Skip to content

Minor fixes in ConvNets #748

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

Merged
nv-kkudrynski merged 1 commit into from
Nov 9, 2020
Merged

Minor fixes in ConvNets #748

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
1 change: 0 additions & 1 deletion PyTorch/Classification/ConvNets/triton/deployer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -61,7 +61,6 @@ def initialize_model(args):
model.load_state_dict(
{k.replace("module.", ""): v for k, v in state_dict.items()}
)
model.load_state_dict(state_dict)
return model.half() if args.fp16 else model


Expand Down
45 changes: 0 additions & 45 deletions TensorFlow/Classification/ConvNets/utils/cmdline_helper.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -173,51 +173,6 @@ def parse_cmdline(available_arch):
help="Quantize weights and activations during training using symmetric quantization."
)

p.add_argument(
'--finetune_checkpoint',
required=False,
default=None,
type=str,
help="Path to pre-trained checkpoint which will be used for fine-tuning"
)

_add_bool_argument(
parser=p, name="use_final_conv", default=False, required=False, help="Use cosine learning rate schedule."
)

p.add_argument(
'--quant_delay',
type=int,
default=0,
required=False,
help="Number of steps to be run before quantization starts to happen"
)

_add_bool_argument(
parser=p,
name="quantize",
default=False,
required=False,
help="Quantize weights and activations during training. (Defaults to Assymmetric quantization)"
)

_add_bool_argument(
parser=p,
name="use_qdq",
default=False,
required=False,
help="Use QDQV3 op instead of FakeQuantWithMinMaxVars op for quantization. QDQv3 does only scaling"
)

_add_bool_argument(
parser=p,
name="symmetric",
default=False,
required=False,
help="Quantize weights and activations during training using symmetric quantization."
)


p.add_argument(
'--log_filename',
type=str,
Expand Down