Comments (4)
@volcacius, PTAL at this issue.
CC: @jinchen62
from brevitas.
Which Pytorch version are you on? With 2.1.1 I see:
Torch Fx graph:
graph():
%arg0_1 : [num_users=1] = placeholder[target=arg0_1]
%arg1_1 : [num_users=1] = placeholder[target=arg1_1]
%arg2_1 : [num_users=1] = placeholder[target=arg2_1]
%_scaled_dot_product_flash_attention : [num_users=9] = call_function[target=torch.ops.aten._scaled_dot_product_flash_attention.default](args = (%arg0_1, %arg1_1, %arg2_1, 0.0, True), kwargs = {})
%getitem : [num_users=1] = call_function[target=operator.getitem](args = (%_scaled_dot_product_flash_attention, 0), kwargs = {})
%getitem_1 : [num_users=0] = call_function[target=operator.getitem](args = (%_scaled_dot_product_flash_attention, 1), kwargs = {})
%getitem_2 : [num_users=0] = call_function[target=operator.getitem](args = (%_scaled_dot_product_flash_attention, 2), kwargs = {})
%getitem_3 : [num_users=0] = call_function[target=operator.getitem](args = (%_scaled_dot_product_flash_attention, 3), kwargs = {})
%getitem_4 : [num_users=0] = call_function[target=operator.getitem](args = (%_scaled_dot_product_flash_attention, 4), kwargs = {})
%getitem_5 : [num_users=0] = call_function[target=operator.getitem](args = (%_scaled_dot_product_flash_attention, 5), kwargs = {})
%getitem_6 : [num_users=0] = call_function[target=operator.getitem](args = (%_scaled_dot_product_flash_attention, 6), kwargs = {})
%getitem_7 : [num_users=0] = call_function[target=operator.getitem](args = (%_scaled_dot_product_flash_attention, 7), kwargs = {})
%getitem_8 : [num_users=0] = call_function[target=operator.getitem](args = (%_scaled_dot_product_flash_attention, 8), kwargs = {})
return getitem
Brevitas Fx graph:
graph():
%arg0_1 : [#users=1] = placeholder[target=arg0_1]
%arg1_1 : [#users=1] = placeholder[target=arg1_1]
%arg2_1 : [#users=1] = placeholder[target=arg2_1]
%_scaled_dot_product_flash_attention_default : [#users=9] = call_function[target=torch.ops.aten._scaled_dot_product_flash_attention.default](args = (%arg0_1, %arg1_1, %arg2_1, 0.0, True), kwargs = {})
%getitem : [#users=1] = call_function[target=operator.getitem](args = (%_scaled_dot_product_flash_attention_default, 0), kwargs = {})
%getitem_1 : [#users=0] = call_function[target=operator.getitem](args = (%_scaled_dot_product_flash_attention_default, 1), kwargs = {})
%getitem_2 : [#users=0] = call_function[target=operator.getitem](args = (%_scaled_dot_product_flash_attention_default, 2), kwargs = {})
%getitem_3 : [#users=0] = call_function[target=operator.getitem](args = (%_scaled_dot_product_flash_attention_default, 3), kwargs = {})
%getitem_4 : [#users=0] = call_function[target=operator.getitem](args = (%_scaled_dot_product_flash_attention_default, 4), kwargs = {})
%getitem_5 : [#users=0] = call_function[target=operator.getitem](args = (%_scaled_dot_product_flash_attention_default, 5), kwargs = {})
%getitem_6 : [#users=0] = call_function[target=operator.getitem](args = (%_scaled_dot_product_flash_attention_default, 6), kwargs = {})
%getitem_7 : [#users=0] = call_function[target=operator.getitem](args = (%_scaled_dot_product_flash_attention_default, 7), kwargs = {})
%getitem_8 : [#users=0] = call_function[target=operator.getitem](args = (%_scaled_dot_product_flash_attention_default, 8), kwargs = {})
return getitem
from brevitas.
Which Pytorch version are you on? With 2.1.1 I see:
torch 2.2.0.dev20231115+cpu
from brevitas.
It should be fixed in this PR #763 I just merged in dev. Thanks for spotting it.
from brevitas.
Related Issues (20)
- Implement context-manager based export
- Missing Proxy tests
- Export ONNX QOperator HOT 5
- Fix Value Tracer
- Activation Equalization co-optimize flag
- Update entrypoint for LLM
- Add squeeze / unsqueeze operations to quant invariant functions in `torch_handler.py` HOT 4
- Add support for minifloat ptq with fx backend on residual models
- Implement `torch.where` STE for minifloat clamping
- Remove maximum assumptions about NaN/inf values for minifloat configurations
- Change way of setting `NaN` and `inf` values for custom minifloat formats
- Update signature check
- Deprecate use of MacOS (Darwin) runners in CI
- Adding tests for "quantize" function for CNN PTQ HOT 7
- Call for better/more documentation
- Per-channel zero points but per-tensor scales HOT 6
- Documentation setup thoughts HOT 3
- update dependencies=2.0.1 requirement HOT 4
- Mac OSX Tests for `torch==1.9.1` fail when installing dependencies HOT 3
- Weights not quantized after using qnn.QuantConv2d layers for QAT HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from brevitas.