File size: 10,018 Bytes
5960497
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
import os
import numpy as np
import tensorflow as tf
from collections import deque

def sample(logits):
    noise = tf.random.uniform(tf.shape(input=logits))
    return tf.argmax(input=logits - tf.math.log(-tf.math.log(noise)), axis=1)

def cat_entropy(logits):
    a0 = logits - tf.reduce_max(input_tensor=logits, axis=1, keepdims=True)
    ea0 = tf.exp(a0)
    z0 = tf.reduce_sum(input_tensor=ea0, axis=1, keepdims=True)
    p0 = ea0 / z0
    return tf.reduce_sum(input_tensor=p0 * (tf.math.log(z0) - a0), axis=1)

def cat_entropy_softmax(p0):
    return - tf.reduce_sum(input_tensor=p0 * tf.math.log(p0 + 1e-6), axis = 1)

def ortho_init(scale=1.0):
    def _ortho_init(shape, dtype, partition_info=None):
        #lasagne ortho init for tf
        shape = tuple(shape)
        if len(shape) == 2:
            flat_shape = shape
        elif len(shape) == 4: # assumes NHWC
            flat_shape = (np.prod(shape[:-1]), shape[-1])
        else:
            raise NotImplementedError
        a = np.random.normal(0.0, 1.0, flat_shape)
        u, _, v = np.linalg.svd(a, full_matrices=False)
        q = u if u.shape == flat_shape else v # pick the one with the correct shape
        q = q.reshape(shape)
        return (scale * q[:shape[0], :shape[1]]).astype(np.float32)
    return _ortho_init

def conv(x, scope, *, nf, rf, stride, pad='VALID', init_scale=1.0, data_format='NHWC', one_dim_bias=False):
    if data_format == 'NHWC':
        channel_ax = 3
        strides = [1, stride, stride, 1]
        bshape = [1, 1, 1, nf]
    elif data_format == 'NCHW':
        channel_ax = 1
        strides = [1, 1, stride, stride]
        bshape = [1, nf, 1, 1]
    else:
        raise NotImplementedError
    bias_var_shape = [nf] if one_dim_bias else [1, nf, 1, 1]
    try:
        nin = x.get_shape()[channel_ax].value
    except:
        nin = x.get_shape()[channel_ax]
    wshape = [rf, rf, nin, nf]
    with tf.compat.v1.variable_scope(scope):
        w = tf.compat.v1.get_variable("w", wshape, initializer=ortho_init(init_scale))
        b = tf.compat.v1.get_variable("b", bias_var_shape, initializer=tf.compat.v1.constant_initializer(0.0))
        if not one_dim_bias and data_format == 'NHWC':
            b = tf.reshape(b, bshape)
        return tf.nn.conv2d(input=x, filters=w, strides=strides, padding=pad, data_format=data_format) + b

def fc(x, scope, nh, *, init_scale=1.0, init_bias=0.0):
    with tf.compat.v1.variable_scope(scope):
        try:
            nin = x.get_shape()[1].value
        except:
            nin = x.get_shape()[1]
        w = tf.compat.v1.get_variable("w", [nin, nh], initializer=ortho_init(init_scale))
        b = tf.compat.v1.get_variable("b", [nh], initializer=tf.compat.v1.constant_initializer(init_bias))
        return tf.matmul(x, w)+b

def batch_to_seq(h, nbatch, nsteps, flat=False):
    if flat:
        h = tf.reshape(h, [nbatch, nsteps])
    else:
        h = tf.reshape(h, [nbatch, nsteps, -1])
    return [tf.squeeze(v, [1]) for v in tf.split(axis=1, num_or_size_splits=nsteps, value=h)]

def seq_to_batch(h, flat = False):
    shape = h[0].get_shape().as_list()
    if not flat:
        assert(len(shape) > 1)
        nh = h[0].get_shape()[-1].value
        return tf.reshape(tf.concat(axis=1, values=h), [-1, nh])
    else:
        return tf.reshape(tf.stack(values=h, axis=1), [-1])

def lstm(xs, ms, s, scope, nh, init_scale=1.0):
    nbatch, nin = [v.value for v in xs[0].get_shape()]
    with tf.compat.v1.variable_scope(scope):
        wx = tf.compat.v1.get_variable("wx", [nin, nh*4], initializer=ortho_init(init_scale))
        wh = tf.compat.v1.get_variable("wh", [nh, nh*4], initializer=ortho_init(init_scale))
        b = tf.compat.v1.get_variable("b", [nh*4], initializer=tf.compat.v1.constant_initializer(0.0))

    c, h = tf.split(axis=1, num_or_size_splits=2, value=s)
    for idx, (x, m) in enumerate(zip(xs, ms)):
        c = c*(1-m)
        h = h*(1-m)
        z = tf.matmul(x, wx) + tf.matmul(h, wh) + b
        i, f, o, u = tf.split(axis=1, num_or_size_splits=4, value=z)
        i = tf.nn.sigmoid(i)
        f = tf.nn.sigmoid(f)
        o = tf.nn.sigmoid(o)
        u = tf.tanh(u)
        c = f*c + i*u
        h = o*tf.tanh(c)
        xs[idx] = h
    s = tf.concat(axis=1, values=[c, h])
    return xs, s

def _ln(x, g, b, e=1e-5, axes=[1]):
    u, s = tf.nn.moments(x=x, axes=axes, keepdims=True)
    x = (x-u)/tf.sqrt(s+e)
    x = x*g+b
    return x

def lnlstm(xs, ms, s, scope, nh, init_scale=1.0):
    nbatch, nin = [v.value for v in xs[0].get_shape()]
    with tf.compat.v1.variable_scope(scope):
        wx = tf.compat.v1.get_variable("wx", [nin, nh*4], initializer=ortho_init(init_scale))
        gx = tf.compat.v1.get_variable("gx", [nh*4], initializer=tf.compat.v1.constant_initializer(1.0))
        bx = tf.compat.v1.get_variable("bx", [nh*4], initializer=tf.compat.v1.constant_initializer(0.0))

        wh = tf.compat.v1.get_variable("wh", [nh, nh*4], initializer=ortho_init(init_scale))
        gh = tf.compat.v1.get_variable("gh", [nh*4], initializer=tf.compat.v1.constant_initializer(1.0))
        bh = tf.compat.v1.get_variable("bh", [nh*4], initializer=tf.compat.v1.constant_initializer(0.0))

        b = tf.compat.v1.get_variable("b", [nh*4], initializer=tf.compat.v1.constant_initializer(0.0))

        gc = tf.compat.v1.get_variable("gc", [nh], initializer=tf.compat.v1.constant_initializer(1.0))
        bc = tf.compat.v1.get_variable("bc", [nh], initializer=tf.compat.v1.constant_initializer(0.0))

    c, h = tf.split(axis=1, num_or_size_splits=2, value=s)
    for idx, (x, m) in enumerate(zip(xs, ms)):
        c = c*(1-m)
        h = h*(1-m)
        z = _ln(tf.matmul(x, wx), gx, bx) + _ln(tf.matmul(h, wh), gh, bh) + b
        i, f, o, u = tf.split(axis=1, num_or_size_splits=4, value=z)
        i = tf.nn.sigmoid(i)
        f = tf.nn.sigmoid(f)
        o = tf.nn.sigmoid(o)
        u = tf.tanh(u)
        c = f*c + i*u
        h = o*tf.tanh(_ln(c, gc, bc))
        xs[idx] = h
    s = tf.concat(axis=1, values=[c, h])
    return xs, s

def conv_to_fc(x):
    try:
        nh = np.prod([v.value for v in x.get_shape()[1:]])
    except:
        nh = np.prod([v for v in x.get_shape()[1:]])
    x = tf.reshape(x, [-1, nh])
    return x

def discount_with_dones(rewards, dones, gamma):
    discounted = []
    r = 0
    for reward, done in zip(rewards[::-1], dones[::-1]):
        r = reward + gamma*r*(1.-done) # fixed off by one bug
        discounted.append(r)
    return discounted[::-1]

def find_trainable_variables(key):
    return tf.compat.v1.trainable_variables(key)

def make_path(f):
    return os.makedirs(f, exist_ok=True)

def constant(p):
    return 1

def linear(p):
    return 1-p

def middle_drop(p):
    eps = 0.75
    if 1-p<eps:
        return eps*0.1
    return 1-p

def double_linear_con(p):
    p *= 2
    eps = 0.125
    if 1-p<eps:
        return eps
    return 1-p

def double_middle_drop(p):
    eps1 = 0.75
    eps2 = 0.25
    if 1-p<eps1:
        if 1-p<eps2:
            return eps2*0.5
        return eps1*0.1
    return 1-p

schedules = {
    'linear':linear,
    'constant':constant,
    'double_linear_con': double_linear_con,
    'middle_drop': middle_drop,
    'double_middle_drop': double_middle_drop
}

class Scheduler(object):

    def __init__(self, v, nvalues, schedule):
        self.n = 0.
        self.v = v
        self.nvalues = nvalues
        self.schedule = schedules[schedule]

    def value(self):
        current_value = self.v*self.schedule(self.n/self.nvalues)
        self.n += 1.
        return current_value

    def value_steps(self, steps):
        return self.v*self.schedule(steps/self.nvalues)


class EpisodeStats:
    def __init__(self, nsteps, nenvs):
        self.episode_rewards = []
        for i in range(nenvs):
            self.episode_rewards.append([])
        self.lenbuffer = deque(maxlen=40)  # rolling buffer for episode lengths
        self.rewbuffer = deque(maxlen=40)  # rolling buffer for episode rewards
        self.nsteps = nsteps
        self.nenvs = nenvs

    def feed(self, rewards, masks):
        rewards = np.reshape(rewards, [self.nenvs, self.nsteps])
        masks = np.reshape(masks, [self.nenvs, self.nsteps])
        for i in range(0, self.nenvs):
            for j in range(0, self.nsteps):
                self.episode_rewards[i].append(rewards[i][j])
                if masks[i][j]:
                    l = len(self.episode_rewards[i])
                    s = sum(self.episode_rewards[i])
                    self.lenbuffer.append(l)
                    self.rewbuffer.append(s)
                    self.episode_rewards[i] = []

    def mean_length(self):
        if self.lenbuffer:
            return np.mean(self.lenbuffer)
        else:
            return 0  # on the first params dump, no episodes are finished

    def mean_reward(self):
        if self.rewbuffer:
            return np.mean(self.rewbuffer)
        else:
            return 0


# For ACER
def get_by_index(x, idx):
    assert(len(x.get_shape()) == 2)
    assert(len(idx.get_shape()) == 1)
    idx_flattened = tf.range(0, x.shape[0]) * x.shape[1] + idx
    y = tf.gather(tf.reshape(x, [-1]),  # flatten input
                  idx_flattened)  # use flattened indices
    return y

def check_shape(ts,shapes):
    i = 0
    for (t,shape) in zip(ts,shapes):
        assert t.get_shape().as_list()==shape, "id " + str(i) + " shape " + str(t.get_shape()) + str(shape)
        i += 1

def avg_norm(t):
    return tf.reduce_mean(input_tensor=tf.sqrt(tf.reduce_sum(input_tensor=tf.square(t), axis=-1)))

def gradient_add(g1, g2, param):
    print([g1, g2, param.name])
    assert (not (g1 is None and g2 is None)), param.name
    if g1 is None:
        return g2
    elif g2 is None:
        return g1
    else:
        return g1 + g2

def q_explained_variance(qpred, q):
    _, vary = tf.nn.moments(x=q, axes=[0, 1])
    _, varpred = tf.nn.moments(x=q - qpred, axes=[0, 1])
    check_shape([vary, varpred], [[]] * 2)
    return 1.0 - (varpred / vary)