Remove inline-aliases-convolution concept (#1303)

Author: clinssen

doc/nestml_language/nestml_language_concepts.rst

@@ -1000,21 +1000,6 @@ The ``recordable`` keyword can be used to make the variable in inline expression
        ...
        recordable inline V_m mV = V_rel + E_L
 
-During simulation, one or more state variables are used to maintain the dynamical state of each convolution across time. To be able to reference these variables from within the model, a special case occurs when an inline expression is defined as a convolution and marked ``recordable``:
-
-.. code-block:: nestml
-
-   recordable inline I_syn pA = convolve(alpha_kernel, spiking_input_port) * pA
-
-Then, the state variables corresponding to this convolution can be referenced in the rest of the model, for instance:
-
-.. code-block:: nestml
-
-   update:
-     # reset the state of synaptic integration
-     I_syn = 0 pA
-     I_syn' = 0 * s**-1
-
 
 Kernel functions
 ~~~~~~~~~~~~~~~~

doc/nestml_language/neurons_in_nestml.rst

@@ -138,49 +138,6 @@ The incoming spikes could have been equivalently handled with an ``onReceive`` e
 Note that in this example, the intended physical unit (pA) was assigned by multiplying the type of the input port ``spikes`` (which is 1/s) by pA·s, resulting in a unit of pA for ``I_syn``.
 
 
-(Re)setting synaptic integration state
-^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-
-When convolutions are used, additional state variables are required for each pair *(shape, spike input port)* that appears as the parameters in a convolution. These variables track the dynamical state of that kernel, for that input port. The number of variables created corresponds to the dimensionality of the kernel. For example, in the code block above, the one-dimensional kernel ``G`` is used in a convolution with spiking input port ``spikes``. During code generation, a new state variable called ``G__conv__spikes`` is created for this combination, by joining together the name of the kernel with the name of the spike buffer using (by default) the string “__conv__”. If the same kernel is used later in a convolution with another spiking input port, say ``spikes_GABA``, then the resulting generated variable would be called ``G__conv__spikes_GABA``, allowing independent synaptic integration between input ports but allowing the same kernel to be used more than once.
-
-The process of generating extra state variables for keeping track of convolution state is normally hidden from the user. For some models, however, it might be required to set or reset the state of synaptic integration, which is stored in these internally generated variables. For example, we might want to set the synaptic current (and its rate of change) to 0 when firing a dendritic action potential. Although we would like to set the generated variable ``G__conv__spikes`` to 0 in the running example, a variable by this name is only generated during code generation, and does not exist in the namespace of the NESTML model to begin with. To still allow referring to this state in the context of the model, it is recommended to use an inline expression, with only a convolution on the right-hand side.
-
-For example, suppose we define:
-
-.. code-block:: nestml
-
-   inline g_dend pA = convolve(G, spikes)
-
-Then the name ``g_dend`` can be used as a target for assignment:
-
-.. code-block:: nestml
-
-   update:
-       g_dend = 42 pA
-
-This also works for higher-order kernels, e.g. for the second-order alpha kernel :math:`H(t)`:
-
-.. code-block:: nestml
-
-   kernel H'' = (-2/tau_syn) * H' - 1/tau_syn**2) * H
-
-We can define an inline expression with the same port as before, ``spikes``:
-
-.. code-block:: nestml
-
-   inline h_dend pA = convolve(H, spikes)
-
-The name ``h_dend`` now acts as an alias for this particular convolution. We can now assign to the inline defined variable up to the order of the kernel:
-
-.. code-block:: nestml
-
-   update:
-       h_dend = 42 pA
-       h_dend' = 10 pA/ms
-
-For more information, see the :doc:`Active dendrite tutorial </tutorials/active_dendrite/nestml_active_dendrite_tutorial>`.
-
-
 Multiple input ports
 ^^^^^^^^^^^^^^^^^^^^
 

doc/tutorials/active_dendrite/nestml_active_dendrite_tutorial.ipynb

@@ -74,8 +74,8 @@ model iaf_psc_exp_nonlineardendrite_neuron:
         t_ref ms = 10 ms                                # refractory period
         ref_timeout_ticks integer = steps(t_ref)
 
-        I_dend_incr pA/ms = pA * exp(1) / tau_syn2
-
+    internals:
+        unit_psr pA/ms = pA * exp(1) / tau_syn2         # Unitary postsynaptic response amplitude
 
     input:
         I_1 <- spike
@@ -86,7 +86,7 @@ model iaf_psc_exp_nonlineardendrite_neuron:
         spike
 
     onReceive(I_2):
-        I_dend$ += I_2 * s * I_dend_incr
+        I_dend$ += unit_psr * I_2 * s
 
     update:
         # solve ODEs
@@ -126,6 +126,7 @@ model iaf_psc_exp_nonlineardendrite_neuron:
                 V_m = V_reset
                 dAP_counts = 0
                 I_dend = 0 pA
+                I_dend$ = 0 pA/s
                 active_dendrite = false
                 active_dendrite_readout = 0.
         else:
@@ -135,4 +136,5 @@ model iaf_psc_exp_nonlineardendrite_neuron:
             active_dendrite_readout = 0.
             dAP_counts = 0
             I_dend = 0 pA
+            I_dend$ = 0 pA/s
 

doc/tutorials/sequence_learning/iaf_psc_exp_nonlineardendrite_neuron.nestml

@@ -431,8 +431,6 @@
     "    p[\"soma_params\"][\"tau_dAP\"] = 60.       # time window over which the dendritic current clamp is active\n",
     "    p[\"soma_params\"][\"theta_dAP\"] = 59.        # current threshold for a dendritic action potential\n",
     "    \n",
-    "    p[\"soma_params\"][\"I_dend_incr\"] = 2.71 / (p[\"soma_params\"][\"tau_syn2\"])\n",
-    "    \n",
     "    p[\"fixed_somatic_delay\"] = 2          # this is an approximate time of how long it takes the soma to fire\n",
     "                                          # upon receiving an external stimulus \n",
     "    \n",
@@ -892,8 +890,6 @@
     "    p[\"fixed_somatic_delay\"] = 2          # this is an approximate time of how long it takes the soma to fire\n",
     "                                          # upon receiving an external stimulus \n",
     "    \n",
-    "    p[\"soma_params\"][\"I_dend_incr\"] = 2.71 / (p[\"soma_params\"][\"tau_syn2\"])\n",
-    "\n",
     "    # synaptic parameters\n",
     "    p[\"J_EX_psp\"] = 1.1 * p[\"soma_params\"][\"V_th\"]     # somatic PSP as a response to an external input\n",
     "    p[\"convergence\"] = 5\n",
@@ -2410,8 +2406,6 @@
     "    p[\"soma_params\"][\"tau_dAP\"] = 60.       # time window over which the dendritic current clamp is active\n",
     "    p[\"soma_params\"][\"theta_dAP\"] = 59.        # current threshold for a dendritic action potential\n",
     "    \n",
-    "    p[\"soma_params\"][\"I_dend_incr\"] = 2.71 / (p[\"soma_params\"][\"tau_syn2\"])\n",
-    "    \n",
     "    p[\"fixed_somatic_delay\"] = 2          # this is an approximate time of how long it takes the soma to fire\n",
     "                                          # upon receiving an external stimulus \n",
     "        \n",
@@ -2623,7 +2617,7 @@
       "PSC maximum J_EX:  4112.209148 pA\n",
       "PSC maximum J_IE:  581.197349 pA\n",
       "PSC maximum J_EI:  -12915.496650 pA\n",
-      "Model parameters: {'dt': 0.1, 'print_simulation_progress': False, 'soma_model': 'iaf_psc_exp_nonlineardendrite_neuron_nestml__with_stdsp_synapse_nestml', 'soma_params': {'C_m': 250.0, 'E_L': 0.0, 'V_m': 0.0, 'V_reset': 0.0, 'V_th': 20.0, 't_ref': 10.0, 'tau_m': 10.0, 'tau_syn1': 2.0, 'tau_syn2': 5.0, 'tau_syn3': 1.0, 'I_p': 200.0, 'tau_dAP': 60.0, 'theta_dAP': 59.0, 'I_dend_incr': 0.542}, 'fixed_somatic_delay': 2, 'M': 6, 'n_E': 150, 'n_I': 1, 'L': 1, 'pattern_size': 20, 'inhibit_model': 'iaf_psc_exp', 'inhibit_params': {'C_m': 250.0, 'E_L': 0.0, 'I_e': 0.0, 'V_m': 0.0, 'V_reset': 0.0, 'V_th': 15.0, 't_ref': 2.0, 'tau_m': 5.0, 'tau_syn_ex': 0.5, 'tau_syn_in': 1.65}, 'J_EX_psp': 22.0, 'J_IE_psp': 0.9, 'J_EI_psp': -40.0, 'convergence': 5, 'rule': 'fixed_indegree', 'connection_prob': 0.2, 'syn_dict_ee': {'weight': 0.01, 'synapse_model': 'stdsp_synapse_nestml__with_iaf_psc_exp_nonlineardendrite_neuron_nestml', 'permanence_threshold': 10.0, 'tau_pre_trace': 20.0, 'delay': 2.0, 'receptor_type': 2, 'lambda_plus': 0.08, 'zt': 1.0, 'lambda_h': 0.014, 'Wmax': 12.98, 'permanence_max': 20.0, 'permanence_min': 1.0, 'lambda_minus': 0.0015, 'dt_min': 4.0, 'permanence': <nest.lib.hl_api_types.Parameter object at 0x7f4733fb5450>, 'dt_max': 80.0}, 'permanence_min': 0.0, 'permanence_max': 8.0, 'calibration': 0.0, 'conn_dict_ex': {'rule': 'all_to_all'}, 'syn_dict_ex': {'receptor_type': 1, 'delay': 0.1, 'weight': 4112.209148358356}, 'conn_dict_edx': {'rule': 'fixed_outdegree', 'outdegree': 21}, 'syn_dict_edx': {'receptor_type': 2, 'delay': 0.1, 'weight': 82.6}, 'syn_dict_ie': {'synapse_model': 'static_synapse', 'delay': 0.1, 'weight': 581.1973492566976}, 'syn_dict_ei': {'synapse_model': 'static_synapse', 'delay': 0.1, 'receptor_type': 3, 'weight': -12915.496650148836}, 'DeltaT': 40.0, 'excitation_start': 30.0, 'time_dend_to_somatic': 20.0, 'DeltaT_cue': 80.0, 'overwrite_files': True, 'seed': 111, 'pad_time': 5.0, 'idend_recording_interval': 1.0, 'idend_record_time': 8.0, 'evaluate_performance': True, 'evaluate_replay': False, 'record_idend_last_episode': True, 'store_connections': True, 'load_connections': False, 'sparse_first_char': False, 'active_weight_recorder': False, 'task': {'task_name': 'hard_coded', 'task_type': 1, 'vocab_size': 6, 'seed': 111, 'store_training_data': True}, 'learning_episodes': 40, 'episodes_to_testing': 1, 'conn_dict_ee': {'rule': 'fixed_indegree', 'indegree': 180, 'allow_autapses': False, 'allow_multapses': False}, 'R_m_soma': 0.04, 'R_m_inhibit': 0.02, 'DeltaT_seq': 100.0}\n",
+      "Model parameters: {'dt': 0.1, 'print_simulation_progress': False, 'soma_model': 'iaf_psc_exp_nonlineardendrite_neuron_nestml__with_stdsp_synapse_nestml', 'soma_params': {'C_m': 250.0, 'E_L': 0.0, 'V_m': 0.0, 'V_reset': 0.0, 'V_th': 20.0, 't_ref': 10.0, 'tau_m': 10.0, 'tau_syn1': 2.0, 'tau_syn2': 5.0, 'tau_syn3': 1.0, 'I_p': 200.0, 'tau_dAP': 60.0, 'theta_dAP': 59.0, 'unit_psr': 0.542}, 'fixed_somatic_delay': 2, 'M': 6, 'n_E': 150, 'n_I': 1, 'L': 1, 'pattern_size': 20, 'inhibit_model': 'iaf_psc_exp', 'inhibit_params': {'C_m': 250.0, 'E_L': 0.0, 'I_e': 0.0, 'V_m': 0.0, 'V_reset': 0.0, 'V_th': 15.0, 't_ref': 2.0, 'tau_m': 5.0, 'tau_syn_ex': 0.5, 'tau_syn_in': 1.65}, 'J_EX_psp': 22.0, 'J_IE_psp': 0.9, 'J_EI_psp': -40.0, 'convergence': 5, 'rule': 'fixed_indegree', 'connection_prob': 0.2, 'syn_dict_ee': {'weight': 0.01, 'synapse_model': 'stdsp_synapse_nestml__with_iaf_psc_exp_nonlineardendrite_neuron_nestml', 'permanence_threshold': 10.0, 'tau_pre_trace': 20.0, 'delay': 2.0, 'receptor_type': 2, 'lambda_plus': 0.08, 'zt': 1.0, 'lambda_h': 0.014, 'Wmax': 12.98, 'permanence_max': 20.0, 'permanence_min': 1.0, 'lambda_minus': 0.0015, 'dt_min': 4.0, 'permanence': <nest.lib.hl_api_types.Parameter object at 0x7f4733fb5450>, 'dt_max': 80.0}, 'permanence_min': 0.0, 'permanence_max': 8.0, 'calibration': 0.0, 'conn_dict_ex': {'rule': 'all_to_all'}, 'syn_dict_ex': {'receptor_type': 1, 'delay': 0.1, 'weight': 4112.209148358356}, 'conn_dict_edx': {'rule': 'fixed_outdegree', 'outdegree': 21}, 'syn_dict_edx': {'receptor_type': 2, 'delay': 0.1, 'weight': 82.6}, 'syn_dict_ie': {'synapse_model': 'static_synapse', 'delay': 0.1, 'weight': 581.1973492566976}, 'syn_dict_ei': {'synapse_model': 'static_synapse', 'delay': 0.1, 'receptor_type': 3, 'weight': -12915.496650148836}, 'DeltaT': 40.0, 'excitation_start': 30.0, 'time_dend_to_somatic': 20.0, 'DeltaT_cue': 80.0, 'overwrite_files': True, 'seed': 111, 'pad_time': 5.0, 'idend_recording_interval': 1.0, 'idend_record_time': 8.0, 'evaluate_performance': True, 'evaluate_replay': False, 'record_idend_last_episode': True, 'store_connections': True, 'load_connections': False, 'sparse_first_char': False, 'active_weight_recorder': False, 'task': {'task_name': 'hard_coded', 'task_type': 1, 'vocab_size': 6, 'seed': 111, 'store_training_data': True}, 'learning_episodes': 40, 'episodes_to_testing': 1, 'conn_dict_ee': {'rule': 'fixed_indegree', 'indegree': 180, 'allow_autapses': False, 'allow_multapses': False}, 'R_m_soma': 0.04, 'R_m_inhibit': 0.02, 'DeltaT_seq': 100.0}\n",
       "\n",
       "Duration of a sequence set 440 ms\n",
       "\n",
@@ -3502,7 +3496,7 @@
       "PSC maximum J_EX:  4112.209148 pA\n",
       "PSC maximum J_IE:  77.492980 pA\n",
       "PSC maximum J_EI:  -12915.496650 pA\n",
-      "Model parameters: {'dt': 0.1, 'print_simulation_progress': False, 'soma_model': 'iaf_psc_exp_nonlineardendrite_neuron_nestml__with_stdsp_synapse_nestml', 'soma_params': {'C_m': 250.0, 'E_L': 0.0, 'V_m': 0.0, 'V_reset': 0.0, 'V_th': 5.0, 't_ref': 10.0, 'tau_m': 10.0, 'tau_syn1': 2.0, 'tau_syn2': 5.0, 'tau_syn3': 1.0, 'I_p': 200.0, 'tau_dAP': 60.0, 'theta_dAP': 41.3, 'I_dend_incr': 0.542}, 'fixed_somatic_delay': 2, 'M': 6, 'n_E': 150, 'n_I': 1, 'L': 1, 'pattern_size': 20, 'inhibit_model': 'iaf_psc_exp', 'inhibit_params': {'C_m': 250.0, 'E_L': 0.0, 'I_e': 0.0, 'V_m': 0.0, 'V_reset': 0.0, 'V_th': 15.0, 't_ref': 2.0, 'tau_m': 5.0, 'tau_syn_ex': 0.5, 'tau_syn_in': 1.65}, 'J_EX_psp': 22.0, 'J_IE_psp': 0.12, 'J_EI_psp': -40.0, 'convergence': 5, 'rule': 'fixed_indegree', 'connection_prob': 0.2, 'syn_dict_ee': {'weight': 0.01, 'synapse_model': 'static_synapse', 'permanence_threshold': 10.0, 'tau_pre_trace': 20.0, 'delay': 2.0, 'receptor_type': 2, 'lambda_plus': 0.08, 'zt': 1.0, 'lambda_h': 0.014, 'Wmax': 12.98, 'permanence_max': 20.0, 'permanence_min': 1.0, 'lambda_minus': 0.0015, 'dt_min': 4.0, 'permanence': <nest.lib.hl_api_types.Parameter object at 0x7f46d10b76d0>, 'dt_max': -80.0}, 'permanence_min': 0.0, 'permanence_max': 8.0, 'calibration': 0.0, 'conn_dict_ex': {'rule': 'all_to_all'}, 'syn_dict_ex': {'receptor_type': 1, 'delay': 0.1, 'weight': 4112.209148358356}, 'conn_dict_edx': {'rule': 'fixed_outdegree', 'outdegree': 21}, 'syn_dict_edx': {'receptor_type': 2, 'delay': 0.1, 'weight': 82.6}, 'syn_dict_ie': {'synapse_model': 'static_synapse', 'delay': 0.1, 'weight': 77.49297990089302}, 'syn_dict_ei': {'synapse_model': 'static_synapse', 'delay': 0.1, 'receptor_type': 3, 'weight': -12915.496650148836}, 'DeltaT': 40.0, 'excitation_start': 30.0, 'time_dend_to_somatic': 20.0, 'DeltaT_cue': 250.0, 'overwrite_files': True, 'seed': 111, 'pad_time': 5.0, 'idend_recording_interval': 0.1, 'idend_record_time': 8.0, 'evaluate_performance': False, 'evaluate_replay': True, 'record_idend_last_episode': True, 'store_connections': False, 'load_connections': True, 'sparse_first_char': False, 'active_weight_recorder': False, 'task': {'task_name': 'hard_coded', 'task_type': 1, 'vocab_size': 6, 'seed': 111, 'store_training_data': True}, 'learning_episodes': 4, 'episodes_to_testing': 1, 'store_training_data': False, 'conn_dict_ee': {'rule': 'fixed_indegree', 'indegree': 180, 'allow_autapses': False, 'allow_multapses': False}, 'R_m_soma': 0.04, 'R_m_inhibit': 0.02, 'DeltaT_seq': 100.0}\n",
+      "Model parameters: {'dt': 0.1, 'print_simulation_progress': False, 'soma_model': 'iaf_psc_exp_nonlineardendrite_neuron_nestml__with_stdsp_synapse_nestml', 'soma_params': {'C_m': 250.0, 'E_L': 0.0, 'V_m': 0.0, 'V_reset': 0.0, 'V_th': 5.0, 't_ref': 10.0, 'tau_m': 10.0, 'tau_syn1': 2.0, 'tau_syn2': 5.0, 'tau_syn3': 1.0, 'I_p': 200.0, 'tau_dAP': 60.0, 'theta_dAP': 41.3, 'unit_psr': 0.542}, 'fixed_somatic_delay': 2, 'M': 6, 'n_E': 150, 'n_I': 1, 'L': 1, 'pattern_size': 20, 'inhibit_model': 'iaf_psc_exp', 'inhibit_params': {'C_m': 250.0, 'E_L': 0.0, 'I_e': 0.0, 'V_m': 0.0, 'V_reset': 0.0, 'V_th': 15.0, 't_ref': 2.0, 'tau_m': 5.0, 'tau_syn_ex': 0.5, 'tau_syn_in': 1.65}, 'J_EX_psp': 22.0, 'J_IE_psp': 0.12, 'J_EI_psp': -40.0, 'convergence': 5, 'rule': 'fixed_indegree', 'connection_prob': 0.2, 'syn_dict_ee': {'weight': 0.01, 'synapse_model': 'static_synapse', 'permanence_threshold': 10.0, 'tau_pre_trace': 20.0, 'delay': 2.0, 'receptor_type': 2, 'lambda_plus': 0.08, 'zt': 1.0, 'lambda_h': 0.014, 'Wmax': 12.98, 'permanence_max': 20.0, 'permanence_min': 1.0, 'lambda_minus': 0.0015, 'dt_min': 4.0, 'permanence': <nest.lib.hl_api_types.Parameter object at 0x7f46d10b76d0>, 'dt_max': -80.0}, 'permanence_min': 0.0, 'permanence_max': 8.0, 'calibration': 0.0, 'conn_dict_ex': {'rule': 'all_to_all'}, 'syn_dict_ex': {'receptor_type': 1, 'delay': 0.1, 'weight': 4112.209148358356}, 'conn_dict_edx': {'rule': 'fixed_outdegree', 'outdegree': 21}, 'syn_dict_edx': {'receptor_type': 2, 'delay': 0.1, 'weight': 82.6}, 'syn_dict_ie': {'synapse_model': 'static_synapse', 'delay': 0.1, 'weight': 77.49297990089302}, 'syn_dict_ei': {'synapse_model': 'static_synapse', 'delay': 0.1, 'receptor_type': 3, 'weight': -12915.496650148836}, 'DeltaT': 40.0, 'excitation_start': 30.0, 'time_dend_to_somatic': 20.0, 'DeltaT_cue': 250.0, 'overwrite_files': True, 'seed': 111, 'pad_time': 5.0, 'idend_recording_interval': 0.1, 'idend_record_time': 8.0, 'evaluate_performance': False, 'evaluate_replay': True, 'record_idend_last_episode': True, 'store_connections': False, 'load_connections': True, 'sparse_first_char': False, 'active_weight_recorder': False, 'task': {'task_name': 'hard_coded', 'task_type': 1, 'vocab_size': 6, 'seed': 111, 'store_training_data': True}, 'learning_episodes': 4, 'episodes_to_testing': 1, 'store_training_data': False, 'conn_dict_ee': {'rule': 'fixed_indegree', 'indegree': 180, 'allow_autapses': False, 'allow_multapses': False}, 'R_m_soma': 0.04, 'R_m_inhibit': 0.02, 'DeltaT_seq': 100.0}\n",
       "\n",
       "Duration of a sequence set 440 ms\n",
       "\n",

doc/tutorials/sequence_learning/sequence_learning.ipynb

@@ -68,17 +68,6 @@ def check_co_co(cls, node: ASTModel):
                         inline_expr_names.extend([inline_expr.variable_name for inline_expr in equations_block.get_inline_expressions()])
                         inline_exprs.extend(equations_block.get_inline_expressions())
 
-                    if var.get_name() in inline_expr_names:
-                        inline_expr_idx = inline_expr_names.index(var.get_name())
-                        inline_expr = inline_exprs[inline_expr_idx]
-                        from pynestml.utils.ast_utils import ASTUtils
-                        if ASTUtils.inline_aliases_convolution(inline_expr):
-                            symbol2 = node.get_scope().resolve_to_symbol(var.get_name(), SymbolKind.VARIABLE)
-                            if symbol2 is not None:
-                                # actually, no problem detected, skip error
-                                # XXX: TODO: check that differential order is less than or equal to that of the kernel
-                                continue
-
                     # check if this symbol is actually a type, e.g. "mV" in the expression "(1 + 2) * mV"
                     symbol2 = var.get_scope().resolve_to_symbol(var.get_complete_name(), SymbolKind.TYPE)
                     if symbol2 is not None:

pynestml/cocos/co_co_all_variables_defined.py

@@ -216,7 +216,7 @@ public:
   double {{ var }}_;
 
 {%-     set inline = utils.get_inline_expression_by_constructed_rhs_name(paired_synapse_original_model, var) %}
-{%-     if inline and utils.inline_aliases_convolution(inline) %}
+{%-     if inline %}
   double get_{{ inline.get_variable_name() + "__for_" + paired_synapse_original_model.get_name() }}() const     // getter for an inline expression
 {%-     else %}
   double get_{{ var }}() const