-
Notifications
You must be signed in to change notification settings - Fork 0
/
package.spiproj
183 lines (162 loc) · 3.52 KB
/
package.spiproj
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
// Spiral's ML Library
// 8/24/2024:
// Has the new core libs, the RPS, the Leduc and HU NL Holdem games running on the GPU, and currently tabular CFR training is being worked on.
// The games have a web based UI frontend, and a Flask based backend. If you are reading this in VS Code, press Ctrl + Shift + B to run it.
// Most likely you'll have to run `npm install` and `poetry install` to install the necessary dependencies first.
// Notes:
// The default int and float for this project should be: i32 and f32.
// It hasn't been tested with any other default types.
// TODO (9/13/2024): After finishing Cutlass integration, look into adding noise to the average policy. Some of the weird behavior like
// the trained Leduc agent folding King preflop might not be simply due to model bias, but inefficient exploration as well.
// packages: |core
modules:
corebase/
default-
base-
limit
real_core*
type_arith*
core-
compile_time/
types-
hashsetm
hashmapm
typed/
hashsetm
hashmapm
struct*
listm
loop
optionm
dep_optionm
monad
arraym
sam
dam
sa_listm
da_listm
tuple_listm
tensorm/
utils*
tensor_main-
partitionm
serializerm/
utils*
seriazer_main-
pickler
types-
corepython/
base-
plistm
jsonm
corecuda/
base-
refm
rangem
intrinsic
random
sortm
tensor_cuda
cupy
thread_scope
pipeline
wmma
pragma
cooperative_groups
semaphore
latchm
transposing_loop
coreext/
console/
types-
utils
console_cuda
console_python
console-
tensor
testing
coretests/
serializer
pickler
ml/
helpers-
primitives
matmul/
kernel
run-
layers
models
path_probs
cfr
cfr_models
cem
cem_models
game/
utils
rps/
main
leduc/
base-
cardm
deckm
game
model_cfr
model_cem
types-
play_cfr
train_cfr
play_cem
train_cem
full
tests/
train
nl_hu_holdem/
lib/
types-
cardm
deckm
hand_rankerm
game
types-
utils-
model_cfr
play_cfr
train_cfr
model_cem
play_cem
train_cem
full
tests/
test1
test2
test3
test4
test5
test6
test7
test8
test9
test10
test11
test12
test13
test14
test15
test16
test17
test18
test19
async_loads/
config-
test0
test1
test2
test3
test4
matmul/
test1_kernel
test1_run
test2_kernel
test2_run
test3_kernel
test3_run