

• How to implement weight decay towards the pre-trained model?

  #10738 opened 10 days ago by sedol1339
  2

• Loading 70B model from .nemo checkpoint takes very long time

  #10745 opened 10 days ago by jiaji-huang
  2

• Global shape mismatch for loaded ((1024, 768)) and expected ((512, 768)) tensor for key model.embedding.position_embeddings.weight

  #10715 opened 11 days ago by Alireza3242
  3

• Unable to merge lora weights: "world_size (1) is not divisible by 4"

  #10782 opened a month ago by Elan456
  2

• Resuming from a checkpoint that ended before the epoch ended and your dataloader is not resumable

  #10797 opened a month ago by AudranBert
  1

• [Question] Converting a Megatron-LM ckpt to Nemo

  #10831 opened a month ago by abgoswam
  3

• Learning Rate Sudden Dropped to min_lr after Warm-up Steps

  #11149 opened 16 days ago by zixianwang2022
  1

• Conversion script of phi3 from HF to Nemo

  #10825 opened a month ago by zhouxie000
  1

• Need env.yaml or requirements.txt

  #11231 opened 13 days ago by jugal-sheth
  1

• [Question] Is there a way to get the results of each layer of a model?

  #11224 opened 13 days ago by GO0108
  0

• [NeVa Pretraining] Vision Encoder Created on All GPUs During Pipeline Parallelism

  #10805 opened a month ago by Esthesia
  1

• Nemo run pretrain raised Exception: No dot product attention support for the provided inputs

  #11218 opened 13 days ago by ycchenzheng
  0

• Availability of Meeting MSDD models

  #11207 opened 14 days ago by uro-sh
  0

• How can I get stt_en_fastconformer_ctc_small pretrain model??

  #11204 opened 14 days ago by PhamDangNguyen
  0

• NameError: name' flash_attn_with_kvcache 'is not defined

  #11200 opened 14 days ago by 1074224619
  0

• How to use wandb sweep for hyperparameter search when finetuning with llama2

  #11151 opened 14 days ago by PurvangL
  1

• Would it be possible to use pyctcdecode with ASR cache-aware streaming?

  #11182 opened 15 days ago by gabitza-tech
  0

• tutorials/tts/Inference_ModelSelect.ipynb is throwing error when installing dependancies

  #11178 opened 15 days ago by bnarasimha
  0

• `IPython` should be included in the requirements

  #10772 opened 2 months ago by MahmoudAshraf97
  2

• Deploy ASR STT Streaming model

  #11019 opened a month ago by rkchamp25
  0

• How to export fastconformer based asr model to support torchscript model?

  #11161 opened 16 days ago by zw76859420
  0

• How to finetune the NEST model with CTC Loss for ASR task?

  #11163 opened 16 days ago by zw76859420
  0

• When do add the code for Target Speaker Extraction, thank!

  #10830 opened a month ago by haha010508
  2

• convert_qwen2_hf_to_nemo error

  #11142 opened 17 days ago by huangqingyi-code
  0

• The SDXL Infer output image is full of noise

  #10938 opened a month ago by blacklong28
  6

• How to Visualize Backward Computational Graph

  #11139 opened 18 days ago by zixianwang2022
  0

• Slurm interactive mode, transcribe_speech_parallel.py gets stuck on consecutive runs

  #11105 opened 21 days ago by itzsimpl
  0

• NeMo2.0 nemorun llm export ValueError: PyTorch DDP is not enabled for mcore optimizer

  #10939 opened 22 days ago by lifeiteng
  2

• Megatron Multilingual En Any 500M

  #10976 opened 23 days ago by rogerwelo
  2

• Code adoptability for other pretrained ASR

  #11074 opened 23 days ago by Amg9794
  0

• When training ASR models, it saves .nemo 2 times in a row

  #10798 opened a month ago by AudranBert
  1

• activate subscription: Not able to find the Promotional Code, Serial Number, or Token to run the inference script?

  #11064 opened 23 days ago by Alla-Abdella
  0

• srun issue with nemorun

  #10997 opened a month ago by RachitBansal
  3

• Unable to export MSDD model to pt or ONNX

  #10999 opened a month ago by jingzhaoo
  1

• RuntimeError: Function 'AcosBackward0' returned nan values in its 0th output.

  #11025 opened a month ago by gor2000
  0

• [Question] Pipeline Parallel for Mamba Megatron

  #10900 opened a month ago by zixianwang2022
  1

• NeMO dependency issues on HuggingFace Hub (for ASR models)

  #10940 opened a month ago by bhavnicksm
  2

• Converting Mamba to tp4: RuntimeError: The size of tensor a (18560) must match the size of tensor b (4640) at non-singleton dimension 0

  #10966 opened a month ago by zixianwang2022
  0

• NeuralDiarizer with the telephonic config mix speakers at the very beginning of shorter audio files (less than 2 minutes duration)

  #10988 opened a month ago by uro-sh
  0

• canary-1b is not exportable

  #11004 opened a month ago by pdufour
  0

• Modules fail for Dreambooth example

  #10888 opened a month ago by paulaserna16
  1

• global batch size at different sequence length

  #10905 opened a month ago by erhoo82
  0

• Link Not Found at Mamba Tutorial

  #10899 opened a month ago by zixianwang2022
  0

• Converting trained llama 2 checkpoint to hf gives "invalid key" error

  #10884 opened a month ago by jiaji-huang
  0

• Add Hydrarunner to oomptimizer

  #10882 opened a month ago by bonham79
  0

• SFT stage use context parallel with flash attention error

  #10876 opened a month ago by ARQlalala
  0

• Allow OOMtimizer tokenizer point towards just parent directory

  #10870 opened a month ago by tbartley94
  0