diff options
Diffstat (limited to 'nerv/examples/lmptb/rnn/tnn.lua')
| -rw-r--r-- | nerv/examples/lmptb/rnn/tnn.lua | 200 |
1 files changed, 165 insertions, 35 deletions
diff --git a/nerv/examples/lmptb/rnn/tnn.lua b/nerv/examples/lmptb/rnn/tnn.lua index 8037918..460fcc4 100644 --- a/nerv/examples/lmptb/rnn/tnn.lua +++ b/nerv/examples/lmptb/rnn/tnn.lua @@ -34,9 +34,11 @@ local function discover(id, layers, layer_repo) layer = layer, inputs_m = {}, --storage for computation, inputs_m[time][port] inputs_b = {}, --inputs_g[time][port], whether this input can been computed + inputs_p_matbak = {}, --which is a back-up space to handle some cross-border computation, inputs_p_matbak[port] outputs_m = {}, outputs_b = {}, err_inputs_m = {}, + err_inputs_p_matbak = {}, --which is a back-up space to handle some cross-border computation err_inputs_b = {}, err_outputs_m = {}, err_outputs_b = {}, @@ -57,26 +59,36 @@ nerv.TNN.FC.HAS_INPUT = 1 nerv.TNN.FC.HAS_LABEL = 2 nerv.TNN.FC.SEQ_NORM = bit.bor(nerv.TNN.FC.HAS_INPUT, nerv.TNN.FC.HAS_LABEL) --This instance have both input and label -function TNN.makeInitialStore(st, p, dim, batch_size, chunk_size, global_conf, st_c, p_c) +function TNN.makeInitialStore(st, p, dim, batch_size, chunk_size, global_conf, st_c, p_c, t_c) --Return a table of matrix storage from time (1-chunk_size)..(2*chunk_size) if (type(st) ~= "table") then nerv.error("st should be a table") end - for i = 1 - chunk_size, chunk_size * 2 do + for i = 1 - chunk_size - 1, chunk_size * 2 + 1 do --intentionally allocated more time, should be [1-chunk_size, chunk_size*2] if (st[i] == nil) then st[i] = {} end st[i][p] = global_conf.cumat_type(batch_size, dim) st[i][p]:fill(0) if (st_c ~= nil) then - if (st_c[i] == nil) then - st_c[i] = {} + if (st_c[i + t_c] == nil) then + st_c[i + t_c] = {} end - st_c[i][p_c] = st[i][p] + st_c[i + t_c][p_c] = st[i][p] end end end +function TNN:outOfFeedRange(t) --out of chunk, or no input, for the current feed + if (t < 1 or t > self.chunk_size) then + return true + end + if (self.feeds_now.flagsPack_now[t] == 0 or self.feeds_now.flagsPack_now[t] == nil) then + return true + end + return false +end + function TNN:__init(id, global_conf, layer_conf) local layers = {} local inputs_p = {} --map:port of the TDAGLayer to layer ref and port @@ -109,7 +121,7 @@ function TNN:__init(id, global_conf, layer_conf) outputs_p[port_to] = {["ref"] = ref_from, ["port"] = port_from} ref_from.outputs_m[port_from] = {} --just a place holder else - conn_now = { + local conn_now = { ["src"] = {["ref"] = ref_from, ["port"] = port_from}, ["dst"] = {["ref"] = ref_to, ["port"] = port_to}, ["time"] = time_to @@ -138,9 +150,11 @@ function TNN:init(batch_size, chunk_size) self.chunk_size = chunk_size for i, conn in ipairs(self.parsed_conns) do --init storage for connections inside the NN local _, output_dim - local ref_from, port_from, ref_to, port_to + local ref_from, port_from, ref_to, port_to, time ref_from, port_from = conn.src.ref, conn.src.port ref_to, port_to = conn.dst.ref, conn.dst.port + time = conn.time + local dim = ref_from.dim_out[port_from] if (dim == 0) then nerv.error("layer %s has a zero dim port", ref_from.layer.id) @@ -148,9 +162,9 @@ function TNN:init(batch_size, chunk_size) print("TNN initing storage", ref_from.layer.id, "->", ref_to.layer.id) ref_to.inputs_p_matbak[port_to] = self.gconf.cumat_type(batch_size, dim) - self.makeInitialStore(ref_from.outputs_m, port_from, dim, batch_size, chunk_size, self.gconf, ref_to.inputs_m, port_to) + self.makeInitialStore(ref_from.outputs_m, port_from, dim, batch_size, chunk_size, self.gconf, ref_to.inputs_m, port_to, time) ref_from.err_inputs_p_matbak[port_from] = self.gconf.cumat_type(batch_size, dim) - self.makeInitialStore(ref_from.err_inputs_m, port_from, dim, batch_size, chunk_size, self.gconf, ref_to.err_outputs_m, port_to) + self.makeInitialStore(ref_from.err_inputs_m, port_from, dim, batch_size, chunk_size, self.gconf, ref_to.err_outputs_m, port_to, time) end @@ -159,8 +173,8 @@ function TNN:init(batch_size, chunk_size) for i = 1, #self.dim_out do --Init storage for output ports local ref = self.outputs_p[i].ref local p = self.outputs_p[i].port - self.makeInitialStore(ref.outputs_m, p, self.dim_out[i], batch_size, chunk_size, self.gconf, self.outputs_m, i) - self.makeInitialStore(ref.err_inputs_m, p, self.dim_out[i], batch_size, chunk_size, self.gconf, self.err_inputs_m, i) + self.makeInitialStore(ref.outputs_m, p, self.dim_out[i], batch_size, chunk_size, self.gconf, self.outputs_m, i, 0) + self.makeInitialStore(ref.err_inputs_m, p, self.dim_out[i], batch_size, chunk_size, self.gconf, self.err_inputs_m, i, 0) end self.inputs_m = {} @@ -168,8 +182,8 @@ function TNN:init(batch_size, chunk_size) for i = 1, #self.dim_in do --Init storage for input ports local ref = self.inputs_p[i].ref local p = self.inputs_p[i].port - self.makeInitialStore(ref.inputs_m, p, self.dim_in[i], batch_size, chunk_size, self.gconf, self.inputs_m, i) - self.makeInitialStore(ref.err_outputs_m, p, self.dim_in[i], batch_size, chunk_size, self.gconf, self.err_outputs_m, i) + self.makeInitialStore(ref.inputs_m, p, self.dim_in[i], batch_size, chunk_size, self.gconf, self.inputs_m, i, 0) + self.makeInitialStore(ref.err_outputs_m, p, self.dim_in[i], batch_size, chunk_size, self.gconf, self.err_outputs_m, i, 0) end for id, ref in pairs(self.layers) do --Calling init for child layers @@ -274,45 +288,161 @@ function TNN:getFeedFromReader(reader) return got_new, feeds_now end +function TNN:moveRightToNextMB() --move output history activations of 1..chunk_size to 1-chunk_size..0 + for t = self.chunk_size, 1, -1 do + for id, ref in pairs(self.layers) do + for p = 1, #ref.dim_out do + ref.outputs_m[t - self.chunk_size][p]:copy_fromd(ref.outputs_m[t][p]) + end + end + end +end + function TNN:net_propagate() --propagate according to feeds_now + for t = 1, self.chunk_size, 1 do + for id, ref in pairs(self.layers) do + for p = 1, #ref.dim_out do + ref.outputs_b[t][p] = false + end + for p = 1, #ref.dim_in do + ref.inputs_b[t][p] = false + end + end + end + local feeds_now = self.feeds_now - for t = 1, chunk_size do + for t = 1, self.chunk_size do if (bit.band(feeds_now.flagsPack_now[t], nerv.TNN.FC.HAS_INPUT) > 0) then for i = 1, #self.dim_in do - local ref = inputs_p[i].ref - local p = inputs_p[i].port + local ref = self.inputs_p[i].ref + local p = self.inputs_p[i].port ref.inputs_b[t][p] = true + self:propagate_dfs(ref, t) end - --TODO end end end -function DAGLayer:update(bp_err, input, output) - self:set_err_inputs(bp_err) - self:set_inputs(input) - self:set_outputs(output) - -- print("update") - for id, ref in pairs(self.queue) do - -- print(ref.layer.id) - ref.layer:update(ref.err_inputs, ref.inputs, ref.outputs) +--ref: the TNN_ref of a layer +--t: the current time to propagate +function TNN:propagate_dfs(ref, t) + if (self:outOfFeedRange(t)) then + return + end + if (ref.outputs_b[t][1] == true) then --already propagated, 1 is just a random port + return + end + + --print("debug dfs", ref.layer.id, t) + + local flag = true --whether have all inputs + for _, conn in pairs(ref.conns_i) do + local p = conn.dst.port + if (not (ref.inputs_b[t][p] or self:outOfFeedRange(t - conn.time))) then + flag = false + break + end + end + if (flag == false) then + return + end + + --ok, do propagate + --print("debug ok, propagating"); + ref.layer:propagate(ref.inputs_m[t], ref.outputs_m[t]) + for i = 1, #ref.dim_out do + if (ref.outputs_b[t][i] == true) then + nerv.error("this time's outputs_b should be false") + end + ref.outputs_b[t][i] = true + end + + --try dfs for further layers + for _, conn in pairs(ref.conns_o) do + --print("debug dfs-searching", conn.dst.ref.layer.id) + conn.dst.ref.inputs_b[t + conn.time][conn.dst.port] = true + self:propagate_dfs(conn.dst.ref, t + conn.time) end end -function DAGLayer:back_propagate(bp_err, next_bp_err, input, output) - self:set_err_outputs(next_bp_err) - self:set_err_inputs(bp_err) - self:set_inputs(input) - self:set_outputs(output) - for i = #self.queue, 1, -1 do - local ref = self.queue[i] - -- print(ref.layer.id) - ref.layer:back_propagate(ref.err_inputs, ref.err_outputs, ref.inputs, ref.outputs) +--do_update: bool, whether we are doing back-propagate or updating the parameters +function TNN:net_backpropagate(do_update) --propagate according to feeds_now + if (do_update == nil) then + nerv.error("do_update should not be nil") + end + for t = 1, self.chunk_size, 1 do + for id, ref in pairs(self.layers) do + for p = 1, #ref.dim_out do + ref.err_inputs_b[t][p] = false + end + for p = 1, #ref.dim_in do + ref.err_outputs_b[t][p] = false + end + end + end + + local feeds_now = self.feeds_now + for t = 1, self.chunk_size do + if (bit.band(feeds_now.flagsPack_now[t], nerv.TNN.FC.HAS_LABEL) > 0) then + for i = 1, #self.dim_out do + local ref = self.outputs_p[i].ref + local p = self.outputs_p[i].port + ref.err_inputs_b[t][p] = true + self:backpropagate_dfs(ref, t, do_update) + end + end + end +end + +--ref: the TNN_ref of a layer +--t: the current time to propagate +function TNN:backpropagate_dfs(ref, t, do_update) + if (self:outOfFeedRange(t)) then + return + end + if (ref.err_outputs_b[t][1] == true) then --already back_propagated, 1 is just a random port + return + end + + --print("debug dfs", ref.layer.id, t) + + local flag = true --whether have all inputs + for _, conn in pairs(ref.conns_o) do + local p = conn.src.port + if (not (ref.err_inputs_b[t][p] or self:outOfFeedRange(t + conn.time))) then + flag = false + break + end + end + if (flag == false) then + return + end + + --ok, do back_propagate + --print("debug ok, back-propagating(or updating)") + if (do_update == false) then + ref.layer:back_propagate(ref.err_inputs_m[t], ref.err_outputs_m[t], ref.inputs_m[t], ref.outputs_m[t]) + else + --print(ref.err_inputs_m[t][1]) + ref.layer:update(ref.err_inputs_m[t], ref.inputs_m[t], ref.outputs_m[t]) + end + for i = 1, #ref.dim_in do + if (ref.err_outputs_b[t][i] == true) then + nerv.error("this time's outputs_b should be false") + end + ref.err_outputs_b[t][i] = true + end + + --try dfs for further layers + for _, conn in pairs(ref.conns_i) do + --print("debug dfs-searching", conn.src.ref.layer.id) + conn.src.ref.err_inputs_b[t - conn.time][conn.src.port] = true + self:backpropagate_dfs(conn.src.ref, t - conn.time, do_update) end end --Return: nerv.ParamRepo -function DAGLayer:get_params() +function TNN:get_params() local param_repos = {} for id, ref in pairs(self.queue) do table.insert(param_repos, ref.layer:get_params()) |