Merge pull request #214 from ROCmSoftwarePlatform/batchnorm-rewrite

Improve batchnorm rewrite

src/fwd_conv_batchnorm_rewrite.cpp

@@ -3,6 +3,7 @@
 #include <migraphx/instruction.hpp>
 #include <migraphx/operators.hpp>
 #include <migraphx/iterator_for.hpp>
+#include <migraphx/ranges.hpp>
 #include <migraphx/dfor.hpp>
 
 namespace migraphx {
@@ -14,32 +15,30 @@ void fwd_conv_batchnorm_rewrite::apply(program& p) const
     {
         if(ins->name() != "batch_norm_inference")
             continue;
-        if(not std::all_of(ins->inputs().begin() + 1, ins->inputs().end(), [](auto arg) {
-               return arg->name() == "@literal";
-           }))
+        // Get scale, bias, mean, variance from inputs
+        auto gamma    = ins->inputs()[1]->eval();
+        auto bias     = ins->inputs()[2]->eval();
+        auto mean     = ins->inputs()[3]->eval();
+        auto variance = ins->inputs()[4]->eval();
+        if(any_of({gamma, bias, mean, variance}, [](auto arg) { return arg.empty(); }))
             continue;
 
         auto conv_ins = ins->inputs()[0];
         if(conv_ins->name() != "convolution")
             continue;
-        if(conv_ins->inputs()[1]->name() != "@literal")
+        // Get convolution weights
+        auto weights = conv_ins->inputs()[1]->eval();
+        if(weights.empty())
             continue;
-        // Get scale, bias, mean, variance from instruction_ref
-        const auto& gamma    = ins->inputs()[1]->get_literal();
-        const auto& bias     = ins->inputs()[2]->get_literal();
-        const auto& mean     = ins->inputs()[3]->get_literal();
-        const auto& variance = ins->inputs()[4]->get_literal();
         // Get epsilon
         auto bn_op   = any_cast<op::batch_norm_inference>(ins->get_operator());
         auto epsilon = bn_op.epsilon;
-        // Get convolution weights
-        const auto& weights = conv_ins->inputs()[1]->get_literal();
         // Get convolution op
         auto conv_op      = conv_ins->get_operator();
         auto weights_lens = weights.get_shape().lens();
         auto conv_lens    = conv_ins->get_shape().lens();
         argument new_weights{weights.get_shape()};
-        argument new_bias{bias.get_shape()};
+        argument new_bias{{bias.get_shape().type(), {bias.get_shape().elements()}}};
         visit_all(weights, gamma, bias, mean, variance, new_weights, new_bias)(
             [&](auto weights2,
                 auto gamma2,
@@ -51,11 +50,11 @@ void fwd_conv_batchnorm_rewrite::apply(program& p) const
                 dfor(weights_lens[0], weights_lens[1], weights_lens[2], weights_lens[3])(
                     [&](std::size_t k, std::size_t c, std::size_t h, std::size_t w) {
                         new_weights2(k, c, h, w) =
-                            gamma2(k) / std::sqrt(variance2(k) + epsilon) * weights2(k, c, h, w);
+                            gamma2[k] / std::sqrt(variance2[k] + epsilon) * weights2(k, c, h, w);
                     });
                 dfor(new_bias.get_shape().elements())([&](std::size_t c) {
-                    new_bias2(c) =
-                        bias2(c) - (gamma2(c) * mean2(c) / std::sqrt(variance2(c) + epsilon));
+                    new_bias2[c] =
+                        bias2[c] - (gamma2[c] * mean2[c] / std::sqrt(variance2[c] + epsilon));
                 });
             });
         // Replace convolution instruction with updated weights

src/include/migraphx/check_shapes.hpp

@@ -18,6 +18,11 @@ struct check_shapes
     {
     }
 
+    template <class Op>
+    check_shapes(const shape* b, const shape* e, const Op& op) : begin(b), end(e), name(op.name())
+    {
+    }
+
     check_shapes(const std::vector<shape>& s) : begin(s.data()), end(s.data() + s.size()) {}
 
     template <class Op>
@@ -119,6 +124,13 @@ struct check_shapes
         return *this;
     }
 
+    const check_shapes& elements(std::size_t n) const
+    {
+        if(!this->all_of([&](const shape& s) { return s.elements() == n; }))
+            MIGRAPHX_THROW(prefix() + "Wrong number of elements");
+        return *this;
+    }
+
     template <class F>
     bool same(F f) const
     {

src/include/migraphx/operators.hpp

@@ -56,6 +56,9 @@ struct batch_norm_inference
     shape compute_shape(std::vector<shape> inputs) const
     {
         check_shapes{inputs, *this}.has(5);
+        check_shapes{inputs.data(), inputs.data() + 1, *this}.only_dims(4);
+        check_shapes{inputs.data() + 1, inputs.data() + inputs.size(), *this}.same_shape().elements(
+            inputs.front().lens()[1]);
         return inputs.front();
     }
 };

src/include/migraphx/ranges.hpp

@@ -71,6 +71,30 @@ bool all_of(const std::initializer_list<T>& c, const Predicate& p)
     return std::all_of(c.begin(), c.end(), p);
 }
 
+template <class C, class Predicate>
+bool any_of(const C& c, const Predicate& p)
+{
+    return std::any_of(c.begin(), c.end(), p);
+}
+
+template <class T, class Predicate>
+bool any_of(const std::initializer_list<T>& c, const Predicate& p)
+{
+    return std::any_of(c.begin(), c.end(), p);
+}
+
+template <class C, class Predicate>
+bool none_of(const C& c, const Predicate& p)
+{
+    return std::none_of(c.begin(), c.end(), p);
+}
+
+template <class T, class Predicate>
+bool none_of(const std::initializer_list<T>& c, const Predicate& p)
+{
+    return std::none_of(c.begin(), c.end(), p);
+}
+
 template <class Range, class Iterator>
 void copy(Range&& r, Iterator it)
 {

src/targets/cpu/lowering.cpp

@@ -75,10 +75,10 @@ struct cpu_batch_norm_inference
 
                     par_dfor(num_batch, num_channels, image_height, image_width)(
                         [&](std::size_t n, std::size_t c, std::size_t h, std::size_t w) {
-                            assert((variance(c) + epsilon) > 0);
-                            result(n, c, h, w) = gamma(c) * (buffer(n, c, h, w) - mean(c)) /
-                                                     std::sqrt(variance(c) + epsilon) +
-                                                 bias(c);
+                            assert((variance[c] + epsilon) > 0);
+                            result(n, c, h, w) = gamma[c] * (buffer(n, c, h, w) - mean[c]) /
+                                                     std::sqrt(variance[c] + epsilon) +
+                                                 bias[c];
                         });
                 });
         }

src/targets/gpu/fuse_ops.cpp

@@ -140,6 +140,8 @@ MIGRAPHX_PRED_MATCHER(fusable_conv, instruction_ref ins)
     auto conv = any_cast<miopen_convolution>(ins->get_operator());
     if(conv.op.group > 1)
         return false;
+    if(conv.op.padding_mode != op::padding_mode_t::default_)
+        return false;
     if(wei.lens()[1] > 512 and conv.algo != miopenConvolutionFwdAlgoWinograd)
         return false;
     auto op = conv.op;

test/fwd_conv_batchnorm_rewrite_test.cpp

@@ -3,9 +3,13 @@
 #include <migraphx/cpu/target.hpp>
 #include <migraphx/operators.hpp>
 #include <migraphx/instruction.hpp>
+#include <migraphx/generate.hpp>
+#include <migraphx/ranges.hpp>
 #include <test.hpp>
 #include <migraphx/verify.hpp>
 
+bool is_batch_norm(migraphx::instruction& ins) { return ins.name() == "batch_norm_inference"; }
+
 TEST_CASE(fwd_conv_batchnorm_rewrite_test)
 {
     std::vector<float> xdata = {
@@ -65,4 +69,105 @@ TEST_CASE(fwd_conv_batchnorm_rewrite_test)
     EXPECT(migraphx::verify_range(results_vector1, results_vector2));
 }
 
+TEST_CASE(non_literal)
+{
+
+    migraphx::shape xs{migraphx::shape::float_type, {1, 3, 8, 8}};
+    migraphx::shape ws{migraphx::shape::float_type, {4, 3, 1, 1}};
+    migraphx::shape vars{migraphx::shape::float_type, {4}};
+    auto create_program = [&]() {
+        migraphx::program p;
+
+        auto x        = p.add_parameter("x", xs);
+        auto w        = p.add_parameter("w", ws);
+        auto conv     = p.add_instruction(migraphx::op::convolution{}, x, w);
+        auto scale    = p.add_literal(migraphx::abs(migraphx::generate_literal(vars, 1)));
+        auto bias     = p.add_literal(migraphx::abs(migraphx::generate_literal(vars, 2)));
+        auto mean     = p.add_literal(migraphx::abs(migraphx::generate_literal(vars, 3)));
+        auto variance = p.add_literal(migraphx::abs(migraphx::generate_literal(vars, 4)));
+        p.add_instruction(migraphx::op::batch_norm_inference{}, conv, scale, bias, mean, variance);
+        return p;
+    };
+
+    migraphx::program p1 = create_program();
+    migraphx::program p2 = create_program();
+    migraphx::fwd_conv_batchnorm_rewrite opt;
+    opt.apply(p2);
+    EXPECT(any_of(p1, &is_batch_norm));
+    EXPECT(any_of(p2, &is_batch_norm));
+}
+
+TEST_CASE(as_literal)
+{
+
+    migraphx::shape xs{migraphx::shape::float_type, {1, 3, 8, 8}};
+    migraphx::shape ws{migraphx::shape::float_type, {4, 3, 1, 1}};
+    migraphx::shape vars{migraphx::shape::float_type, {4}};
+    auto create_program = [&]() {
+        migraphx::program p;
+
+        auto x        = p.add_literal(migraphx::generate_literal(xs, 1));
+        auto w        = p.add_literal(migraphx::generate_literal(ws, 1));
+        auto conv     = p.add_instruction(migraphx::op::convolution{}, x, w);
+        auto scale    = p.add_literal(migraphx::abs(migraphx::generate_literal(vars, 1)));
+        auto bias     = p.add_literal(migraphx::abs(migraphx::generate_literal(vars, 2)));
+        auto mean     = p.add_literal(migraphx::abs(migraphx::generate_literal(vars, 3)));
+        auto variance = p.add_literal(migraphx::abs(migraphx::generate_literal(vars, 4)));
+        p.add_instruction(migraphx::op::batch_norm_inference{}, conv, scale, bias, mean, variance);
+        return p;
+    };
+
+    migraphx::program p1 = create_program();
+    migraphx::program p2 = create_program();
+    migraphx::fwd_conv_batchnorm_rewrite opt;
+    opt.apply(p2);
+    EXPECT(any_of(p1, &is_batch_norm));
+    EXPECT(none_of(p2, &is_batch_norm));
+
+    p1.compile(migraphx::cpu::target{});
+    p2.compile(migraphx::cpu::target{});
+
+    auto result1 = p1.eval({});
+    auto result2 = p2.eval({});
+    visit_all(result1, result2)([&](auto r1, auto r2) { EXPECT(migraphx::verify_range(r1, r2)); });
+}
+
+TEST_CASE(literal_reshape)
+{
+    migraphx::shape xs{migraphx::shape::float_type, {1, 3, 8, 8}};
+    migraphx::shape ws{migraphx::shape::float_type, {4, 3, 1, 1}};
+    migraphx::shape vars{migraphx::shape::float_type, {4}};
+
+    auto create_program = [&]() {
+        migraphx::program p;
+        auto reshape = [&](auto ins) {
+            return p.add_instruction(migraphx::op::reshape{{1, 4, 1, 1}}, ins);
+        };
+
+        auto x        = p.add_literal(migraphx::generate_literal(xs, 1));
+        auto w        = p.add_literal(migraphx::generate_literal(ws, 1));
+        auto conv     = p.add_instruction(migraphx::op::convolution{}, x, w);
+        auto scale    = reshape(p.add_literal(migraphx::abs(migraphx::generate_literal(vars, 1))));
+        auto bias     = reshape(p.add_literal(migraphx::abs(migraphx::generate_literal(vars, 2))));
+        auto mean     = reshape(p.add_literal(migraphx::abs(migraphx::generate_literal(vars, 3))));
+        auto variance = reshape(p.add_literal(migraphx::abs(migraphx::generate_literal(vars, 4))));
+        p.add_instruction(migraphx::op::batch_norm_inference{}, conv, scale, bias, mean, variance);
+        return p;
+    };
+
+    migraphx::program p1 = create_program();
+    migraphx::program p2 = create_program();
+    migraphx::fwd_conv_batchnorm_rewrite opt;
+    opt.apply(p2);
+    EXPECT(any_of(p1, &is_batch_norm));
+    EXPECT(none_of(p2, &is_batch_norm));
+
+    p1.compile(migraphx::cpu::target{});
+    p2.compile(migraphx::cpu::target{});
+
+    auto result1 = p1.eval({});
+    auto result2 = p2.eval({});
+    visit_all(result1, result2)([&](auto r1, auto r2) { EXPECT(migraphx::verify_range(r1, r2)); });
+}
+
 int main(int argc, const char* argv[]) { test::run(argc, argv); }