function build_trainer(ifname) local param_repo = nerv.ParamRepo() param_repo:import(ifname, nil, gconf) local sublayer_repo = make_sublayer_repo(param_repo) local layer_repo = make_layer_repo(sublayer_repo, param_repo) local crit = get_criterion_layer(sublayer_repo) local network = get_network(layer_repo) local input_order = get_input_order() local iterative_trainer = function (prefix, scp_file, bp) gconf.randomize = bp -- build buffer local buffer = make_buffer(make_readers(scp_file, layer_repo)) -- initialize the network network:init(gconf.batch_size) gconf.cnt = 0 err_input = {nerv.CuMatrixFloat(256, 1)} err_input[1]:fill(1) for data in buffer.get_data, buffer do -- prine stat periodically gconf.cnt = gconf.cnt + 1 if gconf.cnt == 1000 then print_stat(sublayer_repo) nerv.CuMatrix.print_profile() nerv.CuMatrix.clear_profile() gconf.cnt = 0 -- break end local input = {} -- if gconf.cnt == 100 then break end for i, id in ipairs(input_order) do if data[id] == nil then nerv.error("input data %s not found", id) end table.insert(input, data[id]) end local output = {nerv.CuMatrixFloat(256, 1)} err_output = {input[1]:create()} network:propagate(input, output) if bp then network:back_propagate(err_input, err_output, input, output) network:update(err_input, input, output) end -- collect garbage in-time to save GPU memory collectgarbage("collect") end print_stat(sublayer_repo) nerv.CuMatrix.print_profile() nerv.CuMatrix.clear_profile() if (not bp) and prefix ~= nil then nerv.info("writing back...") local fname = string.format("%s_cv%.3f.nerv", prefix, get_accuracy(sublayer_repo)) network:get_params():export(fname, nil) end return get_accuracy(sublayer_repo) end return iterative_trainer end dofile(arg[1]) start_halving_inc = 0.5 halving_factor = 0.6 end_halving_inc = 0.1 min_iter = 1 max_iter = 20 min_halving = 5 gconf.batch_size = 256 gconf.buffer_size = 81920 local pf0 = gconf.initialized_param local trainer = build_trainer(pf0) --local trainer = build_trainer("c3.nerv") local accu_best = trainer(nil, gconf.cv_scp, false) local do_halving = false nerv.info("initial cross validation: %.3f", accu_best) for i = 1, max_iter do nerv.info("[NN] begin iteration %d with lrate = %.6f", i, gconf.lrate) local accu_tr = trainer(nil, gconf.tr_scp, true) nerv.info("[TR] training set %d: %.3f", i, accu_tr) local accu_new = trainer( string.format("%s_%s_iter_%d_lr%f_tr%.3f", string.gsub( (string.gsub(pf0[1], "(.*/)(.*)", "%2")), "(.*)%..*", "%1"), os.date("%Y%m%d%H%M%S"), i, gconf.lrate, accu_tr), gconf.cv_scp, false) nerv.info("[CV] cross validation %d: %.3f", i, accu_new) -- TODO: revert the weights local accu_diff = accu_new - accu_best if do_halving and accu_diff < end_halving_inc and i > min_iter then break end if accu_diff < start_halving_inc and i >= min_halving then do_halving = true end if do_halving then gconf.lrate = gconf.lrate * halving_factor end if accu_new > accu_best then accu_best = accu_new end -- nerv.Matrix.print_profile() end