Test model refactor (#2720)

* template funcion fill_meta

* add function get_plan_max_idx

* add function get_mdn_max_idx

* remove temp builder variables

* get_mdn_max_idx->get_lead_data

* get_pan_max_idx->get_plan_data

* fill_lane_line

* combine fill_lane_line&fill_path into one function

* fill_lead_vw

* using unique_ptr

* prefer using constexpr

* more constexpr,remove duplicate defined TRAJECTORY_SIZE

* remove suffix _arr from variable name

* misc

* remove extern c

* refactor model_publish

* remove unused paramaters

* traffic_convention to c style array

* c style array:prev_desire&pulse_desire

* fix error&make easy for review

* const mat3 &transform

* move cl_command_queue into ModelState

* use maco LEAD_MHP_SELECTION

* move constexpr from .h to .cc

* remove #define MODEL_NAME

* modeldata.h: contexpr

* remove param temporal from model_init

* helper function get_best_data

* fix probs

* int

Co-authored-by: deanlee <deanlee3@gmail.com>
pull/2702/head
HaraldSchafer 5 years ago committed by GitHub
parent bbe868d312
commit 630c7309a5
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
  1. 16
      selfdrive/common/modeldata.h
  2. 16
      selfdrive/modeld/modeld.cc
  3. 2
      selfdrive/modeld/models/commonmodel.cc
  4. 17
      selfdrive/modeld/models/commonmodel.h
  5. 320
      selfdrive/modeld/models/driving.cc
  6. 67
      selfdrive/modeld/models/driving.h

@ -1,10 +1,10 @@
#pragma once
#define MODEL_PATH_DISTANCE 192
#define TRAJECTORY_SIZE 33
#define MIN_DRAW_DISTANCE 10.0
#define MAX_DRAW_DISTANCE 100.0
#define POLYFIT_DEGREE 4
#define SPEED_PERCENTILES 10
#define DESIRE_PRED_SIZE 32
#define OTHER_META_SIZE 4
constexpr int MODEL_PATH_DISTANCE = 192;
constexpr int TRAJECTORY_SIZE = 33;
constexpr float MIN_DRAW_DISTANCE = 10.0;
constexpr float MAX_DRAW_DISTANCE = 100.0;
constexpr int POLYFIT_DEGREE = 4;
constexpr int SPEED_PERCENTILES = 10;
constexpr int DESIRE_PRED_SIZE = 32;
constexpr int OTHER_META_SIZE = 4;

@ -117,11 +117,10 @@ int main(int argc, char **argv) {
// cl init
cl_device_id device_id = cl_get_device_id(CL_DEVICE_TYPE_DEFAULT);
cl_context context = CL_CHECK_ERR(clCreateContext(NULL, 1, &device_id, NULL, NULL, &err));
cl_command_queue q = CL_CHECK_ERR(clCreateCommandQueue(context, device_id, 0, &err));
// init the models
ModelState model;
model_init(&model, device_id, context, true);
model_init(&model, device_id, context);
LOGW("models loaded, modeld starting");
// loop
@ -185,8 +184,8 @@ int main(int argc, char **argv) {
visionbuf_sync(&yuv_ion, VISIONBUF_SYNC_TO_DEVICE);
ModelDataRaw model_buf =
model_eval_frame(&model, q, yuv_ion.buf_cl, buf_info.width, buf_info.height,
model_transform, NULL, vec_desire);
model_eval_frame(&model, yuv_ion.buf_cl, buf_info.width, buf_info.height,
model_transform, vec_desire);
mt2 = millis_since_boot();
float model_execution_time = (mt2 - mt1) / 1000.0;
@ -196,10 +195,9 @@ int main(int argc, char **argv) {
if (run_count < 10) frames_dropped = 0; // let frame drops warm up
float frame_drop_ratio = frames_dropped / (1 + frames_dropped);
const float* raw_pred_ptr = send_raw_pred ? (const float *)model.output : nullptr;
model_publish(pm, extra.frame_id, frame_id, vipc_dropped_frames, frame_drop_ratio, model_buf, raw_pred_ptr, extra.timestamp_eof, model_execution_time);
model_publish_v2(pm, extra.frame_id, frame_id, vipc_dropped_frames, frame_drop_ratio, model_buf, raw_pred_ptr, extra.timestamp_eof, model_execution_time);
posenet_publish(pm, extra.frame_id, frame_id, vipc_dropped_frames, frame_drop_ratio, model_buf, extra.timestamp_eof);
const float *raw_pred_ptr = send_raw_pred ? &model.output[0] : nullptr;
model_publish(pm, extra.frame_id, frame_id, frame_drop_ratio, model_buf, raw_pred_ptr, extra.timestamp_eof, model_execution_time);
posenet_publish(pm, extra.frame_id, vipc_dropped_frames, model_buf, extra.timestamp_eof);
LOGD("model process: %.2fms, from last %.2fms, vipc_frame_id %zu, frame_id, %zu, frame_drop %.3f", mt2-mt1, mt1-last, extra.frame_id, frame_id, frame_drop_ratio);
last = mt1;
@ -217,8 +215,6 @@ int main(int argc, char **argv) {
err = pthread_join(live_thread_handle, NULL);
assert(err == 0);
CL_CHECK(clReleaseContext(context));
CL_CHECK(clReleaseCommandQueue(q));
pthread_mutex_destroy(&transform_lock);
return 0;
}

@ -25,7 +25,7 @@ void frame_init(ModelFrame* frame, int width, int height,
float *frame_prepare(ModelFrame* frame, cl_command_queue q,
cl_mem yuv_cl, int width, int height,
mat3 transform) {
const mat3 &transform) {
transform_queue(&frame->transform, q,
yuv_cl, width, height,
frame->transformed_y_cl, frame->transformed_u_cl, frame->transformed_v_cl,

@ -1,6 +1,4 @@
#ifndef COMMONMODEL_H
#define COMMONMODEL_H
#pragma once
#define CL_USE_DEPRECATED_OPENCL_1_2_APIS
#ifdef __APPLE__
#include <OpenCL/cl.h>
@ -14,10 +12,6 @@
#include "transforms/transform.h"
#include "transforms/loadyuv.h"
#ifdef __cplusplus
extern "C" {
#endif
const bool send_raw_pred = getenv("SEND_RAW_PRED") != NULL;
void softmax(const float* input, float* output, size_t len);
@ -37,12 +31,5 @@ void frame_init(ModelFrame* frame, int width, int height,
cl_device_id device_id, cl_context context);
float *frame_prepare(ModelFrame* frame, cl_command_queue q,
cl_mem yuv_cl, int width, int height,
mat3 transform);
const mat3 &transform);
void frame_free(ModelFrame* frame);
#ifdef __cplusplus
}
#endif
#endif

@ -8,25 +8,42 @@
#include "common/timing.h"
#include "common/params.h"
#include "driving.h"
#define MIN_VALID_LEN 10.0
#define TRAJECTORY_SIZE 33
#define TRAJECTORY_TIME 10.0
#define TRAJECTORY_DISTANCE 192.0
#define PLAN_IDX 0
#define LL_IDX PLAN_IDX + PLAN_MHP_N*(PLAN_MHP_GROUP_SIZE)
#define LL_PROB_IDX LL_IDX + 4*2*2*33
#define RE_IDX LL_PROB_IDX + 4
#define LEAD_IDX RE_IDX + 2*2*2*33
#define LEAD_PROB_IDX LEAD_IDX + LEAD_MHP_N*(LEAD_MHP_GROUP_SIZE)
#define DESIRE_STATE_IDX LEAD_PROB_IDX + 3
#define META_IDX DESIRE_STATE_IDX + DESIRE_LEN
#define POSE_IDX META_IDX + OTHER_META_SIZE + DESIRE_PRED_SIZE
#define OUTPUT_SIZE POSE_IDX + POSE_SIZE
#include "clutil.h"
constexpr int MODEL_WIDTH = 512;
constexpr int MODEL_HEIGHT = 256;
constexpr int MODEL_FRAME_SIZE = MODEL_WIDTH * MODEL_HEIGHT * 3 / 2;
constexpr int PLAN_MHP_N = 5;
constexpr int PLAN_MHP_COLUMNS = 30;
constexpr int PLAN_MHP_VALS = 30*33;
constexpr int PLAN_MHP_SELECTION = 1;
constexpr int PLAN_MHP_GROUP_SIZE = (2*PLAN_MHP_VALS + PLAN_MHP_SELECTION);
constexpr int LEAD_MHP_N = 5;
constexpr int LEAD_MHP_VALS = 4;
constexpr int LEAD_MHP_SELECTION = 3;
constexpr int LEAD_MHP_GROUP_SIZE = (2*LEAD_MHP_VALS + LEAD_MHP_SELECTION);
constexpr int POSE_SIZE = 12;
constexpr int MIN_VALID_LEN = 10;
constexpr int TRAJECTORY_TIME = 10;
constexpr float TRAJECTORY_DISTANCE = 192.0;
constexpr int PLAN_IDX = 0;
constexpr int LL_IDX = PLAN_IDX + PLAN_MHP_N*PLAN_MHP_GROUP_SIZE;
constexpr int LL_PROB_IDX = LL_IDX + 4*2*2*33;
constexpr int RE_IDX = LL_PROB_IDX + 4;
constexpr int LEAD_IDX = RE_IDX + 2*2*2*33;
constexpr int LEAD_PROB_IDX = LEAD_IDX + LEAD_MHP_N*(LEAD_MHP_GROUP_SIZE);
constexpr int DESIRE_STATE_IDX = LEAD_PROB_IDX + 3;
constexpr int META_IDX = DESIRE_STATE_IDX + DESIRE_LEN;
constexpr int POSE_IDX = META_IDX + OTHER_META_SIZE + DESIRE_PRED_SIZE;
constexpr int OUTPUT_SIZE = POSE_IDX + POSE_SIZE;
#ifdef TEMPORAL
#define TEMPORAL_SIZE 512
constexpr int TEMPORAL_SIZE = 512;
#else
#define TEMPORAL_SIZE 0
constexpr int TEMPORAL_SIZE = 0;
#endif
// #define DUMP_YUV
@ -35,36 +52,26 @@ Eigen::Matrix<float, MODEL_PATH_DISTANCE, POLYFIT_DEGREE - 1> vander;
float X_IDXS[TRAJECTORY_SIZE];
float T_IDXS[TRAJECTORY_SIZE];
void model_init(ModelState* s, cl_device_id device_id, cl_context context, int temporal) {
void model_init(ModelState* s, cl_device_id device_id, cl_context context) {
frame_init(&s->frame, MODEL_WIDTH, MODEL_HEIGHT, device_id, context);
s->input_frames = (float*)calloc(MODEL_FRAME_SIZE * 2, sizeof(float));
const int output_size = OUTPUT_SIZE + TEMPORAL_SIZE;
s->output = (float*)calloc(output_size, sizeof(float));
s->input_frames = std::make_unique<float[]>(MODEL_FRAME_SIZE * 2);
s->m = new DefaultRunModel("../../models/supercombo.dlc", s->output, output_size, USE_GPU_RUNTIME);
constexpr int output_size = OUTPUT_SIZE + TEMPORAL_SIZE;
s->output = std::make_unique<float[]>(output_size);
s->m = std::make_unique<DefaultRunModel>("../../models/supercombo.dlc", &s->output[0], output_size, USE_GPU_RUNTIME);
#ifdef TEMPORAL
assert(temporal);
s->m->addRecurrent(&s->output[OUTPUT_SIZE], TEMPORAL_SIZE);
#endif
#ifdef DESIRE
s->prev_desire = std::make_unique<float[]>(DESIRE_LEN);
s->pulse_desire = std::make_unique<float[]>(DESIRE_LEN);
s->m->addDesire(s->pulse_desire.get(), DESIRE_LEN);
s->m->addDesire(s->pulse_desire, DESIRE_LEN);
#endif
#ifdef TRAFFIC_CONVENTION
s->traffic_convention = std::make_unique<float[]>(TRAFFIC_CONVENTION_LEN);
s->m->addTrafficConvention(s->traffic_convention.get(), TRAFFIC_CONVENTION_LEN);
bool is_rhd = Params().read_db_bool("IsRHD");
if (is_rhd) {
s->traffic_convention[1] = 1.0;
} else {
s->traffic_convention[0] = 1.0;
}
const int idx = Params().read_db_bool("IsRHD") ? 1 : 0;
s->traffic_convention[idx] = 1.0;
s->m->addTrafficConvention(s->traffic_convention, TRAFFIC_CONVENTION_LEN);
#endif
// Build Vandermonde matrix
@ -75,12 +82,12 @@ void model_init(ModelState* s, cl_device_id device_id, cl_context context, int t
vander(i, j) = pow(X_IDXS[i], POLYFIT_DEGREE-j-1);
}
}
s->q = CL_CHECK_ERR(clCreateCommandQueue(context, device_id, 0, &err));
}
ModelDataRaw model_eval_frame(ModelState* s, cl_command_queue q,
cl_mem yuv_cl, int width, int height,
mat3 transform, void* sock,
float *desire_in) {
ModelDataRaw model_eval_frame(ModelState* s, cl_mem yuv_cl, int width, int height,
const mat3 &transform, float *desire_in) {
#ifdef DESIRE
if (desire_in != NULL) {
for (int i = 1; i < DESIRE_LEN; i++) {
@ -98,10 +105,10 @@ ModelDataRaw model_eval_frame(ModelState* s, cl_command_queue q,
//for (int i = 0; i < OUTPUT_SIZE + TEMPORAL_SIZE; i++) { printf("%f ", s->output[i]); } printf("\n");
float *new_frame_buf = frame_prepare(&s->frame, q, yuv_cl, width, height, transform);
float *new_frame_buf = frame_prepare(&s->frame, s->q, yuv_cl, width, height, transform);
memmove(&s->input_frames[0], &s->input_frames[MODEL_FRAME_SIZE], sizeof(float)*MODEL_FRAME_SIZE);
memmove(&s->input_frames[MODEL_FRAME_SIZE], new_frame_buf, sizeof(float)*MODEL_FRAME_SIZE);
s->m->execute(s->input_frames, MODEL_FRAME_SIZE*2);
s->m->execute(&s->input_frames[0], MODEL_FRAME_SIZE*2);
#ifdef DUMP_YUV
FILE *dump_yuv_file = fopen("/sdcard/dump.yuv", "wb");
@ -110,7 +117,7 @@ ModelDataRaw model_eval_frame(ModelState* s, cl_command_queue q,
assert(1==2);
#endif
clEnqueueUnmapMemObject(q, s->frame.net_input, (void*)new_frame_buf, 0, NULL, NULL);
clEnqueueUnmapMemObject(s->q, s->frame.net_input, (void*)new_frame_buf, 0, NULL, NULL);
// net outputs
ModelDataRaw net_outputs;
@ -126,10 +133,8 @@ ModelDataRaw model_eval_frame(ModelState* s, cl_command_queue q,
}
void model_free(ModelState* s) {
free(s->output);
free(s->input_frames);
frame_free(&s->frame);
delete s->m;
CL_CHECK(clReleaseCommandQueue(s->q));
}
void poly_fit(float *in_pts, float *in_stds, float *out, int valid_len) {
@ -157,48 +162,53 @@ void poly_fit(float *in_pts, float *in_stds, float *out, int valid_len) {
out[3] = y0;
}
void fill_path(cereal::ModelData::PathData::Builder path, const float * data, float valid_len, int valid_len_idx) {
float points_arr[TRAJECTORY_SIZE];
float stds_arr[TRAJECTORY_SIZE];
float poly_arr[POLYFIT_DEGREE];
float std;
for (int i=0; i<TRAJECTORY_SIZE; i++) {
// negative sign because mpc has left positive
points_arr[i] = -data[30*i + 16];
stds_arr[i] = exp(data[30*(33 + i) + 16]);
static const float *get_best_data(const float *data, int size, int group_size, int offset) {
int max_idx = 0;
for (int i = 1; i < size; i++) {
if (data[(i + 1) * group_size + offset] >
data[(max_idx + 1) * group_size + offset]) {
max_idx = i;
}
}
std = stds_arr[0];
poly_fit(points_arr, stds_arr, poly_arr, valid_len_idx);
return &data[max_idx * group_size];
}
path.setPoly(poly_arr);
path.setProb(1.0);
path.setStd(std);
path.setValidLen(valid_len);
static const float *get_plan_data(float *plan) {
return get_best_data(plan, PLAN_MHP_N, PLAN_MHP_GROUP_SIZE, -1);
}
void fill_lane_line(cereal::ModelData::PathData::Builder path, const float * data, int ll_idx, float valid_len, int valid_len_idx, float prob) {
float points_arr[TRAJECTORY_SIZE];
float stds_arr[TRAJECTORY_SIZE];
float poly_arr[POLYFIT_DEGREE];
float std;
static const float *get_lead_data(const float *lead, int t_offset) {
return get_best_data(lead, LEAD_MHP_N, LEAD_MHP_GROUP_SIZE, t_offset - LEAD_MHP_SELECTION);
}
void fill_path(cereal::ModelData::PathData::Builder path, const float *data, const float prob,
float valid_len, int valid_len_idx, int ll_idx) {
float points[TRAJECTORY_SIZE] = {};
float stds[TRAJECTORY_SIZE] = {};
float poly[POLYFIT_DEGREE] = {};
for (int i=0; i<TRAJECTORY_SIZE; i++) {
// negative sign because mpc has left positive
points_arr[i] = -data[2*33*ll_idx + 2*i];
stds_arr[i] = exp(data[2*33*(4 + ll_idx) + 2*i]);
if (ll_idx == 0) {
points[i] = -data[30 * i + 16];
stds[i] = exp(data[30 * (33 + i) + 16]);
} else {
points[i] = -data[2 * 33 * ll_idx + 2 * i];
stds[i] = exp(data[2 * 33 * (4 + ll_idx) + 2 * i]);
}
}
std = stds_arr[0];
poly_fit(points_arr, stds_arr, poly_arr, valid_len_idx);
const float std = stds[0];
poly_fit(points, stds, poly, valid_len_idx);
path.setPoly(poly_arr);
path.setPoly(poly);
path.setProb(prob);
path.setStd(std);
path.setValidLen(valid_len);
}
void fill_lead_v2(cereal::ModelDataV2::LeadDataV2::Builder lead, const float * data, float prob, float t) {
lead.setProb(prob);
void fill_lead_v2(cereal::ModelDataV2::LeadDataV2::Builder lead, const float *lead_data, const float *prob, int t_offset, float t) {
const float *data = get_lead_data(lead_data, t_offset);
lead.setProb(sigmoid(prob[t_offset]));
lead.setT(t);
float xyva_arr[LEAD_MHP_VALS];
float xyva_stds_arr[LEAD_MHP_VALS];
@ -210,8 +220,9 @@ void fill_lead_v2(cereal::ModelDataV2::LeadDataV2::Builder lead, const float * d
lead.setXyvaStd(xyva_stds_arr);
}
void fill_lead(cereal::ModelData::LeadData::Builder lead, const float * data, float prob) {
lead.setProb(prob);
void fill_lead(cereal::ModelData::LeadData::Builder lead, const float *lead_data, const float *prob, int t_offset) {
const float *data = get_lead_data(lead_data, t_offset);
lead.setProb(sigmoid(prob[t_offset]));
lead.setDist(data[0]);
lead.setStd(exp(data[LEAD_MHP_VALS]));
// TODO make all msgs same format
@ -223,7 +234,8 @@ void fill_lead(cereal::ModelData::LeadData::Builder lead, const float * data, fl
lead.setRelAStd(exp(data[LEAD_MHP_VALS + 3]));
}
void fill_meta(cereal::ModelData::MetaData::Builder meta, const float * meta_data) {
template <class MetaBuilder>
void fill_meta(MetaBuilder meta, const float *meta_data) {
float desire_state_softmax[DESIRE_LEN];
float desire_pred_softmax[4*DESIRE_LEN];
softmax(&meta_data[0], desire_state_softmax, DESIRE_LEN);
@ -239,22 +251,6 @@ void fill_meta(cereal::ModelData::MetaData::Builder meta, const float * meta_dat
meta.setDesirePrediction(desire_pred_softmax);
}
void fill_meta_v2(cereal::ModelDataV2::MetaData::Builder meta, const float * meta_data) {
float desire_state_softmax[DESIRE_LEN] = {};
float desire_pred_softmax[4*DESIRE_LEN] = {};
softmax(&meta_data[0], desire_state_softmax, DESIRE_LEN);
for (int i=0; i<4; i++) {
softmax(&meta_data[DESIRE_LEN + OTHER_META_SIZE + i*DESIRE_LEN],
&desire_pred_softmax[i*DESIRE_LEN], DESIRE_LEN);
}
meta.setDesireState(desire_state_softmax);
meta.setEngagedProb(sigmoid(meta_data[DESIRE_LEN]));
meta.setGasDisengageProb(sigmoid(meta_data[DESIRE_LEN + 1]));
meta.setBrakeDisengageProb(sigmoid(meta_data[DESIRE_LEN + 2]));
meta.setSteerOverrideProb(sigmoid(meta_data[DESIRE_LEN + 3]));
meta.setDesirePrediction(desire_pred_softmax);
}
void fill_xyzt(cereal::ModelDataV2::XYZTData::Builder xyzt, const float * data,
int columns, int column_offset, float * plan_t_arr) {
float x_arr[TRAJECTORY_SIZE] = {};
@ -292,34 +288,9 @@ void fill_xyzt(cereal::ModelDataV2::XYZTData::Builder xyzt, const float * data,
xyzt.setT(t_arr);
}
void model_publish_v2(PubMaster &pm, uint32_t vipc_frame_id, uint32_t frame_id,
uint32_t vipc_dropped_frames, float frame_drop,
const ModelDataRaw &net_outputs, const float* raw_pred, uint64_t timestamp_eof,
float model_execution_time) {
// make msg
MessageBuilder msg;
auto framed = msg.initEvent().initModelV2();
uint32_t frame_age = (frame_id > vipc_frame_id) ? (frame_id - vipc_frame_id) : 0;
framed.setFrameId(vipc_frame_id);
framed.setFrameAge(frame_age);
framed.setFrameDropPerc(frame_drop * 100);
framed.setTimestampEof(timestamp_eof);
framed.setModelExecutionTime(model_execution_time);
if (send_raw_pred) {
framed.setRawPred(kj::arrayPtr((const uint8_t*)raw_pred, (OUTPUT_SIZE+TEMPORAL_SIZE)*sizeof(float)));
}
void fill_model(cereal::ModelDataV2::Builder &framed, const ModelDataRaw &net_outputs) {
// plan
int plan_mhp_max_idx = 0;
for (int i=1; i<PLAN_MHP_N; i++) {
if (net_outputs.plan[(i + 1)*(PLAN_MHP_GROUP_SIZE) - 1] >
net_outputs.plan[(plan_mhp_max_idx + 1)*(PLAN_MHP_GROUP_SIZE) - 1]) {
plan_mhp_max_idx = i;
}
}
float * best_plan = &net_outputs.plan[plan_mhp_max_idx*(PLAN_MHP_GROUP_SIZE)];
const float *best_plan = get_plan_data(net_outputs.plan);
float plan_t_arr[TRAJECTORY_SIZE];
for (int i=0; i<TRAJECTORY_SIZE; i++) {
plan_t_arr[i] = best_plan[i*PLAN_MHP_COLUMNS + 15];
@ -352,59 +323,24 @@ void model_publish_v2(PubMaster &pm, uint32_t vipc_frame_id, uint32_t frame_id,
framed.setRoadEdgeStds(road_edge_stds_arr);
// meta
auto meta = framed.initMeta();
fill_meta_v2(meta, net_outputs.meta);
fill_meta(framed.initMeta(), net_outputs.meta);
// leads
auto leads = framed.initLeads(LEAD_MHP_SELECTION);
float t_offsets[LEAD_MHP_SELECTION] = {0.0, 2.0, 4.0};
for (int t_offset=0; t_offset<LEAD_MHP_SELECTION; t_offset++) {
int mdn_max_idx = 0;
for (int i=1; i<LEAD_MHP_N; i++) {
if (net_outputs.lead[(i+1)*(LEAD_MHP_GROUP_SIZE) + t_offset - LEAD_MHP_SELECTION] >
net_outputs.lead[(mdn_max_idx + 1)*(LEAD_MHP_GROUP_SIZE) + t_offset - LEAD_MHP_SELECTION]) {
mdn_max_idx = i;
}
}
fill_lead_v2(leads[t_offset], &net_outputs.lead[mdn_max_idx * (LEAD_MHP_GROUP_SIZE)],
sigmoid(net_outputs.lead_prob[t_offset]), t_offsets[t_offset]);
fill_lead_v2(leads[t_offset], net_outputs.lead, net_outputs.lead_prob, t_offset, t_offsets[t_offset]);
}
pm.send("modelV2", msg);
}
void model_publish(PubMaster &pm, uint32_t vipc_frame_id, uint32_t frame_id,
uint32_t vipc_dropped_frames, float frame_drop,
const ModelDataRaw &net_outputs, const float* raw_pred, uint64_t timestamp_eof,
float model_execution_time) {
uint32_t frame_age = (frame_id > vipc_frame_id) ? (frame_id - vipc_frame_id) : 0;
MessageBuilder msg;
auto framed = msg.initEvent().initModel();
framed.setFrameId(vipc_frame_id);
framed.setFrameAge(frame_age);
framed.setFrameDropPerc(frame_drop * 100);
framed.setTimestampEof(timestamp_eof);
framed.setModelExecutionTime(model_execution_time);
if (send_raw_pred) {
framed.setRawPred(kj::arrayPtr((const uint8_t*)raw_pred, (OUTPUT_SIZE+TEMPORAL_SIZE)*sizeof(float)));
}
void fill_model(cereal::ModelData::Builder &framed, const ModelDataRaw &net_outputs) {
// Find the distribution that corresponds to the most probable plan
int plan_mhp_max_idx = 0;
for (int i=1; i<PLAN_MHP_N; i++) {
if (net_outputs.plan[(i + 1)*(PLAN_MHP_GROUP_SIZE) - 1] >
net_outputs.plan[(plan_mhp_max_idx + 1)*(PLAN_MHP_GROUP_SIZE) - 1]) {
plan_mhp_max_idx = i;
}
}
const float *best_plan = get_plan_data(net_outputs.plan);
// x pos at 10s is a good valid_len
float valid_len = 0;
float valid_len_candidate;
for (int i=1; i<TRAJECTORY_SIZE; i++) {
valid_len_candidate = net_outputs.plan[plan_mhp_max_idx*(PLAN_MHP_GROUP_SIZE) + 30*i];
if (valid_len_candidate >= valid_len){
valid_len = valid_len_candidate;
if (const float len = best_plan[30*i]; len >= valid_len){
valid_len = len;
}
}
// clamp to 10 and MODEL_PATH_DISTANCE
@ -415,47 +351,39 @@ void model_publish(PubMaster &pm, uint32_t vipc_frame_id, uint32_t frame_id,
valid_len_idx = i;
}
}
fill_path(framed.initPath(), best_plan, 1.0, valid_len, valid_len_idx, 0);
fill_path(framed.initLeftLane(), net_outputs.lane_lines, sigmoid(net_outputs.lane_lines_prob[1]), valid_len, valid_len_idx, 1);
fill_path(framed.initRightLane(), net_outputs.lane_lines, sigmoid(net_outputs.lane_lines_prob[2]), valid_len, valid_len_idx, 2);
auto lpath = framed.initPath();
fill_path(lpath, &net_outputs.plan[plan_mhp_max_idx*(PLAN_MHP_GROUP_SIZE)], valid_len, valid_len_idx);
auto left_lane = framed.initLeftLane();
int ll_idx = 1;
fill_lane_line(left_lane, net_outputs.lane_lines, ll_idx, valid_len, valid_len_idx,
sigmoid(net_outputs.lane_lines_prob[ll_idx]));
auto right_lane = framed.initRightLane();
ll_idx = 2;
fill_lane_line(right_lane, net_outputs.lane_lines, ll_idx, valid_len, valid_len_idx,
sigmoid(net_outputs.lane_lines_prob[ll_idx]));
// Find the distribution that corresponds to the current lead
int mdn_max_idx = 0;
int t_offset = 0;
for (int i=1; i<LEAD_MHP_N; i++) {
if (net_outputs.lead[(i+1)*(LEAD_MHP_GROUP_SIZE) + t_offset - 3] >
net_outputs.lead[(mdn_max_idx + 1)*(LEAD_MHP_GROUP_SIZE) + t_offset - 3]) {
mdn_max_idx = i;
}
}
fill_lead(framed.initLead(), &net_outputs.lead[mdn_max_idx*(LEAD_MHP_GROUP_SIZE)], sigmoid(net_outputs.lead_prob[t_offset]));
// Find the distribution that corresponds to the lead in 2s
mdn_max_idx = 0;
t_offset = 1;
for (int i=1; i<LEAD_MHP_N; i++) {
if (net_outputs.lead[(i+1)*(LEAD_MHP_GROUP_SIZE) + t_offset - 3] >
net_outputs.lead[(mdn_max_idx + 1)*(LEAD_MHP_GROUP_SIZE) + t_offset - 3]) {
mdn_max_idx = i;
}
}
fill_lead(framed.initLeadFuture(), &net_outputs.lead[mdn_max_idx*(LEAD_MHP_GROUP_SIZE)], sigmoid(net_outputs.lead_prob[t_offset]));
fill_lead(framed.initLead(), net_outputs.lead, net_outputs.lead_prob, 0);
fill_lead(framed.initLeadFuture(), net_outputs.lead, net_outputs.lead_prob, 1);
fill_meta(framed.initMeta(), net_outputs.meta);
}
pm.send("model", msg);
void model_publish(PubMaster &pm, uint32_t vipc_frame_id, uint32_t frame_id, float frame_drop,
const ModelDataRaw &net_outputs, const float *raw_pred, uint64_t timestamp_eof,
float model_execution_time) {
const uint32_t frame_age = (frame_id > vipc_frame_id) ? (frame_id - vipc_frame_id) : 0;
auto do_publish = [&](auto init_model_func, const char *pub_name) {
MessageBuilder msg;
auto framed = (msg.initEvent().*(init_model_func))();
framed.setFrameId(vipc_frame_id);
framed.setFrameAge(frame_age);
framed.setFrameDropPerc(frame_drop * 100);
framed.setTimestampEof(timestamp_eof);
framed.setModelExecutionTime(model_execution_time);
if (send_raw_pred) {
framed.setRawPred(kj::arrayPtr((const uint8_t *)raw_pred, (OUTPUT_SIZE + TEMPORAL_SIZE) * sizeof(float)));
}
fill_model(framed, net_outputs);
pm.send(pub_name, msg);
};
do_publish(&cereal::Event::Builder::initModel, "model");
do_publish(&cereal::Event::Builder::initModelV2, "modelV2");
}
void posenet_publish(PubMaster &pm, uint32_t vipc_frame_id, uint32_t frame_id,
uint32_t vipc_dropped_frames, float frame_drop,
void posenet_publish(PubMaster &pm, uint32_t vipc_frame_id, uint32_t vipc_dropped_frames,
const ModelDataRaw &net_outputs, uint64_t timestamp_eof) {
float trans_arr[3];
float trans_std_arr[3];

@ -1,12 +1,10 @@
#ifndef MODEL_H
#define MODEL_H
#pragma once
// gate this here
#define TEMPORAL
#define DESIRE
#define TRAFFIC_CONVENTION
#include "common/mat.h"
#include "common/util.h"
#include "common/modeldata.h"
@ -17,29 +15,9 @@
#include <memory>
#include "messaging.hpp"
#define MODEL_NAME "supercombo_dlc"
#define MODEL_WIDTH 512
#define MODEL_HEIGHT 256
#define MODEL_FRAME_SIZE MODEL_WIDTH * MODEL_HEIGHT * 3 / 2
#define DESIRE_LEN 8
#define TRAFFIC_CONVENTION_LEN 2
#define PLAN_MHP_N 5
#define PLAN_MHP_COLUMNS 30
#define PLAN_MHP_VALS 30*33
#define PLAN_MHP_SELECTION 1
#define PLAN_MHP_GROUP_SIZE (2*PLAN_MHP_VALS + PLAN_MHP_SELECTION)
#define LEAD_MHP_N 5
#define LEAD_MHP_VALS 4
#define LEAD_MHP_SELECTION 3
#define LEAD_MHP_GROUP_SIZE (2*LEAD_MHP_VALS + LEAD_MHP_SELECTION)
#define POSE_SIZE 12
#define MODEL_FREQ 20
constexpr int DESIRE_LEN = 8;
constexpr int TRAFFIC_CONVENTION_LEN = 2;
constexpr int MODEL_FREQ = 20;
struct ModelDataRaw {
float *plan;
float *lane_lines;
@ -56,33 +34,26 @@ struct ModelDataRaw {
typedef struct ModelState {
ModelFrame frame;
float *output;
float *input_frames;
RunModel *m;
std::unique_ptr<float[]> output;
std::unique_ptr<float[]> input_frames;
std::unique_ptr<RunModel> m;
cl_command_queue q;
#ifdef DESIRE
std::unique_ptr<float[]> prev_desire;
std::unique_ptr<float[]> pulse_desire;
float prev_desire[DESIRE_LEN] = {};
float pulse_desire[DESIRE_LEN] = {};
#endif
#ifdef TRAFFIC_CONVENTION
std::unique_ptr<float[]> traffic_convention;
float traffic_convention[TRAFFIC_CONVENTION_LEN] = {};
#endif
} ModelState;
void model_init(ModelState* s, cl_device_id device_id,
cl_context context, int temporal);
ModelDataRaw model_eval_frame(ModelState* s, cl_command_queue q,
cl_mem yuv_cl, int width, int height,
mat3 transform, void* sock, float *desire_in);
void model_init(ModelState* s, cl_device_id device_id, cl_context context);
ModelDataRaw model_eval_frame(ModelState* s, cl_mem yuv_cl, int width, int height,
const mat3 &transform, float *desire_in);
void model_free(ModelState* s);
void poly_fit(float *in_pts, float *in_stds, float *out);
void model_publish(PubMaster &pm, uint32_t vipc_frame_id, uint32_t frame_id,
uint32_t vipc_dropped_frames, float frame_drop, const ModelDataRaw &data, const float* raw_pred,
uint64_t timestamp_eof, float model_execution_time);
void model_publish_v2(PubMaster &pm, uint32_t vipc_frame_id, uint32_t frame_id,
uint32_t vipc_dropped_frames, float frame_drop, const ModelDataRaw &data, const float* raw_pred,
uint64_t timestamp_eof, float model_execution_time);
void posenet_publish(PubMaster &pm, uint32_t vipc_frame_id, uint32_t frame_id,
uint32_t vipc_dropped_frames, float frame_drop, const ModelDataRaw &data,
uint64_t timestamp_eof);
#endif
void model_publish(PubMaster &pm, uint32_t vipc_frame_id, uint32_t frame_id, float frame_drop,
const ModelDataRaw &net_outputs, const float *raw_pred, uint64_t timestamp_eof,
float model_execution_time);
void posenet_publish(PubMaster &pm, uint32_t vipc_frame_id, uint32_t vipc_dropped_frames,
const ModelDataRaw &net_outputs, uint64_t timestamp_eof);

Loading…
Cancel
Save