pytorch-labs/segment-anything-fast issues and pull requests

#100 - Fix load local config for non-A100 setup

Pull Request - State: closed - Opened by cpuhrsch 11 months ago
Labels: CLA Signed

#100 - Fix load local config for non-A100 setup

Pull Request - State: closed - Opened by cpuhrsch 11 months ago
Labels: CLA Signed

#99 - ValueError: Expected query, key, and value to all be be jagged at dimension 2, but got query._ragged_idx: 1, key._ragged_idx: 1 and value._ragged_idx: 1 instead.

Issue - State: closed - Opened by sanshe316 11 months ago - 4 comments

#99 - ValueError: Expected query, key, and value to all be be jagged at dimension 2, but got query._ragged_idx: 1, key._ragged_idx: 1 and value._ragged_idx: 1 instead.

Issue - State: closed - Opened by sanshe316 11 months ago - 4 comments

#98 - error for run mask_generator.generate(image) on nvidia 3090 and A10

Issue - State: open - Opened by qingfengmingyue 11 months ago - 3 comments

#98 - error for run mask_generator.generate(image) on nvidia 3090 and A10

Issue - State: open - Opened by qingfengmingyue 11 months ago - 3 comments

#97 - RTX 4090: RuntimeError: CUDA error: an illegal memory access was encountered

Issue - State: closed - Opened by Scass0807 11 months ago - 1 comment

#97 - RTX 4090: RuntimeError: CUDA error: an illegal memory access was encountered

Issue - State: closed - Opened by Scass0807 11 months ago - 1 comment

#96 - Fix torch bloat16 -> numpy float32 conversion for compile max-autotune

Pull Request - State: closed - Opened by mawanda-jun 11 months ago - 3 comments
Labels: CLA Signed

#96 - Fix torch bloat16 -> numpy float32 conversion for compile max-autotune

Pull Request - State: closed - Opened by mawanda-jun 11 months ago - 3 comments
Labels: CLA Signed

#95 - use sam_model_fast_registry on NVIDIA GeForce RTX 4090 ，get error

Issue - State: open - Opened by Aatroy 11 months ago - 3 comments

#95 - use sam_model_fast_registry on NVIDIA GeForce RTX 4090 ，get error

Issue - State: open - Opened by Aatroy 11 months ago - 3 comments

#94 - Running on CPU.

Issue - State: open - Opened by shersoni610 12 months ago - 1 comment

#94 - Running on CPU.

Issue - State: open - Opened by shersoni610 12 months ago - 1 comment

#93 - Min Memory requirement?

Issue - State: open - Opened by shersoni610 12 months ago - 8 comments

#93 - Min Memory requirement?

Issue - State: open - Opened by shersoni610 12 months ago - 8 comments

#92 - Update and Organize documentation

Pull Request - State: open - Opened by subashbasnyat 12 months ago
Labels: CLA Signed

#92 - Update and Organize documentation

Pull Request - State: open - Opened by subashbasnyat 12 months ago
Labels: CLA Signed

#91 - Enhanced and Updated Documentation for Experiment Setup and Execution Process

Pull Request - State: closed - Opened by subashbasnyat 12 months ago - 1 comment

#91 - Enhanced and Updated Documentation for Experiment Setup and Execution Process

Pull Request - State: closed - Opened by subashbasnyat 12 months ago - 1 comment

#90 - Question of behavior of storing custom configuration of autotune function

Issue - State: closed - Opened by SangbumChoi 12 months ago - 3 comments

#90 - Question of behavior of storing custom configuration of autotune function

Issue - State: closed - Opened by SangbumChoi 12 months ago - 3 comments

#89 - Update flash_4.py

Pull Request - State: open - Opened by SangbumChoi 12 months ago - 2 comments
Labels: CLA Signed

#89 - Update flash_4.py

Pull Request - State: open - Opened by SangbumChoi 12 months ago - 2 comments
Labels: CLA Signed

#88 - error for int8 inference on Nvidia 3090

Issue - State: open - Opened by rbavery 12 months ago - 7 comments

#88 - error for int8 inference on Nvidia 3090

Issue - State: open - Opened by rbavery 12 months ago - 7 comments

#87 - ValueError: ('Unsupported kind: ', 'FRAGMENT')

Issue - State: open - Opened by 0verlook 12 months ago - 2 comments

#87 - ValueError: ('Unsupported kind: ', 'FRAGMENT')

Issue - State: open - Opened by 0verlook 12 months ago - 2 comments

#86 - Inference on A10 GPU become slower than original SAM

Issue - State: closed - Opened by GovernorHo 12 months ago - 1 comment

#86 - Inference on A10 GPU become slower than original SAM

Issue - State: closed - Opened by GovernorHo 12 months ago - 1 comment

#85 - error with flash attention

Issue - State: open - Opened by staple-pi 12 months ago - 9 comments

#85 - error with flash attention

Issue - State: open - Opened by staple-pi 12 months ago - 9 comments

#84 - Do I have to use the latest version of PyTorch?

Issue - State: open - Opened by SSongkang 12 months ago - 5 comments

#84 - Do I have to use the latest version of PyTorch?

Issue - State: open - Opened by SSongkang 12 months ago - 5 comments

#83 - 一定要使用最新的pytorch吗？

Issue - State: closed - Opened by SSongkang 12 months ago

#83 - 一定要使用最新的pytorch吗？

Issue - State: closed - Opened by SSongkang 12 months ago

#82 - can't use it in gtx 4080

Issue - State: closed - Opened by staple-pi 12 months ago

#82 - can't use it in gtx 4080

Issue - State: closed - Opened by staple-pi 12 months ago

#81 - Compatible with torch.export?

Issue - State: closed - Opened by dbbert 12 months ago - 2 comments

#81 - Compatible with torch.export?

Issue - State: closed - Opened by dbbert 12 months ago - 2 comments

#80 - sam_model_fast_registry ， Speed problem

Issue - State: closed - Opened by KnSun99 12 months ago - 8 comments

#80 - sam_model_fast_registry ， Speed problem

Issue - State: closed - Opened by KnSun99 12 months ago - 8 comments

#79 - Labeling tool ISAT have supported segment-anything-fast.

Issue - State: closed - Opened by yatengLG 12 months ago - 2 comments

#79 - Labeling tool ISAT have supported segment-anything-fast.

Issue - State: closed - Opened by yatengLG 12 months ago - 2 comments

#78 - How to train SAM using this repo? Can I get similar gain for training?

Issue - State: closed - Opened by yxchng 12 months ago - 2 comments

#78 - How to train SAM using this repo? Can I get similar gain for training?

Issue - State: closed - Opened by yxchng 12 months ago - 2 comments

#77 - Update README.md

Pull Request - State: open - Opened by eltociear 12 months ago
Labels: CLA Signed

#77 - Update README.md

Pull Request - State: open - Opened by eltociear 12 months ago
Labels: CLA Signed

#76 - Worse performance on SamAutomaticMaskGenerator

Issue - State: open - Opened by BrainWWW 12 months ago - 6 comments

#76 - Worse performance on SamAutomaticMaskGenerator

Issue - State: open - Opened by BrainWWW 12 months ago - 6 comments

#75 - Add comments to summary_chart.py

Pull Request - State: open - Opened by ColeCurrent 12 months ago - 2 comments
Labels: CLA Signed

#75 - Add comments to summary_chart.py

Pull Request - State: open - Opened by ColeCurrent 12 months ago - 2 comments
Labels: CLA Signed

#74 - Analysis pytorch profiler

Issue - State: closed - Opened by zhanglei1172 12 months ago - 3 comments

#74 - Analysis pytorch profiler

Issue - State: closed - Opened by zhanglei1172 12 months ago - 3 comments

#73 - Create new BEST_CONFIGS when loading wasn't possible

Pull Request - State: closed - Opened by cpuhrsch 12 months ago
Labels: CLA Signed

#73 - Create new BEST_CONFIGS when loading wasn't possible

Pull Request - State: closed - Opened by cpuhrsch 12 months ago
Labels: CLA Signed

#72 - Performance on GTX 4070

Issue - State: open - Opened by shersoni610 12 months ago - 2 comments

#72 - Performance on GTX 4070

Issue - State: open - Opened by shersoni610 12 months ago - 2 comments

#71 - fix triton attention bug

Pull Request - State: closed - Opened by zhanglei1172 12 months ago - 3 comments
Labels: CLA Signed

#71 - fix triton attention bug

Pull Request - State: closed - Opened by zhanglei1172 12 months ago - 3 comments
Labels: CLA Signed

#70 - Update README.md

Pull Request - State: closed - Opened by cpuhrsch 12 months ago
Labels: CLA Signed

#70 - Update README.md

Pull Request - State: closed - Opened by cpuhrsch 12 months ago
Labels: CLA Signed

#69 - Faster AMG

Pull Request - State: closed - Opened by cpuhrsch 12 months ago
Labels: CLA Signed

#69 - Faster AMG

Pull Request - State: closed - Opened by cpuhrsch 12 months ago
Labels: CLA Signed

#68 - support for aot compilation with torch.export?

Issue - State: closed - Opened by rbavery 12 months ago - 3 comments

#68 - support for aot compilation with torch.export?

Issue - State: closed - Opened by rbavery 12 months ago - 3 comments

#67 - Autotuning not on A100, instructions, single line API for model optimizations.

Pull Request - State: closed - Opened by cpuhrsch 12 months ago
Labels: CLA Signed

#67 - Autotuning not on A100, instructions, single line API for model optimizations.

Pull Request - State: closed - Opened by cpuhrsch 12 months ago
Labels: CLA Signed

#66 - _attention_rel_h_rel_w_kernel_aligned_device not receiving correct dtype

Issue - State: closed - Opened by geodavic 12 months ago - 11 comments

#66 - _attention_rel_h_rel_w_kernel_aligned_device not receiving correct dtype

Issue - State: closed - Opened by geodavic 12 months ago - 11 comments

#65 - Use torchao for quantization

Pull Request - State: closed - Opened by cpuhrsch almost 1 year ago - 2 comments
Labels: CLA Signed

#65 - Use torchao for quantization

Pull Request - State: closed - Opened by cpuhrsch almost 1 year ago - 2 comments
Labels: CLA Signed

#64 - Update README.md

Pull Request - State: closed - Opened by cpuhrsch about 1 year ago
Labels: CLA Signed

#64 - Update README.md

Pull Request - State: closed - Opened by cpuhrsch about 1 year ago
Labels: CLA Signed

#63 - Shared memory out of resource.Need 135200M memory！

Issue - State: closed - Opened by DESEOUMAIGA about 1 year ago - 14 comments

#63 - Shared memory out of resource.Need 135200M memory！

Issue - State: closed - Opened by DESEOUMAIGA about 1 year ago - 14 comments

#62 - Update results.csv using p4d results

Pull Request - State: closed - Opened by cpuhrsch about 1 year ago
Labels: CLA Signed

#62 - Update results.csv using p4d results

Pull Request - State: closed - Opened by cpuhrsch about 1 year ago
Labels: CLA Signed

#61 - Updated p4d results using new int_mm kernel

Pull Request - State: closed - Opened by cpuhrsch about 1 year ago
Labels: CLA Signed