Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Re-enable detectron2 model accuracy checks. #2317

Closed
wants to merge 2 commits into from
Closed

Re-enable detectron2 model accuracy checks. #2317

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
c30d44d
Re-enable detectron2 model accuracy checks.
xuzhao9 Jun 19, 2024
794d12b
Fix detectron2 maskrcnn
xuzhao9 Jun 19, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 0 additions & 2 deletions torchbenchmark/models/detectron2_fasterrcnn_r_101_c4/metadata.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,5 +6,3 @@ eval_deterministic: false
eval_nograd: true
train_benchmark: false
train_deterministic: false
not_implemented:
- test: example
1 change: 0 additions & 1 deletion torchbenchmark/models/detectron2_fasterrcnn_r_101_dc5/metadata.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,5 @@ eval_deterministic: false
eval_nograd: true
not_implemented:
- device: cpu
- test: example
train_benchmark: false
train_deterministic: false
1 change: 0 additions & 1 deletion torchbenchmark/models/detectron2_fasterrcnn_r_101_fpn/metadata.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,5 @@ eval_deterministic: false
eval_nograd: true
not_implemented:
- device: cpu
- test: example
train_benchmark: false
train_deterministic: false
1 change: 0 additions & 1 deletion torchbenchmark/models/detectron2_fasterrcnn_r_50_c4/metadata.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,5 @@ eval_deterministic: false
eval_nograd: true
not_implemented:
- device: cpu
- test: example
train_benchmark: false
train_deterministic: false
1 change: 0 additions & 1 deletion torchbenchmark/models/detectron2_fasterrcnn_r_50_dc5/metadata.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,5 @@ eval_deterministic: false
eval_nograd: true
not_implemented:
- device: cpu
- test: example
train_benchmark: false
train_deterministic: false
1 change: 0 additions & 1 deletion torchbenchmark/models/detectron2_fasterrcnn_r_50_fpn/metadata.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,5 @@ eval_deterministic: false
eval_nograd: true
not_implemented:
- device: cpu
- test: example
train_benchmark: false
train_deterministic: false
4 changes: 1 addition & 3 deletions torchbenchmark/models/detectron2_maskrcnn/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -58,9 +58,7 @@ class Model(BenchmarkModel):
model_file = os.path.join(MODEL_DIR, ".data", f"{MODEL_NAME}.pkl")
DEFAULT_TRAIN_BSIZE = 1
DEFAULT_EVAL_BSIZE = 1
# Skip correctness check, because the output tensor can't be verified using
# cosine similarity or torch.close()
SKIP_CORRECTNESS_CHECK = True
DISABLE_DETERMINISM = True

def __init__(self, test, device, batch_size=None, extra_args=[]):
super().__init__(
Expand Down
1 change: 0 additions & 1 deletion torchbenchmark/models/detectron2_maskrcnn/metadata.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,5 @@ eval_deterministic: false
eval_nograd: true
not_implemented:
- device: cpu
- test: example
train_benchmark: false
train_deterministic: false
1 change: 0 additions & 1 deletion torchbenchmark/models/detectron2_maskrcnn_r_101_c4/metadata.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,5 @@ eval_deterministic: false
eval_nograd: true
not_implemented:
- device: cpu
- test: example
train_benchmark: false
train_deterministic: false
1 change: 0 additions & 1 deletion torchbenchmark/models/detectron2_maskrcnn_r_101_fpn/metadata.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,5 @@ eval_deterministic: false
eval_nograd: true
not_implemented:
- device: cpu
- test: example
train_benchmark: false
train_deterministic: false
1 change: 0 additions & 1 deletion torchbenchmark/models/detectron2_maskrcnn_r_50_c4/metadata.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,5 @@ eval_deterministic: false
eval_nograd: true
not_implemented:
- device: cpu
- test: example
train_benchmark: false
train_deterministic: false
1 change: 0 additions & 1 deletion torchbenchmark/models/detectron2_maskrcnn_r_50_fpn/metadata.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,5 @@ eval_deterministic: false
eval_nograd: true
not_implemented:
- device: cpu
- test: example
train_benchmark: false
train_deterministic: false
3 changes: 0 additions & 3 deletions torchbenchmark/models/maml/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,9 +16,6 @@ class Model(BenchmarkModel):
DEFAULT_EVAL_BSIZE = 1
ALLOW_CUSTOMIZE_BSIZE = False
CANNOT_SET_CUSTOM_OPTIMIZER = True
# Skip correctness check, because maml runs backward and optimizer in eval()
# Which will return non-deterministic results
SKIP_CORRECTNESS_CHECK = True

def __init__(self, test, device, batch_size=None, extra_args=[]):
super().__init__(
Expand Down
4 changes: 1 addition & 3 deletions torchbenchmark/util/framework/detectron2/model_factory.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -90,9 +90,7 @@ class Detectron2Model(BenchmarkModel):
# Default batch sizes
DEFAULT_TRAIN_BSIZE = 1
DEFAULT_EVAL_BSIZE = 1
# Skip correctness check, because the output tensor can't be verified using
# cosine similarity or torch.close()
SKIP_CORRECTNESS_CHECK = True
DISABLE_DETERMINISM = True

def __init__(self, variant, test, device, batch_size=None, extra_args=[]):
super().__init__(
Expand Down
Loading