Got most of the neural network stuff working. Some parameter updates to manage in NeuralNetworkSpecifications
youainti
5 years ago
parent
7b51044057
commit
d8fff40288
@ -0,0 +1,109 @@
|
||||
import torch
|
||||
import combined as c
|
||||
|
||||
"""
|
||||
This module holds the neural networks I am going to use to estimate
|
||||
the functions of interest.
|
||||
"""
|
||||
|
||||
class LaunchFnEstimand(torch.nn.Module):
|
||||
"""
|
||||
This is used to estimate the launch function
|
||||
"""
|
||||
def __init__(self, state_tensor_size,layers_size,number_constellations):
|
||||
super().__init__()
|
||||
self.number_constellations = number_constellations
|
||||
self.layers_size = layers_size
|
||||
self.state_tensor_size = state_tensor_size
|
||||
|
||||
|
||||
#Layers
|
||||
self.linear_1 = torch.nn.Linear(in_features=state_tensor_size, out_features=layers_size)
|
||||
self.relu = torch.nn.ReLU()
|
||||
self.linear_3 = torch.nn.Linear(in_features=layers_size, out_features=layers_size)
|
||||
self.linear_5 = torch.nn.Linear(in_features=layers_size, out_features=number_constellations)
|
||||
|
||||
|
||||
def forward(self, input_values):
|
||||
|
||||
intermediate_values = self.relu(input_values) #states should be positive anyway.
|
||||
intermediate_values = self.linear_1(intermediate_values)
|
||||
intermediate_values = self.linear_3(intermediate_values)
|
||||
intermediate_values = self.linear_5(intermediate_values)
|
||||
intermediate_values = self.relu(intermediate_values) #launches are always positive
|
||||
|
||||
return intermediate_values
|
||||
|
||||
class PartialDerivativesEstimand(torch.nn.Module):
|
||||
"""
|
||||
This is used to estimate the partial derivatives of the value functions
|
||||
"""
|
||||
def __init__(self,batch_size, number_constellations, number_states, layer_size=12):
|
||||
super().__init__()
|
||||
self.batch_size = batch_size
|
||||
self.number_constellations = number_constellations
|
||||
self.number_states = number_states
|
||||
self.layer_size = layer_size
|
||||
|
||||
|
||||
#preprocess (single linear layer in case there is anything that needs to happen to all states)
|
||||
self.preprocess = torch.nn.Sequential(
|
||||
torch.nn.ReLU() #cleanup as states must be positive
|
||||
,torch.nn.Linear(in_features = self.number_states, out_features=self.number_states)
|
||||
)
|
||||
#upscale to get the basic dimensionality correct. From (batch,State) to (batch, constellation, state). Includes a reshape
|
||||
self.upsample = lambda x: torch.nn.Upsample(scale_factor=self.number_constellations)(x).view(self.batch_size
|
||||
,self.number_constellations
|
||||
,self.number_states)
|
||||
|
||||
#sequential steps
|
||||
self.sequential = torch.nn.Sequential(
|
||||
torch.nn.Linear(in_features=number_states, out_features=layer_size)
|
||||
#who knows if a convolution might help here.
|
||||
,torch.nn.Linear(in_features=layer_size, out_features=layer_size)
|
||||
,torch.nn.Linear(in_features=layer_size, out_features=layer_size)
|
||||
)
|
||||
|
||||
#reduce the feature axis to match expected results
|
||||
self.feature_reduction = torch.nn.Linear(in_features=layer_size, out_features=number_states)
|
||||
|
||||
def forward(self, input_values):
|
||||
#Note that the input values are just going to be the state variables
|
||||
#TODO:check that input values match the prepared dimension?
|
||||
|
||||
#preprocess
|
||||
intermediate = self.preprocess(input_values)
|
||||
|
||||
#upscale the input values
|
||||
intermediate = self.upsample(intermediate)
|
||||
|
||||
#intermediate processing
|
||||
intermediate = self.sequential(intermediate)
|
||||
|
||||
#reduce feature axis to match the expected number of partials
|
||||
intermediate = self.feature_reduction(intermediate)
|
||||
|
||||
|
||||
return intermediate
|
||||
|
||||
class EstimandNN(torch.nn.Module):
|
||||
"""
|
||||
This neural network takes the current states as input values and returns both
|
||||
the partial derivatives of the value function and the launch function.
|
||||
"""
|
||||
def __init__(self, state_tensor_size,layers_size,number_constellations):
|
||||
super().__init__()
|
||||
|
||||
#So, this next section constructs different layers within the NN
|
||||
#sinlge linear section
|
||||
pass
|
||||
#TODO:verify these are correct
|
||||
self.partials_estimator = PartialDerivativesEstimand(state_tensor_size,layers_size,number_constellations) #TODO
|
||||
self.launch_estimator = LaunchFnEstimand(state_tensor_size,layers_size,number_constellations)
|
||||
|
||||
def forward(self, input_values):
|
||||
pass
|
||||
partials = self.partials_estimator(input_values)
|
||||
launch = self.launch_estimator(input_values)
|
||||
|
||||
return c.EstimandInterface(partials,launch)
|
||||
@ -0,0 +1,203 @@
|
||||
{
|
||||
"cells": [
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 1,
|
||||
"id": "least-cooling",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"import torch"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"id": "statistical-temperature",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"The purpose of this notebook is to allow me to investigate proper shaping of inputs.\n",
|
||||
"\n",
|
||||
"Typically pytorch chooses a tensor specification\n",
|
||||
"$$\n",
|
||||
"(N, .*)\n",
|
||||
"$$\n",
|
||||
"where $N$ is the batch size.\n",
|
||||
"For example a Convolutional NN layer expects\n",
|
||||
"$$\n",
|
||||
" NCHW\n",
|
||||
"$$\n",
|
||||
"for BatchSize,ChannelSize,Height,Width.\n",
|
||||
"On the other hand, Linear expects\n",
|
||||
"$$\n",
|
||||
" N.*H\n",
|
||||
"$$\n",
|
||||
"for BatchSize,any number of other dimensions, in_features\n",
|
||||
"\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 15,
|
||||
"id": "japanese-poultry",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"class PartialDerivativesEstimand(torch.nn.Module):\n",
|
||||
" def __init__(self,batch_size, number_constellations, number_states,scale_factor=4, layer_size=12):\n",
|
||||
" \"\"\"\n",
|
||||
" \n",
|
||||
" \"\"\"\n",
|
||||
" super().__init__()\n",
|
||||
" self.batch_size = batch_size\n",
|
||||
" self.number_constellations = number_constellations\n",
|
||||
" self.number_states = number_states\n",
|
||||
" self.scale_factor = scale_factor\n",
|
||||
" self.layer_size = layer_size\n",
|
||||
" \n",
|
||||
" \n",
|
||||
" #preprocess (single linear layer in case there is anything that needs to happen to all states)\n",
|
||||
" self.preprocess = torch.nn.Sequential(\n",
|
||||
" torch.nn.ReLU() #cleanup as states must be positive\n",
|
||||
" ,torch.nn.Linear(in_features = self.number_states, out_features=self.number_states)\n",
|
||||
" )\n",
|
||||
" #upscale to get the basic dimensionality correct. From (batch,State) to (batch, constellation, state). Includes a reshape\n",
|
||||
" self.upsample = lambda x: torch.nn.Upsample(scale_factor=self.number_constellations)(x).view(self.batch_size\n",
|
||||
" ,self.number_constellations\n",
|
||||
" ,self.number_states)\n",
|
||||
" \n",
|
||||
" #sequential steps\n",
|
||||
" self.sequential = torch.nn.Sequential(\n",
|
||||
" torch.nn.Linear(in_features=number_states, out_features=layer_size)\n",
|
||||
" #who knows if a convolution might help here.\n",
|
||||
" ,torch.nn.Linear(in_features=layer_size, out_features=layer_size)\n",
|
||||
" ,torch.nn.Linear(in_features=layer_size, out_features=layer_size)\n",
|
||||
" )\n",
|
||||
"\n",
|
||||
" #reduce axis to match expectation\n",
|
||||
" self.feature_reduction = torch.nn.Linear(in_features=layer_size, out_features=number_states)\n",
|
||||
" \n",
|
||||
" def forward(self, input_values):\n",
|
||||
" #Note that the input values are just going to be the state variables\n",
|
||||
" #TODO:check that input values match the prepared dimension?\n",
|
||||
" \n",
|
||||
" #preprocess\n",
|
||||
" intermediate = self.preprocess(input_values)\n",
|
||||
" \n",
|
||||
" #upscale the input values\n",
|
||||
" intermediate = self.upsample(intermediate)\n",
|
||||
" \n",
|
||||
" #intermediate processing\n",
|
||||
" intermediate = self.sequential(intermediate)\n",
|
||||
" \n",
|
||||
" #reduce feature axis to match the expected number of partials\n",
|
||||
" intermediate = self.feature_reduction(intermediate)\n",
|
||||
" \n",
|
||||
" \n",
|
||||
" return intermediate"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 19,
|
||||
"id": "communist-teach",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"batch_size = 2\n",
|
||||
"constellations = 2\n",
|
||||
"number_states = constellations+1\n",
|
||||
"\n",
|
||||
"#initialize the NN\n",
|
||||
"a = PartialDerivativesEstimand(batch_size,constellations,number_states,scale_factor=2)\n",
|
||||
"\n",
|
||||
"#example state\n",
|
||||
"s = torch.rand(size=(batch_size,1,number_states))"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 22,
|
||||
"id": "chemical-revolution",
|
||||
"metadata": {},
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"tensor([[[0.9283, 0.9414, 0.3426]],\n",
|
||||
"\n",
|
||||
" [[0.1902, 0.0369, 0.4699]]])"
|
||||
]
|
||||
},
|
||||
"execution_count": 22,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"s"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 23,
|
||||
"id": "directed-lobby",
|
||||
"metadata": {},
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"tensor([[[-0.1991, 0.1335, 0.2821],\n",
|
||||
" [-0.3549, 0.0213, 0.2322]],\n",
|
||||
"\n",
|
||||
" [[-0.1701, 0.1557, 0.2954],\n",
|
||||
" [-0.3017, 0.0690, 0.2419]]], grad_fn=<AddBackward0>)"
|
||||
]
|
||||
},
|
||||
"execution_count": 23,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"a(s)"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"id": "placed-coating",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": []
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"id": "advised-execution",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": []
|
||||
}
|
||||
],
|
||||
"metadata": {
|
||||
"kernelspec": {
|
||||
"display_name": "Python 3",
|
||||
"language": "python",
|
||||
"name": "python3"
|
||||
},
|
||||
"language_info": {
|
||||
"codemirror_mode": {
|
||||
"name": "ipython",
|
||||
"version": 3
|
||||
},
|
||||
"file_extension": ".py",
|
||||
"mimetype": "text/x-python",
|
||||
"name": "python",
|
||||
"nbconvert_exporter": "python",
|
||||
"pygments_lexer": "ipython3",
|
||||
"version": "3.8.8"
|
||||
}
|
||||
},
|
||||
"nbformat": 4,
|
||||
"nbformat_minor": 5
|
||||
}
|
||||
Loading…
Reference in New Issue