aboutsummaryrefslogtreecommitdiff
path: root/nerv/layer/lstm_gate.lua
diff options
context:
space:
mode:
Diffstat (limited to 'nerv/layer/lstm_gate.lua')
-rw-r--r--nerv/layer/lstm_gate.lua97
1 files changed, 0 insertions, 97 deletions
diff --git a/nerv/layer/lstm_gate.lua b/nerv/layer/lstm_gate.lua
deleted file mode 100644
index 39a3ff7..0000000
--- a/nerv/layer/lstm_gate.lua
+++ /dev/null
@@ -1,97 +0,0 @@
-local LSTMGateLayer = nerv.class('nerv.LSTMGateLayer', 'nerv.Layer')
--- NOTE: this is a full matrix gate
-
-function LSTMGateLayer:__init(id, global_conf, layer_conf)
- nerv.Layer.__init(self, id, global_conf, layer_conf)
- self.param_type = layer_conf.param_type
- self:check_dim_len(-1, 1) --accept multiple inputs
- self:bind_params()
-end
-
-function LSTMGateLayer:bind_params()
- local lconf = self.lconf
- lconf.no_update_ltp1 = lconf.no_update_ltp1 or lconf.no_update_ltp
- for i = 1, #self.dim_in do
- local pid = "ltp" .. i
- local pid_list = i == 1 and {pid, "ltp"} or pid
- self["ltp" .. i] = self:find_param(pid_list, lconf, self.gconf,
- nerv.LinearTransParam,
- {self.dim_in[i], self.dim_out[1]})
- if self.param_type[i] == 'D' then
- self["ltp" .. i].trans:diagonalize()
- end
- local no_update = lconf["no_update_ltp" .. i]
- if (no_update ~= nil) and no_update or lconf.no_update_all then
- self["ltp" .. i].no_update = true
- end
- end
- self.ltp = self.ltp1 -- alias of ltp1
- self.bp = self:find_param("bp", lconf, self.gconf,
- nerv.BiasParam, {1, self.dim_out[1]},
- nerv.Param.gen_zero)
- local no_update = lconf["no_update_bp"]
- if (no_update ~= nil) and no_update or lconf.no_update_all then
- self.bp.no_update = true
- end
-end
-
-function LSTMGateLayer:init(batch_size)
- if self.dim_out[1] ~= self.bp.trans:ncol() then
- nerv.error("mismatching dimensions of linear transform and bias paramter")
- end
- for i = 1, #self.dim_in do
- if self.dim_in[i] ~= self["ltp" .. i].trans:nrow() then
- nerv.error("mismatching dimensions of linear transform parameter and input")
- end
- if self.dim_out[1] ~= self["ltp" .. i].trans:ncol() then
- nerv.error("mismatching dimensions of linear transform parameter and output")
- end
- self["ltp" .. i]:train_init()
- end
- self.bp:train_init()
- self.err_bakm = self.mat_type(batch_size, self.dim_out[1])
-end
-
-function LSTMGateLayer:batch_resize(batch_size)
- if self.err_m:nrow() ~= batch_size then
- self.err_bakm = self.mat_type(batch_size, self.dim_out[1])
- end
-end
-
-function LSTMGateLayer:propagate(input, output)
- -- apply linear transform
- output[1]:mul(input[1], self.ltp1.trans, 1.0, 0.0, 'N', 'N')
- for i = 2, #self.dim_in do
- output[1]:mul(input[i], self["ltp" .. i].trans, 1.0, 1.0, 'N', 'N')
- end
- -- add bias
- output[1]:add_row(self.bp.trans, 1.0)
- output[1]:sigmoid(output[1])
-end
-
-function LSTMGateLayer:back_propagate(bp_err, next_bp_err, input, output)
- self.err_bakm:sigmoid_grad(bp_err[1], output[1])
- for i = 1, #self.dim_in do
- next_bp_err[i]:mul(self.err_bakm, self["ltp" .. i].trans, 1.0, 0.0, 'N', 'T')
- self["ltp" .. i]:back_propagate_by_err_input(self.err_bakm, input[i])
- end
- self.bp:back_propagate_by_gradient(self.err_bakm:colsum())
-end
-
-function LSTMGateLayer:update()
- for i = 1, #self.dim_in do
- self["ltp" .. i]:update_by_err_input()
- if self.param_type[i] == 'D' then
- self["ltp" .. i].trans:diagonalize()
- end
- end
- self.bp:update_by_gradient()
-end
-
-function LSTMGateLayer:get_params()
- local pr = nerv.ParamRepo({self.bp}, self.loc_type)
- for i = 1, #self.dim_in do
- pr:add(self["ltp" .. i])
- end
- return pr
-end