Skip to content
This repository was archived by the owner on Apr 28, 2023. It is now read-only.

Commit 2cd79b8

Browse files
nicolasvasilachenicolasvasilache
committed
Update C++ examples to follow TC coding guide
1 parent 5b2a75e commit 2cd79b8

File tree

4 files changed

+55
-55
lines changed

4 files changed

+55
-55
lines changed

‎benchmarks/MLP_model.cc‎

Lines changed: 43 additions & 43 deletions
Original file line numberDiff line numberDiff line change
@@ -64,23 +64,23 @@ DEFINE_uint32(Q, 2, "W4_h");
6464
// float(E1, D) LUT1, int32(B, L1) I1,
6565
// float(E2, D) LUT2, int32(B, L2) I2) -> (O1, O2)
6666
// {
67-
// O1(i, j) +=! LUT1(I1(i, k), j)
68-
// O2(i, j) +=! LUT2(I2(i, k), j)
67+
// O1(b, d) +=! LUT1(I1(b, r_l1), d)
68+
// O2(b, d) +=! LUT2(I2(b, r_l2), d)
6969
// }
7070
// def _3FCRELU(
7171
// float(B,M) I, float(O,N) W2, float(O) B2,
7272
// float(P,O) W3, float(P) B3, float(Q,P) W4,
7373
// float(Q) B4) -> (O1, O2, O3, O4)
7474
// {
75-
// O2(b, o) = B2(o)
76-
// O2(b, o) += O1(b, n) * W2(o, n)
77-
// O2(b, o) = fmax(O2(b, o), 0)
78-
// O3(b, p) = B3(p)
79-
// O3(b, p) += O2(b, o) * W3(p, o)
80-
// O3(b, p) = fmax(O3(b, p), 0)
81-
// O4(b, q) = B4(q)
82-
// O4(b, q) += O3(b, p) * W4(q, p)
83-
// O4(b, q) = fmax(O4(b, q), 0)
75+
// O2(b, o) = B2(o)
76+
// O2(b, o) += O1(b, n) * W2(o, n)
77+
// O2(b, o) = fmax(O2(b, o), 0)
78+
// O3(b, p) = B3(p)
79+
// O3(b, p) += O2(b, o) * W3(p, o)
80+
// O3(b, p) = fmax(O3(b, p), 0)
81+
// O4(b, q) = B4(q)
82+
// O4(b, q) += O3(b, p) * W4(q, p)
83+
// O4(b, q) = fmax(O4(b, q), 0)
8484
// }
8585
// def prod_model(float(E1, D) LUT1, int32(B, L1) I1,
8686
// float(E2, D) LUT2, int32(B, L2) I2,
@@ -91,15 +91,15 @@ DEFINE_uint32(Q, 2, "W4_h");
9191
// float(Q,P) W4, float(Q) B4)
9292
// -> (C1, C2, C3, I, O1, O2, O3, O4)
9393
// {
94-
// (C1, C2) = _2LUT(LUT1, I1, LUT2, I2)
95-
// C3(b, wy) += I3(b, wxx) * W(wy, wxx)
96-
// I(b, m) = Concat(C1, C2, C3) // not in TC atm
97-
// O1(b, n) = B1(n)
98-
// O1(b, n) += I(b, m) * W1(n, m)
99-
// O1(b, n) = fmax(O1(b, n), 0)
94+
// (C1, C2) = _2LUT(LUT1, I1, LUT2, I2)
95+
// C3(b, wy) +=! I3(b, r_wx) * W(wy, r_wx)
96+
// I(b, m) = Concat(C1, C2, C3) // not in TC atm
97+
// O1(b, n) = B1(n)
98+
// O1(b, n) +=! I(b, m) * W1(n, m)
99+
// O1(b, n) = fmax(O1(b, n), 0)
100100
// (O2, O3, O4) =
101-
// _3FCRELU(I, W1, B1, W2, B2, W3, B3, W4, B4)
102-
// # O4 goes out to binary classifier, omitted here
101+
// _3FCRELU(I, W1, B1, W2, B2, W3, B3, W4, B4)
102+
// # O4 goes out to binary classifier, omitted here
103103
// }
104104

105105
class ProductionModel : public Benchmark {
@@ -191,9 +191,9 @@ void ProductionModel::run1LUT(
191191

192192
std::vector<at::Tensor> inputs = {LUT1, IDX1};
193193
std::string tc = R"(
194-
def _1LUT(float(E1, D) LUT1, int32(B, L1) I1) -> (O1) {
195-
O1(i, j) +=! LUT1(I1(i, k), j)
196-
}
194+
def _1LUT(float(E1, D) LUT1, int32(B, L1) I1) -> (O1) {
195+
O1(b, d) +=! LUT1(I1(b, r_l1), d)
196+
}
197197
)";
198198

199199
std::string suffix = std::string("_B_") + std::to_string(FLAGS_B) +
@@ -294,10 +294,10 @@ void ProductionModel::run2LUT(
294294

295295
std::vector<at::Tensor> inputs = {LUT1, IDX1, LUT2, IDX2};
296296
std::string tc = R"(
297-
def _2LUT(float(E1, D) LUT1, int32(B, L1) I1, float(E2, D) LUT2, int32(B, L2) I2) -> (O1, O2) {
298-
O1(i, j) +=! LUT1(I1(i, k), j)
299-
O2(i, j) +=! LUT2(I2(i, k), j)
300-
}
297+
def _2LUT(float(E1, D) LUT1, int32(B, L1) I1, float(E2, D) LUT2, int32(B, L2) I2) -> (O1, O2) {
298+
O1(b, d) +=! LUT1(I1(b, r_l1), d)
299+
O2(b, d) +=! LUT2(I2(b, r_l2), d)
300+
}
301301
)";
302302

303303
std::string suffix = std::string("_B_") + std::to_string(FLAGS_B) +
@@ -353,9 +353,9 @@ void ProductionModel::runC3(
353353

354354
std::vector<at::Tensor> inputs = {I, W};
355355
std::string tc = R"TC(
356-
def _C3(float(B,WX) I, float(WY, WX) W) -> (C3) {
357-
C3(b, wy) +=! I(b, wxx) * W(wy, wxx)
358-
}
356+
def _C3(float(B,WX) I, float(WY, WX) W) -> (C3) {
357+
C3(b, wy) +=! I(b, r_wx) * W(wy, r_wx)
358+
}
359359
)TC";
360360

361361
std::string suffix = std::string("_B_") + std::to_string(FLAGS_B) +
@@ -408,11 +408,11 @@ void ProductionModel::runMLP1(
408408

409409
std::vector<at::Tensor> inputs = {I, W1, B1};
410410
std::string tc = R"TC(
411-
def mlp1(float(B,M) I, float(M, N) W1, float(N) B1) -> (O1) {
412-
O1(b, n) +=! I(b, mm) * W1(mm, n)
413-
O1(b, n) = O1(b, n) + B1(n)
414-
O1(b, n) = fmax(O1(b, n), 0)
415-
}
411+
def mlp1(float(B,M) I, float(M, N) W1, float(N) B1) -> (O1) {
412+
O1(b, n) +=! I(b, r_m) * W1(r_m, n)
413+
O1(b, n) = O1(b, n) + B1(n)
414+
O1(b, n) = fmax(O1(b, n), 0)
415+
}
416416
)TC";
417417

418418
std::string suffix = std::string("_B_") + std::to_string(FLAGS_B) +
@@ -474,17 +474,17 @@ void ProductionModel::runMLP3(
474474

475475
std::vector<at::Tensor> inputs = {I, W2, B2, W3, B3, W4, B4};
476476
std::string tc = R"TC(
477-
def mlp3(float(B,N) I, float(O,N) W2, float(O) B2, float(P,O) W3, float(P) B3, float(Q,P) W4, float(Q) B4) -> (O2, O3, O4) {
478-
O2(b, o) +=! I(b, n) * W2(o, n)
479-
O2(b, o) = O2(b, o) + B2(o)
480-
O2(b, o) = fmax(O2(b, o), 0)
477+
def mlp3(float(B,N) I, float(O,N) W2, float(O) B2, float(P,O) W3, float(P) B3, float(Q,P) W4, float(Q) B4) -> (O2, O3, O4) {
478+
O2(b, o) +=! I(b, n) * W2(o, n)
479+
O2(b, o) = O2(b, o) + B2(o)
480+
O2(b, o) = fmax(O2(b, o), 0)
481481
O3(b, p) +=! O2(b, o) * W3(p, o)
482-
O3(b, p) = O3(b, p) + B3(p)
483-
O3(b, p) = fmax(O3(b, p), 0)
482+
O3(b, p) = O3(b, p) + B3(p)
483+
O3(b, p) = fmax(O3(b, p), 0)
484484
O4(b, q) +=! O3(b, p) * W4(q, p)
485-
O4(b, q) = O4(b, q) + B4(q)
486-
O4(b, q) = fmax(O4(b, q), 0)
487-
}
485+
O4(b, q) = O4(b, q) + B4(q)
486+
O4(b, q) = fmax(O4(b, q), 0)
487+
}
488488
)TC";
489489

490490
std::string suffix = std::string("_B_") + std::to_string(FLAGS_B) +

‎benchmarks/batchmatmul.cc‎

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -76,9 +76,9 @@ void BatchMatMul::runBatchMatMul(
7676

7777
std::vector<at::Tensor> inputs = {X, Y};
7878
std::string tc = R"(
79-
def batch_matmul(float(B, N, M) X, float(B, M, K) Y) -> (Z) {
80-
Z(b, n, k) +=! X(b, n, mm) * Y(b, mm, k)
81-
}
79+
def batch_matmul(float(B, N, M) X, float(B, M, K) Y) -> (Z) {
80+
Z(b, n, k) +=! X(b, n, r_m) * Y(b, r_m, k)
81+
}
8282
)";
8383

8484
std::string suffix = std::string("_B_") + std::to_string(FLAGS_B) +

‎benchmarks/group_convolution.cc‎

Lines changed: 6 additions & 6 deletions
Original file line numberDiff line numberDiff line change
@@ -122,13 +122,13 @@ void GroupConvolution::runGroupConvolution(
122122
.resize_({G, F});
123123
std::vector<at::Tensor> inputs = {tI, tW, tB};
124124
std::string tc = R"(
125-
def group_convolution(float(N,G,C,H,W) I, float(G,F,C,KH,KW) W1, float(G,F) B)
126-
-> (O)
127-
{
125+
def group_convolution(float(N,G,C,H,W) I, float(G,F,C,KH,KW) W1, float(G,F) B)
126+
-> (O)
127+
{
128128
O(n, g, f, h, w) +=!
129-
I(n, g, c, h + kh, w + kw) * W1(g, f, c, kh, kw)
130-
O(n, g, f, h, w) = O(n, g, f, h, w) + B(g, f)
131-
}
129+
I(n, g, r_c, h + r_kh, w + r_kw) * W1(g, f, r_c, r_kh, r_kw)
130+
O(n, g, f, h, w) = O(n, g, f, h, w) + B(g, f)
131+
}
132132
)";
133133

134134
std::string suffix = std::string("_N_") + std::to_string(FLAGS_N) +

‎benchmarks/tmm.cc‎

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -73,9 +73,9 @@ void TransposedMatMul::runTransposedMatMul(
7373

7474
std::vector<at::Tensor> inputs = {A, B};
7575
std::string tc = R"TC(
76-
def tmm(float(M,K) A, float(N,K) B) -> (C) {
77-
C(m, n) +=! A(m, kk) * B(n, kk)
78-
}
76+
def tmm(float(M,K) A, float(N,K) B) -> (C) {
77+
C(m, n) +=! A(m, r_k) * B(n, r_k)
78+
}
7979
)TC";
8080

8181
std::string suffix = std::string("_M_") + std::to_string(FLAGS_M) +

0 commit comments

Comments
 (0)