Aritra Dutta issues

Results 7 issues of


                                            Aritra Dutta

How to calculate and retrieve the best matching document for a query from a list of document?

I have been trying to fetch the best matched document given a query. But the code given in the examples seems to work only pairwise, given a query and document...

How to get rid of the "Duplicate GPU detected : rank 0 and rank 1 both on CUDA device ca000" error while training of the ColBERTv1.9 modell?

I am trying to finetune the ColBERT v1.9 on my specific dataset for retrieval, but unable to do so. I encountered the below error:- torch.distributed.DistBackendError: NCCL error in: /opt/conda/conda-bld/pytorch_1702400431970/work/torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:1333, invalid...

[BUG] <title> When using MiniCPM as the language Model, getting "data fetch error", bug in dataset.py file.

### 是否已有关于该错误的issue或讨论？ | Is there an existing issue / discussion for this? - [X] 我已经搜索过已有的issues和讨论 | I have searched the existing issues / discussions ### 该问题是否在FAQ中有解答？ | Is there an...

How to get raw_logits from the Moondream model?

We have the method "model.batch_answer( batch )", but what if I just want to get the logits before the decoding part. Something like using only the forward method "logits =...

I trained and loaded the xlora model as shown in the tutorial, but I couldn't get the trained xlora model loaded, getting "AttributeError: 'NoneType' object has no attribute 'items'" error.

I am using a Quantised Qwen VL 2B model, I loaded that as a base model then passed the same to initialise a trained xlora model as follows: ``` base_model...

The vision encoder doesn't return hidden states causing the model.generate() method to fail. How to fix it?

If we check the stack trace, we would see that the forward pass of the vision mode,l which is basically a SigLip Encoder, is failing, causing the ultimate model.generate() fail....