mirrors
/
Segment-Everything-Everywhere-All-At-Once
mirror of https://github.com/UX-Decoder/Segment-Everything-Everywhere-All-At-Once.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Update EVAL.md

pull/139/head
Xueyan Zou 2023-10-04 22:34:54 -05:00 committed by GitHub
parent 5cdd628908
commit 8465e5f606
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 136 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

136
assets/readmes/EVAL.md
View File

@ -0,0 +1,136 @@
## X-Decoder
**Focal-T:**
```
CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 mpirun -n 8 python entry.py evaluate \
--conf_files configs/xdecoder/focalt_unicl_lang.yaml \
--overrides \
COCO.INPUT.IMAGE_SIZE 1024 \
MODEL.DECODER.CAPTIONING.ENABLED True \
MODEL.DECODER.RETRIEVAL.ENABLED True \
MODEL.DECODER.GROUNDING.ENABLED True \
COCO.TEST.BATCH_SIZE_TOTAL 8 \
COCO.TRAIN.BATCH_SIZE_TOTAL 8 \
COCO.TRAIN.BATCH_SIZE_PER_GPU 1 \
VLP.TEST.BATCH_SIZE_TOTAL 8 \
VLP.TRAIN.BATCH_SIZE_TOTAL 8 \
VLP.TRAIN.BATCH_SIZE_PER_GPU 1 \
MODEL.DECODER.HIDDEN_DIM 512 \
MODEL.ENCODER.CONVS_DIM 512 \
MODEL.ENCODER.MASK_DIM 512 \
ADE20K.TEST.BATCH_SIZE_TOTAL 8 \
FP16 True \
WEIGHT True \
RESUME_FROM /pth/to/xdecoder_data/xdecoder/xdecoder_focalt_last.pt
```
**Focal-L:**
```
CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 mpirun -n 8 python entry.py evaluate \
--conf_files configs/xdecoder/focall_unicl_lang.yaml \
--overrides \
COCO.INPUT.IMAGE_SIZE 1024 \
MODEL.DECODER.CAPTIONING.ENABLED True \
MODEL.DECODER.RETRIEVAL.ENABLED True \
MODEL.DECODER.GROUNDING.ENABLED True \
COCO.TEST.BATCH_SIZE_TOTAL 8 \
COCO.TRAIN.BATCH_SIZE_TOTAL 8 \
COCO.TRAIN.BATCH_SIZE_PER_GPU 1 \
VLP.TEST.BATCH_SIZE_TOTAL 8 \
VLP.TRAIN.BATCH_SIZE_TOTAL 8 \
VLP.TRAIN.BATCH_SIZE_PER_GPU 1 \
MODEL.DECODER.HIDDEN_DIM 512 \
MODEL.ENCODER.CONVS_DIM 512 \
MODEL.ENCODER.MASK_DIM 512 \
ADE20K.TEST.BATCH_SIZE_TOTAL 8 \
FP16 True \
WEIGHT True \
RESUME_FROM /pth/to/xdecoder_data/xdecoder/xdecoder_focall_last.pt
```
## SEEM
**Focal-T v0:**
```sh
CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 mpirun -n 8 python entry.py evaluate \
--conf_files configs/seem/focalt_unicl_lang_v0.yaml \
--overrides \
COCO.INPUT.IMAGE_SIZE 1024 \
MODEL.DECODER.HIDDEN_DIM 512 \
MODEL.ENCODER.CONVS_DIM 512 \
MODEL.ENCODER.MASK_DIM 512 \
VOC.TEST.BATCH_SIZE_TOTAL 8 \
TEST.BATCH_SIZE_TOTAL 8 \
REF.TEST.BATCH_SIZE_TOTAL 8 \
FP16 True \
WEIGHT True \
RESUME_FROM /pth/to/xdecoder_data/seem/seem_focalt_v0.pt
```
**Focal-T v1:**
```sh
CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 mpirun -n 8 python entry.py evaluate \
--conf_files configs/seem/focalt_unicl_lang_v1.yaml \
--overrides \
COCO.INPUT.IMAGE_SIZE 1024 \
MODEL.DECODER.HIDDEN_DIM 512 \
MODEL.ENCODER.CONVS_DIM 512 \
MODEL.ENCODER.MASK_DIM 512 \
VOC.TEST.BATCH_SIZE_TOTAL 8 \
TEST.BATCH_SIZE_TOTAL 8 \
REF.TEST.BATCH_SIZE_TOTAL 8 \
FP16 True \
WEIGHT True \
RESUME_FROM /pth/to/xdecoder_data/seem/seem_focalt_v1.pt
```
**ViT-B SAM v1:**
```sh
CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 mpirun -n 8 python entry.py evaluate \
--conf_files configs/seem/samvitb_unicl_lang_v1.yaml \
--overrides \
COCO.INPUT.IMAGE_SIZE 1024 \
MODEL.DECODER.HIDDEN_DIM 512 \
MODEL.ENCODER.CONVS_DIM 512 \
MODEL.ENCODER.MASK_DIM 512 \
VOC.TEST.BATCH_SIZE_TOTAL 8 \
TEST.BATCH_SIZE_TOTAL 8 \
REF.TEST.BATCH_SIZE_TOTAL 8 \
FP16 True \
WEIGHT True \
RESUME_FROM /pth/to/xdecoder_data/seem/seem_samvitb_v1.pt
```
**ViT-L SAM v1:**
```sh
CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 mpirun -n 8 python entry.py evaluate \
--conf_files configs/seem/samvitl_unicl_lang_v1.yaml \
--overrides \
COCO.INPUT.IMAGE_SIZE 1024 \
MODEL.DECODER.HIDDEN_DIM 512 \
MODEL.ENCODER.CONVS_DIM 512 \
MODEL.ENCODER.MASK_DIM 512 \
VOC.TEST.BATCH_SIZE_TOTAL 8 \
TEST.BATCH_SIZE_TOTAL 8 \
REF.TEST.BATCH_SIZE_TOTAL 8 \
FP16 True \
WEIGHT True \
RESUME_FROM /pth/to/xdecoder_data/seem/seem_samvitl_v1.pt
```
**Focal-L v0:**
```sh
CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 mpirun -n 8 python entry.py evaluate \
--conf_files configs/seem/focall_unicl_lang_v0.yaml \
--overrides \
COCO.INPUT.IMAGE_SIZE 1024 \
MODEL.DECODER.HIDDEN_DIM 512 \
MODEL.ENCODER.CONVS_DIM 512 \
MODEL.ENCODER.MASK_DIM 512 \
VOC.TEST.BATCH_SIZE_TOTAL 8 \
TEST.BATCH_SIZE_TOTAL 8 \
REF.TEST.BATCH_SIZE_TOTAL 8 \
FP16 True \
WEIGHT True \
RESUME_FROM /pth/to/xdecoder_data/seem/seem_focall_v0.pt
```