dvlab-research / LISA

Project Page for "LISA: Reasoning Segmentation via Large Language Model"

dvlab-research/LISA Issues

IndexError: The shape of the mask [1, 304] at index 1 does not match the shape of the indexed tensor [1, 624, 256] at index 1
Updated 18 days ago7
No output segmentation image when running chat.py
Updated 24 days ago1
LISA model_forward bug, only occurs when inference.
Updated 24 days ago1
AttributeError: 'NoneType' object has no attribute 'shape'
Updated 24 days ago2
Code error about scheduler in train_ds.py
Updated a month ago
AttributeError: 'LlamaAttention' object has no attribute 'rope_theta'
Updated a month ago1
How many samples were iterated in total?
Updated a month ago
Why apply the causal mask to image tokens in the attention operations of LLM?
Updated a month ago
How to generate ReasonSeg Dataset
Updated 2 months ago
KeyError: 'LlavaConfig'
Updated 2 months ago3
Segmentation Performance Evaluation Result is significantly different than paper gIoU:60.0 cIoU:67.8 vs gIoU:0.0759 cIoU:0.0648
Updated 2 months ago3
Token indices sequence length is longer than the specified maximum sequence length for this model (565 > 512). Running this sequence through the model will result in indexing errors
Updated 2 months ago
Is LISA model on Hugginface?
Updated 2 months ago
Which file is the lora weight?
Updated 2 months ago3
Inference speed on refcocog dataset
Updated 2 months ago
Comparison with GroundedDINO
Updated 2 months ago
val.json missing
Closed 2 months ago
Token indices sequence length is longer than the specified maximum sequence length for this model (XXX > 512). Running this sequence through the model will result in indexing errors
Updated 3 months ago1
How do I get the results in Table 1 ?
Updated 3 months ago
question about text_hidden_fcs
Updated 3 months ago
gt_mask shape is (0, x, y) while pred_mask shape is (some number, x, y), which raise assertion error for vqa and reason seg datasets
Updated 3 months ago
unexpected_keys= base_model.model.model.layers.0.mlp.gate_proj.lora_A.default.weight'
Updated 4 months ago
`xinlai/LISA-13B-llama2-v1-explanatory` model regression or incorrect user setting?
Updated 4 months ago
Plan to open source LISA++
Updated 5 months ago
What's the diffenence between evaluate(...) and model_forward(..., inference=True) in LISA.py?
Updated 5 months ago
Lisa++ dataset release
Updated 5 months ago
'llava' is already used by a Transformers config, pick another name.
Updated 5 months ago2
online demo
Updated 5 months ago1
cuda out of memory
Updated 6 months ago
Issues with downloading ReasonSeg Dataset : (Train , Val zip files)
Closed 6 months ago1
I found train and val dataset in the link you shared but there is no test dataset. Could you please share it?
Updated 6 months ago
Only train my dataset with huggingface checkpoint repo LISA-13B-llama2-v1
Updated 6 months ago
distributed training error
Updated 6 months ago1
Question about the indexing of the hidden state
Updated 6 months ago
Are all of your 8 NVIDIA 3090 on a single node?
Updated 6 months ago
pytorch_model.bin
Closed 6 months ago
about the batch_size and training speed
Updated 6 months ago
Why use_mm_start_end=True, does it make any difference if it is set to False to train from scratch?
Updated 6 months ago
Facing issue with Quantization.
Closed 6 months ago1
Can not access Online Demo
Closed 6 months ago3
LISA-7B和13B的训练代价
Updated 6 months ago
Batchsize and forward time
Updated 6 months ago
pytorch_model.bin
Updated 6 months ago
Question about [start_token]
Closed 6 months ago1
how to merge lora weight when there are more than 1 model.bin
Updated 7 months ago1
AttributeError: 'LISAForCausalLM' object has no attribute 'ce_loss_weight'
Updated 7 months ago3
default_conversation = conv_vicuna_v0 ？？？
Updated 7 months ago
Validate on my own trained pytorch.bin
Closed 7 months ago2
A issue about the random seed set
Updated 7 months ago
How to get the 4bit model after training is completed
Closed 7 months ago1