Closed BerLancelot closed 3 months ago
Enabling this flag may help you.
Many thanks! I'm using the previous version of Opengait with the Gaitgraph algorithm, but I find that the Rank-1 accuracy is only 5.5% when I'm reproducing it using the GREW dataset. Noticing that the original Gaitgraph paper was not tested using the GREW dataset, I wonder if the low accuracy was due to the difference in the dataset, or if the algorithm itself was not working well?
Here are my parameter settings:
data_cfg:
dataset_name: GREW
dataset_root: E:/PycharmProjects/datasets/processed_datasets/GREW/GREW-pose-pkl
dataset_partition: ./datasets/GREW/GREW.json
test_dataset_name: GREW
num_workers: 1
remove_no_gallery: false
frame_threshold: 16
evaluator_cfg:
enable_float16: false
restore_ckpt_strict: false
restore_hint: 150000
save_name: GaitGraph1_phase1
sampler:
batch_size: 256
frames_num_fixed: 501
frames_num_max: 50
frames_num_min: 25
sample_type: fixed_ordered
frames_skip_num: 0
metric: euc
# eval_func: GREW_submission
eval_func: evaluate_real_scene
transform:
- type: Compose
trf_cfg:
- type: SelectSequenceCenter
sequence_length: 16
- type: GaitGraph1Input
loss_cfg:
- loss_term_weight: 1
temperature: 0.01
type: SupConLoss_Re
log_prefix: SupConLoss
model_cfg:
model: GaitGraph1
joint_format: coco
input_num: 1
reduction: 8
block: Bottleneck # Basic, initial
input_branch:
- 3
- 64
- 64
- 32
main_stream:
- 32
- 128
- 128
- 256
- 256
num_class: 256
tta: true
optimizer_cfg:
lr: 0.01
solver: Adam
weight_decay: 0.00001
scheduler_cfg:
max_lr: 0.01
total_steps: 150000
scheduler: OneCycleLR
trainer_cfg:
enable_float16: false
log_iter: 100
with_test: true
restore_ckpt_strict: true
restore_hint: 0
save_iter: 5000
save_name: GaitGraph1_phase1
sync_BN: true
total_iter: 150000
sampler:
batch_shuffle: true
frames_num_fixed: 501
frames_num_max: 50
frames_num_min: 25
sample_type: fixed_ordered #Repeat sample
frames_skip_num: 0
batch_size: 128
type: CommonSampler
transform:
- type: TwoView
trf_cfg:
- type: MirrorPoses
probability: 0.5
- type: FlipSequence
probability: 0.5
- type: RandomSelectSequence
sequence_length: 16
- type: PointNoise
std: 0.05
- type: JointNoise
std: 0.1
- type: GaitGraph1Input
And the test results are as follows:
Stale issue message
I'm a beginner in gait recognition, and I'd like to ask how to see the accuracy of the results after gaitgraph training under the GREW dataset, at the moment I can only see the supervised comparison loss function. Thanks for the answer!
Currently during training the progress bar prompt will only appear something like "[2024-01-06 18:42:25] [INFO]: Iteration 150000, Cost 53.59s, SupConLoss_loss=0.0237" with no accuracy related information.