tactic clear_tactic(name const & n) {
auto fn = [=](environment const &, io_state const &, proof_state const & _s) -> optional<proof_state> {
if (!_s.get_goals()) {
throw_no_goal_if_enabled(_s);
return none_proof_state();
}
proof_state s = apply_substitution(_s);
goals const & gs = s.get_goals();
goal g = head(gs);
goals tail_gs = tail(gs);
if (auto p = g.find_hyp(n)) {
expr const & h = p->first;
unsigned i = p->second;
buffer<expr> hyps;
g.get_hyps(hyps);
hyps.erase(hyps.size() - i - 1);
if (depends_on(g.get_type(), h)) {
throw_tactic_exception_if_enabled(s, sstream() << "invalid 'clear' tactic, conclusion depends on '"
<< n << "'");
return none_proof_state();
}
if (auto h2 = depends_on(i, hyps.end() - i, h)) {
throw_tactic_exception_if_enabled(s, sstream() << "invalid 'clear' tactic, hypothesis '" << *h2
<< "' depends on '" << n << "'");
return none_proof_state();
}
name_generator ngen = s.get_ngen();
expr new_type = g.get_type();
expr new_meta = mk_app(mk_metavar(ngen.next(), Pi(hyps, new_type)), hyps);
goal new_g(new_meta, new_type);
substitution new_subst = s.get_subst();
assign(new_subst, g, new_meta);
proof_state new_s(s, goals(new_g, tail_gs), new_subst, ngen);
return some_proof_state(new_s);
} else {
throw_tactic_exception_if_enabled(s, sstream() << "invalid 'clear' tactic, goal does not have a hypothesis "
<< " named '" << n << "'");
return none_proof_state();
}
};
return tactic01(fn);
}
tactic change_goal_tactic(elaborate_fn const & elab, expr const & e) {
return tactic([=](environment const & env, io_state const & ios, proof_state const & s) {
proof_state new_s = s;
goals const & gs = new_s.get_goals();
if (!gs) {
throw_no_goal_if_enabled(s);
return proof_state_seq();
}
expr t = head(gs).get_type();
bool report_unassigned = true;
if (auto new_e = elaborate_with_respect_to(env, ios, elab, new_s, e, none_expr(), report_unassigned)) {
goals const & gs = new_s.get_goals();
goal const & g = head(gs);
substitution subst = new_s.get_subst();
auto tc = mk_type_checker(env);
constraint_seq cs;
if (tc->is_def_eq(t, *new_e, justification(), cs)) {
if (cs) {
unifier_config cfg(ios.get_options());
buffer<constraint> cs_buf;
cs.linearize(cs_buf);
to_buffer(new_s.get_postponed(), cs_buf);
unify_result_seq rseq = unify(env, cs_buf.size(), cs_buf.data(), subst, cfg);
return map2<proof_state>(rseq, [=](pair<substitution, constraints> const & p) -> proof_state {
substitution const & subst = p.first;
constraints const & postponed = p.second;
substitution new_subst = subst;
expr final_e = new_subst.instantiate_all(*new_e);
expr M = g.mk_meta(mk_fresh_name(), final_e);
goal new_g(M, final_e);
assign(new_subst, g, M);
return proof_state(new_s, cons(new_g, tail(gs)), new_subst, postponed);
});
}
expr M = g.mk_meta(mk_fresh_name(), *new_e);
goal new_g(M, *new_e);
assign(subst, g, M);
return proof_state_seq(proof_state(new_s, cons(new_g, tail(gs)), subst));
} else {
throw_tactic_exception_if_enabled(new_s, [=](formatter const & fmt) {
format r = format("invalid 'change' tactic, the given type");
r += pp_indent_expr(fmt, *new_e);
r += compose(line(), format("does not match the goal type"));
r += pp_indent_expr(fmt, t);
return r;
});
return proof_state_seq();
}
}
return proof_state_seq();
});
}
tactic revert_tactic(name const & n) {
auto fn = [=](environment const &, io_state const &, proof_state const & s) -> optional<proof_state> {
goals const & gs = s.get_goals();
if (empty(gs)) {
throw_no_goal_if_enabled(s);
return none_proof_state();
}
goal g = head(gs);
goals tail_gs = tail(gs);
if (auto p = g.find_hyp(n)) {
expr const & h = p->first;
unsigned i = p->second;
buffer<expr> hyps;
g.get_hyps(hyps);
hyps.erase(hyps.size() - i - 1);
if (optional<expr> other_h = depends_on(i, hyps.end() - i, h)) {
throw_tactic_exception_if_enabled(s, sstream() << "invalid 'revert' tactic, hypothesis '" << local_pp_name(*other_h)
<< "' depends on '" << local_pp_name(h) << "'");
return none_proof_state(); // other hypotheses depend on h
}
name_generator ngen = s.get_ngen();
expr new_type = Pi(h, g.get_type());
expr new_meta = mk_app(mk_metavar(ngen.next(), Pi(hyps, new_type)), hyps);
goal new_g(new_meta, new_type);
substitution new_subst = s.get_subst();
assign(new_subst, g, mk_app(new_meta, h));
proof_state new_s(s, goals(new_g, tail_gs), new_subst, ngen);
return some_proof_state(new_s);
} else {
throw_tactic_exception_if_enabled(s, sstream() << "invalid 'revert' tactic, unknown hypothesis '" << n << "'");
return none_proof_state();
}
};
return tactic01(fn);
}
tactic intros_tactic(list<name> _ns, bool relax_main_opaque) {
auto fn = [=](environment const & env, io_state const &, proof_state const & s) {
list<name> ns = _ns;
goals const & gs = s.get_goals();
if (empty(gs)) {
throw_no_goal_if_enabled(s);
return optional<proof_state>();
}
goal const & g = head(gs);
name_generator ngen = s.get_ngen();
auto tc = mk_type_checker(env, ngen.mk_child(), relax_main_opaque);
expr t = g.get_type();
expr m = g.get_meta();
bool gen_names = empty(ns);
try {
while (true) {
if (!gen_names && is_nil(ns))
break;
if (!is_pi(t)) {
if (!is_nil(ns)) {
t = tc->ensure_pi(t).first;
} else {
expr new_t = tc->whnf(t).first;
if (!is_pi(new_t))
break;
t = new_t;
}
}
name new_name;
if (!is_nil(ns)) {
new_name = head(ns);
ns = tail(ns);
} else {
new_name = get_unused_name(binding_name(t), m);
}
expr new_local = mk_local(ngen.next(), new_name, binding_domain(t), binding_info(t));
t = instantiate(binding_body(t), new_local);
m = mk_app(m, new_local);
}
goal new_g(m, t);
return some(proof_state(s, goals(new_g, tail(gs)), ngen));
} catch (exception &) {
return optional<proof_state>();
}
};
return tactic01(fn);
}
double HDP_MEDOIDS (vector<Instance*>& data, vector< vector<double> >& means, Lookups* tables, vector<double> lambdas, dist_func df, int FIX_DIM) {
// STEP ZERO: validate input and initialization
int N = tables->nWords;
int D = tables->nDocs;
vector< pair<int, int> > doc_lookup = *(tables->doc_lookup);
double lambda_global = lambdas[0];
double lambda_local = lambdas[1];
vector< vector<double> > global_means (1, vector<double>(FIX_DIM, 0.0));
vector< vector<int> > k (D, vector<int>(1,0)); // global association
vector<int> z (N, 0); // local assignment
vector<int> global_asgn (N, 0); // global assignment
// STEP ONE: a. set initial *global* medoid as global mean
compute_assignment (global_asgn, k, z, tables);
compute_means (data, global_asgn, FIX_DIM, global_means);
double last_cost = compute_cost (data, global_means, k, z, lambdas, tables, df, FIX_DIM);
double new_cost = last_cost;
while (true) {
// 4. for each point x_ij,
for (int j = 0; j < D; j ++) {
for (int i = doc_lookup[j].first; i < doc_lookup[j].second; i++) {
int num_global_means = global_means.size();
vector<double> d_ij (num_global_means, 0.0);
for (int p = 0; p < num_global_means; p ++) {
Instance* temp_ins = vec2ins(global_means[p]);
double euc_dist = df(data[i], temp_ins, FIX_DIM);
d_ij[p] = euc_dist * euc_dist;
delete temp_ins;
}
set<int> temp;
for (int p = 0; p < num_global_means; p ++) temp.insert(p);
int num_local_means = k[j].size();
for (int q = 0; q < num_local_means; q ++) temp.erase(k[j][q]);
set<int>::iterator it;
for (it=temp.begin(); it!=temp.end();++it) d_ij[*it] += lambda_local;
int min_p = -1; double min_dij = INF;
for (int p = 0; p < num_global_means; p ++)
if (d_ij[p] < min_dij) {
min_p = p;
min_dij = d_ij[p];
}
if (min_dij > lambda_global + lambda_local) {
z[i] = num_local_means;
k[j].push_back(num_global_means);
vector<double> new_g(FIX_DIM, 0.0);
for (int f = 0; f < data[i]->fea.size(); f++)
new_g[data[i]->fea[f].first-1] = data[i]->fea[f].second;
global_means.push_back(new_g);
// cout << "global and local increment" << endl;
} else {
bool c_exist = false;
for (int c = 0; c < num_local_means; c ++)
if (k[j][c] == min_p) {
z[i] = c;
c_exist = true;
break;
}
if (!c_exist) {
z[i] = num_local_means;
k[j].push_back(min_p);
// cout << "local increment" << endl;
}
}
}
}
/*
cout << "half..........." << endl;
cout << "#global created: " << global_means.size()
<< ", #global used: " << get_num_global_means(k);
*/
new_cost = compute_cost (data, global_means, k, z, lambdas, tables, df, FIX_DIM);
// 5. for all local clusters,
for (int j = 0; j < D; j ++) {
int begin_i = doc_lookup[j].first;
int end_i = doc_lookup[j].second;
int doc_len = doc_lookup[j].second - doc_lookup[j].first;
int num_local_means = k[j].size();
// all local clusters are distinct to each other
/*
set<int> temp;
for (int y = 0; y < num_local_means; y++)
temp.insert(k[j][y]);
cout << temp.size() << " ==? " << num_local_means << endl;
assert (temp.size() == num_local_means);
*/
// compute means of local clusters
vector< vector<double> > local_means (num_local_means, vector<double>(FIX_DIM, 0.0));
vector<int> local_asgn (z.begin()+begin_i, z.begin()+end_i);
vector<Instance*> local_data (data.begin()+begin_i,data.begin()+end_i);
compute_means (local_data, local_asgn, FIX_DIM, local_means);
assert (num_local_means == local_means.size());
// pre-compute instances for global means
int num_global_means = global_means.size();
vector<Instance*> temp_global_means (num_global_means, NULL);
for (int p = 0; p < num_global_means; p ++)
temp_global_means[p] = vec2ins (global_means[p]);
// pre-compute instances for local means
vector<Instance*> temp_local_means (num_local_means, NULL);
for (int c = 0; c < num_local_means; c ++)
temp_local_means[c] = vec2ins (local_means[c]);
for (int c = 0; c < num_local_means; c++) {
// compute distance of local clusters to each global cluster
num_global_means = global_means.size();
vector<double> d_jcp (num_global_means, 0.0);
double sum_d_ijc = 0.0;
for (int i = doc_lookup[j].first; i < doc_lookup[j].second; i ++) {
if (z[i] != c) continue;
double local_dist = df (data[i], temp_local_means[c], FIX_DIM);
sum_d_ijc += local_dist * local_dist;
for (int p = 0; p < num_global_means; p ++) {
double dist = df (data[i], temp_global_means[p], FIX_DIM);
d_jcp[p] += dist * dist;
}
}
int min_p = -1; double min_d_jcp = INF;
for (int p = 0; p < num_global_means; p ++)
if (d_jcp[p] < min_d_jcp) {
min_p = p;
min_d_jcp = d_jcp[p];
}
assert (min_p >= 0);
// cout << min_d_jcp << " " << lambda_global << " " << sum_d_ijc << endl;
if (min_d_jcp > lambda_global + sum_d_ijc) {
global_means.push_back(local_means[c]); // push mu_jc
temp_global_means.push_back(vec2ins (local_means[c]));
k[j][c] = num_global_means;
// cout << "global increment" << endl;
} else {
k[j][c] = min_p;
}
}
for (int c = 0; c < num_local_means; c ++)
delete temp_local_means[c];
num_global_means = global_means.size();
for (int p = 0; p < num_global_means; p ++)
delete temp_global_means[p];
}
// 6. for each global clusters,
compute_assignment (global_asgn, k, z, tables);
/*
cout << "compute global means.." << endl;
cout << "#global created: " << global_means.size()
<< ", #global used: " << get_num_global_means(k);
*/
compute_means (data, global_asgn, FIX_DIM, global_means);
// 7. convergence?
new_cost = compute_cost (data, global_means, k, z, lambdas, tables, df, FIX_DIM);
if ( new_cost < objmin ) objmin = new_cost;
objmin_trace << omp_get_wtime()-start_time << " " << objmin << endl;
if (new_cost == last_cost)
break;
if (new_cost < last_cost) {
last_cost = new_cost;
} else {
cerr << "failure" << endl;
return INF;
assert(false);
}
}
means = global_means;
return last_cost;
}// entry main function
