-
Notifications
You must be signed in to change notification settings - Fork 2
/
models.py
252 lines (212 loc) · 10.4 KB
/
models.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
import numpy as np
import tensorflow as tf
from tensorflow.keras import layers as tfkl
from tensorflow_probability import distributions as tfd
from tensorflow.keras.mixed_precision import experimental as prec
import tools
class RSSME(tools.Module):
def __init__(self, stoch=30, deter=200, hidden=200, num_models=7, act=tf.nn.elu):
super().__init__()
self._activation = act
self._stoch_size = stoch
self._deter_size = deter
self._hidden_size = hidden
self._cell = tfkl.GRUCell(self._deter_size)
self._k = num_models
def initial(self, batch_size):
dtype = prec.global_policy().compute_dtype
return dict(
mean=tf.zeros([batch_size, self._stoch_size], dtype),
std=tf.zeros([batch_size, self._stoch_size], dtype),
stoch=tf.zeros([batch_size, self._stoch_size], dtype),
deter=self._cell.get_initial_state(None, batch_size, dtype))
def observe(self, embed, action, state=None):
if state is None:
state = self.initial(tf.shape(action)[0])
embed = tf.transpose(embed, [1, 0, 2])
action = tf.transpose(action, [1, 0, 2])
post, prior = tools.static_scan(lambda prev, inputs: self.obs_step(prev[0], *inputs),
(action, embed), (state, state))
post = {k: tf.transpose(v, [1, 0, 2]) for k, v in post.items()}
prior = {k: tf.transpose(v, [1, 0, 2]) for k, v in prior.items()}
return post, prior
def imagine(self, action, state=None):
if state is None:
state = self.initial(tf.shape(action)[0])
assert isinstance(state, dict), state
action = tf.transpose(action, [1, 0, 2])
prior = tools.static_scan(self.img_step, action, state)
prior = {k: tf.transpose(v, [1, 0, 2]) for k, v in prior.items()}
return prior
def get_feat(self, state):
return tf.concat([state['stoch'], state['deter']], -1)
def get_feat_size(self, state):
return self._stoch_size + self._deter_size
def get_dist(self, state):
return tfd.MultivariateNormalDiag(state['mean'], state['std'])
def obs_step(self, prev_state, prev_action, embed):
prior = self.img_step(prev_state, prev_action)
x = tf.concat([prior['deter'], embed], -1)
x = self.get('obs1', tfkl.Dense, self._hidden_size, self._activation)(x)
x = self.get('obs2', tfkl.Dense, 2 * self._stoch_size, None)(x)
mean, std = tf.split(x, 2, -1)
std = tf.nn.softplus(std) + 0.1
stoch = self.get_dist({'mean': mean, 'std': std}).sample()
post = {'mean': mean, 'std': std, 'stoch': stoch, 'deter': prior['deter']}
return post, prior
def img_step(self, prev_state, prev_action, k=None):
if k is None:
k = np.random.choice(self._k)
x = tf.concat([prev_state['stoch'], prev_action], -1)
x = self.get('img1', tfkl.Dense, self._hidden_size, self._activation)(x)
x, deter = self._cell(x, [prev_state['deter']])
deter = deter[0] # Keras wraps the state in a list.
x = self.get('img2_{}'.format(k), tfkl.Dense, self._hidden_size, self._activation)(x)
x = self.get('img3_{}'.format(k), tfkl.Dense, 2 * self._stoch_size, None)(x)
mean, std = tf.split(x, 2, -1)
std = tf.nn.softplus(std) + 0.1
stoch = self.get_dist({'mean': mean, 'std': std}).sample()
prior = {'mean': mean, 'std': std, 'stoch': stoch, 'deter': deter}
return prior
class MultivariateNormalDiag(tools.Module):
def __init__(self, hidden_size, latent_size, scale=None):
super().__init__()
self.latent_size = latent_size
self.scale = scale
self.dense1 = tf.keras.layers.Dense(hidden_size, activation=tf.nn.leaky_relu)
self.dense2 = tf.keras.layers.Dense(hidden_size, activation=tf.nn.leaky_relu)
self.output_layer = tf.keras.layers.Dense(2 * latent_size if self.scale
is None else latent_size)
def __call__(self, *inputs):
if len(inputs) > 1:
inputs = tf.concat(inputs, axis=-1)
else:
inputs, = inputs
out = self.dense1(inputs)
out = self.dense2(out)
out = self.output_layer(out)
loc = out[..., :self.latent_size]
if self.scale is None:
assert out.shape[-1] == 2 * self.latent_size
scale_diag = tf.nn.softplus(out[..., self.latent_size:]) + 1e-5
else:
assert out.shape[-1].value == self.latent_size
scale_diag = tf.ones_like(loc) * self.scale
return loc, scale_diag
class ConstantMultivariateNormalDiag(tools.Module):
def __init__(self, latent_size, scale=None):
super().__init__()
self.latent_size = latent_size
self.scale = scale
def __call__(self, *inputs):
# first input should not have any dimensions after the batch_shape, step_type
batch_shape = tf.shape(inputs[0]) # input is only used to infer batch_shape
shape = tf.concat([batch_shape, [self.latent_size]], axis=0)
loc = tf.zeros(shape)
if self.scale is None:
scale_diag = tf.ones(shape)
else:
scale_diag = tf.ones(shape) * self.scale
return loc, scale_diag
class ConvEncoderLarge(tools.Module):
def __init__(self, depth=32, act=tf.nn.relu):
self._act = act
self._depth = depth
def __call__(self, obs):
kwargs = dict(strides=2, activation=self._act)
x = tf.reshape(obs['image'], (-1,) + tuple(obs['image'].shape[-3:]))
x = self.get('h1', tfkl.Conv2D, 1 * self._depth, 4, **kwargs)(x)
x = self.get('h2', tfkl.Conv2D, 2 * self._depth, 4, **kwargs)(x)
x = self.get('h3', tfkl.Conv2D, 4 * self._depth, 4, **kwargs)(x)
x = self.get('h4', tfkl.Conv2D, 8 * self._depth, 4, **kwargs)(x)
x = self.get('h5', tfkl.Conv2D, 8 * self._depth, 4, **kwargs)(x)
shape = tf.concat([tf.shape(obs['image'])[:-3], [32 * self._depth]], 0)
return tf.reshape(x, shape)
class ConvDecoderLarge(tools.Module):
def __init__(self, depth=32, act=tf.nn.relu, shape=(128, 128, 3)):
self._act = act
self._depth = depth
self._shape = shape
def __call__(self, features):
kwargs = dict(strides=2, activation=self._act)
x = self.get('h1', tfkl.Dense, 32 * self._depth, None)(features)
x = tf.reshape(x, [-1, 1, 1, 32 * self._depth])
x = self.get('h2', tfkl.Conv2DTranspose, 4 * self._depth, 5, **kwargs)(x)
x = self.get('h3', tfkl.Conv2DTranspose, 2 * self._depth, 5, **kwargs)(x)
x = self.get('h4', tfkl.Conv2DTranspose, 1 * self._depth, 5, **kwargs)(x)
x = self.get('h5', tfkl.Conv2DTranspose, 1 * self._depth, 6, **kwargs)(x)
x = self.get('h6', tfkl.Conv2DTranspose, self._shape[-1], 6, strides=2)(x)
mean = tf.reshape(x, tf.concat([tf.shape(features)[:-1], self._shape], 0))
return tfd.Independent(tfd.Normal(mean, 1), len(self._shape))
class ConvEncoder(tools.Module):
def __init__(self, depth=32, act=tf.nn.relu):
self._act = act
self._depth = depth
def __call__(self, obs):
kwargs = dict(strides=2, activation=self._act)
x = tf.reshape(obs['image'], (-1,) + tuple(obs['image'].shape[-3:]))
x = self.get('h1', tfkl.Conv2D, 1 * self._depth, 4, **kwargs)(x)
x = self.get('h2', tfkl.Conv2D, 2 * self._depth, 4, **kwargs)(x)
x = self.get('h3', tfkl.Conv2D, 4 * self._depth, 4, **kwargs)(x)
x = self.get('h4', tfkl.Conv2D, 8 * self._depth, 4, **kwargs)(x)
shape = tf.concat([tf.shape(obs['image'])[:-3], [32 * self._depth]], 0)
return tf.reshape(x, shape)
class ConvDecoder(tools.Module):
def __init__(self, depth=32, act=tf.nn.relu, shape=(64, 64, 3)):
self._act = act
self._depth = depth
self._shape = shape
def __call__(self, features):
kwargs = dict(strides=2, activation=self._act)
x = self.get('h1', tfkl.Dense, 32 * self._depth, None)(features)
x = tf.reshape(x, [-1, 1, 1, 32 * self._depth])
x = self.get('h2', tfkl.Conv2DTranspose, 4 * self._depth, 5, **kwargs)(x)
x = self.get('h3', tfkl.Conv2DTranspose, 2 * self._depth, 5, **kwargs)(x)
x = self.get('h4', tfkl.Conv2DTranspose, 1 * self._depth, 6, **kwargs)(x)
x = self.get('h5', tfkl.Conv2DTranspose, self._shape[-1], 6, strides=2)(x)
mean = tf.reshape(x, tf.concat([tf.shape(features)[:-1], self._shape], 0))
return tfd.Independent(tfd.Normal(mean, 1), len(self._shape))
class DenseDecoder(tools.Module):
def __init__(self, shape, layers, units, dist='normal', act=tf.nn.elu):
self._shape = shape
self._layers = layers
self._units = units
self._dist = dist
self._act = act
def __call__(self, features):
x = features
for index in range(self._layers):
x = self.get(f'h{index}', tfkl.Dense, self._units, self._act)(x)
x = self.get(f'hout', tfkl.Dense, np.prod(self._shape))(x)
x = tf.reshape(x, tf.concat([tf.shape(features)[:-1], self._shape], 0))
if self._dist == 'normal':
return tfd.Independent(tfd.Normal(x, 1), len(self._shape))
if self._dist == 'binary':
return tfd.Independent(tfd.Bernoulli(x), len(self._shape))
raise NotImplementedError(self._dist)
class DenseNetwork(tools.Module):
def __init__(self, shape, layers, units, act=tf.nn.elu):
self._shape = shape
self._layers = layers
self._units = units
self._act = act
def __call__(self, features):
x = features
for index in range(self._layers):
x = self.get(f'h{index}', tfkl.Dense, self._units, self._act)(x)
x = self.get(f'hout', tfkl.Dense, self._shape)(x)
return x
class ActorNetwork(tools.Module):
def __init__(self, shape, layers, units, act=tf.nn.elu, mean_scale=1.0):
self._shape = shape
self._layers = layers
self._units = units
self._act = act
self._mean_scale = mean_scale
def __call__(self, features):
x = features
for index in range(self._layers):
x = self.get(f'h{index}', tfkl.Dense, self._units, self._act)(x)
x = self.get(f'hout', tfkl.Dense, self._shape)(x)
x = self._mean_scale * tf.tanh(x)
return x