Shogo Inadomi

Thank you for your comment (Sorry for the late reply). I notice bricks.py add_brick_() is the code for checking some 3D constraints, but I'm not sure whether it is comprehensive...

- About my question : For example, the green block and red block is mistakenly swapped at step 17. The correct order is green ->red from left to right. If...

- Performance degradation for retraining : How about adding "--load_lpub \" to train_mepnet.sh ? When I deleted "--load_lpub \" from eval_mepnet.sh, the result became worse. I think changing image type...

I solved by myself. `WORLD_SIZE=1 CUDA_VISIBLE_DEVICES=0,1,2,3 python generate.py --load_8bit --base_model 'decapoda-research/llama-7b-hf' --lora_weights 'tloen/alpaca-lora-7b'` Also, as mentioned at https://github.com/tloen/alpaca-lora/issues/21#issuecomment-1473318920 adding device_map setting was necessary. But here I use multi-GPUs, so set...

 Above command was OK for 7B model, but for 13B and 30B model, output was strange when put with `--load_8bit` . When using a single GPU,...

Hi, I attached restore_semantic_last_layer_from_initial_checkpoint : false with a textproto file like model_options { initial_checkpoint: path-to-pretrained-model (for me it was max_deeplab_l_backbone_os16_axial_deeplab_cityscapes_trainfine/ckpt-60000) restore_semantic_last_layer_from_initial_checkpoint: false ... } then it worked for my own...