Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / philschmid/deep-learning-pytorch-huggingface issues and pull requests
#78 - try to run aha-grpo !
Issue -
State: closed - Opened by curater 9 days ago
- 2 comments
#77 - RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:3 and cuda:0!
Issue -
State: open - Opened by wotulong 14 days ago
- 2 comments
#76 - IndexError: pop from an empty deque in run_r1_grpo.py
Issue -
State: open - Opened by JunMa11 17 days ago
- 3 comments
#75 - R1 grpo
Pull Request -
State: closed - Opened by philschmid 18 days ago
#74 - Dpo 2025
Pull Request -
State: closed - Opened by philschmid 26 days ago
#73 - Labels leaking in training data
Issue -
State: open - Opened by hamzahass about 1 month ago
#72 - Typo fine-tune-llms-in-2025.ipynb
Pull Request -
State: open - Opened by ChristianBernhard about 2 months ago
- 3 comments
#71 - Modernberg
Pull Request -
State: closed - Opened by philschmid about 2 months ago
#70 - 2025
Pull Request -
State: closed - Opened by philschmid about 2 months ago
#69 - Added crucial comment in fine-tune-embedding-model-for-rag.ipynb
Pull Request -
State: open - Opened by ChristianBernhard 2 months ago
#68 - Fixed typo in fine-tune-embedding-model-for-rag.ipynb
Pull Request -
State: open - Opened by ChristianBernhard 2 months ago
#67 - Update gemma-lora-example.ipynb
Pull Request -
State: open - Opened by qgallouedec 3 months ago
#66 - [`embeddings`] Update (positive, anchor) to (anchor, positive)
Pull Request -
State: closed - Opened by tomaarsen 3 months ago
#65 - Fine-tune-llm-in-2024-with-trl.ipynb for LLAMA3.2
Issue -
State: open - Opened by anas-zafar 4 months ago
#64 - Vlm
Pull Request -
State: closed - Opened by philschmid 5 months ago
#63 - Fp inference
Pull Request -
State: closed - Opened by philschmid 5 months ago
#62 - Update deepseed-flan-t5-summarization.ipynb
Pull Request -
State: open - Opened by sharmax-vandana 5 months ago
#61 - ValueError: Must flatten tensors with uniform dtype but got torch.bfloat16 and torch.float32
Issue -
State: open - Opened by daje0601 5 months ago
#60 - OOM error in FSDP QLORA setup
Issue -
State: open - Opened by ss8319 6 months ago
#59 - added spec example
Pull Request -
State: closed - Opened by philschmid 6 months ago
#58 - Regarding the OOM issues with fine-tuning Flan-T5-xl
Issue -
State: open - Opened by Alsodream 7 months ago
#57 - Update instruction-tune-llama-2-int4.ipynb
Pull Request -
State: closed - Opened by Rishav-hub 7 months ago
- 1 comment
#56 - Quantization question:
Issue -
State: open - Opened by aptum11 8 months ago
#55 - Not able to run training/fsdp-qlora-distributed-llama3.ipynb
Issue -
State: closed - Opened by aasthavar 8 months ago
- 7 comments
#54 - St
Pull Request -
State: closed - Opened by philschmid 9 months ago
#53 - Clean up some typos; simplify some code; update some comments
Pull Request -
State: closed - Opened by tomaarsen 9 months ago
#52 - Deprecation warnings.
Issue -
State: open - Opened by hohoCode 9 months ago
#51 - Fine-tune-llm-in-2024-with-trl.ipynb not producing the outputs
Issue -
State: open - Opened by scigeek72 9 months ago
#50 - Fsdp qlora
Pull Request -
State: closed - Opened by philschmid 10 months ago
#49 - Out of Memory: Cannot reproduce T5-XXL run on 8xA10G.
Issue -
State: open - Opened by slai-natanijel 11 months ago
- 3 comments
#48 - What's the use of "messages" in dpo step?
Issue -
State: open - Opened by katopz 11 months ago
#47 - question about DeepSpeedPeftCallback
Issue -
State: open - Opened by mickeysun0104 12 months ago
#46 - Gemma
Pull Request -
State: closed - Opened by philschmid 12 months ago
#45 - Re. fine-tune-llms-in-2024-with-trl.ipynb
Issue -
State: open - Opened by andysingal 12 months ago
- 1 comment
#44 - Dpo
Pull Request -
State: closed - Opened by philschmid 12 months ago
#43 - Target modules all-linear not found in the base model.
Issue -
State: closed - Opened by kassemsabeh about 1 year ago
- 6 comments
#42 - Commit Version Bug Fix
Pull Request -
State: closed - Opened by YanSte about 1 year ago
- 1 comment
#41 - Trl
Pull Request -
State: closed - Opened by philschmid about 1 year ago
#40 - flash attention error on instruction tune llama-2 tutorial on Sagemaker notebook
Issue -
State: open - Opened by matthewchung74 over 1 year ago
- 2 comments
#39 - Precision Issue
Issue -
State: open - Opened by zihaohe123 over 1 year ago
- 4 comments
#38 - Falcon-180B "forward() got an unexpected keyword argument 'position_ids'"
Issue -
State: open - Opened by aittalam over 1 year ago
#37 - Does this work for Llama2 - Fine-tune Falcon 180B with DeepSpeed ZeRO, LoRA & Flash Attention?
Issue -
State: open - Opened by ibicdev over 1 year ago
- 11 comments
#36 - Ds lora
Pull Request -
State: closed - Opened by philschmid over 1 year ago
#35 - Instruction tuning of LLama2 is significantly slower compared to documented 3 hours fine-tuning time on A10G.
Issue -
State: open - Opened by mlscientist2 over 1 year ago
- 1 comment
#34 - Compute metrics while using SFT trainer
Issue -
State: open - Opened by shubhamagarwal92 over 1 year ago
- 1 comment
#33 - Cannot load tokenizer for llama2
Issue -
State: closed - Opened by smreddy05 over 1 year ago
- 1 comment
#32 - LLama 2 Flash Attention Patch Not Working For 70B
Issue -
State: open - Opened by mallorbc over 1 year ago
- 6 comments
#31 - Gptq
Pull Request -
State: closed - Opened by philschmid over 1 year ago
#30 - Fix Flash Attention forward for Llama-2 70b
Pull Request -
State: closed - Opened by davidmrau over 1 year ago
- 8 comments
#29 - Is the DataCollator necessary in peft-flan-t5-int8-summarization.ipynb ?
Issue -
State: open - Opened by brooksbp over 1 year ago
#28 - question about the llama instruction code
Issue -
State: closed - Opened by yeontaek over 1 year ago
- 8 comments
#27 - improvements
Pull Request -
State: closed - Opened by philschmid over 1 year ago
#26 - Llama patch for FlashAttention support fails with use_cache
Issue -
State: open - Opened by qmdnls over 1 year ago
- 2 comments
#25 - How to create a json file for create_flan_t5_cnn_dataset.py
Issue -
State: open - Opened by andysingal over 1 year ago
- 1 comment
#24 - gcc/cuda used for training
Issue -
State: open - Opened by danyaljj over 1 year ago
- 1 comment
#23 - fix
Pull Request -
State: closed - Opened by philschmid over 1 year ago
#22 - Flash attention
Pull Request -
State: closed - Opened by philschmid over 1 year ago
- 4 comments
#21 - Falcon int4
Pull Request -
State: closed - Opened by philschmid over 1 year ago
#20 - added container image for training
Pull Request -
State: closed - Opened by philschmid over 1 year ago
#19 - CPU offload when not using offload deepspeed config file
Issue -
State: open - Opened by siddharthvaria over 1 year ago
- 3 comments
#18 - Error when training peft model example
Issue -
State: open - Opened by Tachyon5 over 1 year ago
- 6 comments
#17 - Colab notebook fails
Issue -
State: closed - Opened by TzurV over 1 year ago
- 1 comment
#16 - CUDA OOM error while saving the model
Issue -
State: closed - Opened by aasthavar almost 2 years ago
- 10 comments
#15 - Does deepspeed partition the model to multi GPUs?
Issue -
State: open - Opened by vikki7777 almost 2 years ago
- 4 comments
#14 - ValueError
Issue -
State: open - Opened by Martok10 almost 2 years ago
- 4 comments
#13 - Inference on CNN validation set takes 2+ hours on p4dn.24xlarge machine with 8 A100s, 40GB each
Issue -
State: open - Opened by sverneka almost 2 years ago
- 5 comments
#12 - FLAN-T5 XXL using DeepSpeed fits well for training but gives OOM error for inference.
Issue -
State: open - Opened by irshadbhat almost 2 years ago
- 2 comments
#11 - Sample inference script for FLAN-T5 XXL using DeepSpeed & Hugging Face.
Issue -
State: closed - Opened by irshadbhat almost 2 years ago
- 7 comments
#10 - Error when finetuning Flan-T5-XXL on custom dataset
Issue -
State: open - Opened by ngun7 almost 2 years ago
- 1 comment
#9 - Peft flan
Pull Request -
State: closed - Opened by philschmid almost 2 years ago
#8 - fix small bugs of deepseed-flan-t5-summarization.ipynb
Pull Request -
State: closed - Opened by yao-matrix almost 2 years ago
- 2 comments
#7 - Error (return code -7) when finetuning FLANT5-xxl on 8* A100
Issue -
State: open - Opened by scofield7419 almost 2 years ago
- 3 comments
#6 - OOM when finetuning FLANT5-xxl
Issue -
State: closed - Opened by AndrewZhe almost 2 years ago
- 4 comments
#5 - Deepspeed example
Pull Request -
State: closed - Opened by philschmid about 2 years ago
#4 - Chat Inference Code
Issue -
State: closed - Opened by samarthsarin about 2 years ago
- 2 comments
#3 - compute_metrics() function
Issue -
State: open - Opened by ybagoury about 2 years ago
- 1 comment
#2 - Problem with preprocess_function() in tutorial
Issue -
State: closed - Opened by ybagoury about 2 years ago
- 4 comments
#1 - Add tokenizer to Trainer object
Pull Request -
State: open - Opened by mrm8488 about 2 years ago
- 1 comment