Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
W
winosystolic
Manage
Activity
Members
Labels
Plan
Issues
Issue boards
Milestones
Wiki
Code
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Snippets
Deploy
Releases
Model registry
Monitor
Incidents
Analyze
Value stream analytics
Contributor analytics
Repository analytics
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
Community forum
Contribute to GitLab
Provide feedback
Keyboard shortcuts
?
Snippets
Groups
Projects
Show more breadcrumbs
xliu79
winosystolic
Graph
7ff2d959b220c41bc6db80a57677bfc5fb0aed70
Select Git revision
Branches
1
master
default
protected
1 result
You can move around the graph by using the arrow keys.
Begin with the selected commit
Created with Raphaël 2.2.0
18
Dec
9
17
Aug
22
May
20
14
12
9
28
Apr
9
19
Mar
17
28
Feb
19
13
6
Jan
3
28
Dec
27
26
23
22
18
16
12
10
9
21
Nov
20
15
4
1
8
Oct
7
shrink size to 2 2 2
master
master
somthing
a new version
seperate merge kernel size and kernel size
testing for 1x1
merge weight done
merge weight and strided output
fixed merged kernel input
validate C merge kernel. validate elementwise
add latency model base functions
validate the 8bit p[recision
change input load mode
a temporary work in float model
integration of hardware function to sequence
a back up before modification of adding the result
add float point validation layer
add float point
OK
automated script
finished the script to load format, write a c++ parser to load info
fix some new pragmas
fix some hls pragma that prevents burst read
adding sds pragma
add wino4x4, fixed a bug related to address bitwidth
fix warning problem
fix some macros for t16 winogtrad
add vivado_hls_script
verified pragmas
header file move for convinence of HLS test
set debug conv desc macro
output write back validated
output write back module complete, still need to partition the parameter
validate previous commit with different configurations
validated the computation of wino cell but need further implementation of output write back or add back
mid check point of changing input load and buffering part
weight part done and validated
correct input buffer and weight buffer load, but need to move parameters to conv_desc
flattened input feeding part but no transformation or stream out
fixed a bug in 5x5 model
new version, update input/output structure, update input reading module and validated
Loading