ftw
latest
TensorFlow Modules & Networks
For The Win (FTW) TensorFlow Agent
Utilities for Replay Buffers, Datasets, Hyperparameters & Internal Rewards
Reverb Adders
Reverb Replay Datasets for TensorFlow
Hyperparameters
Internal Rewards
Environment Wrappers
ftw
Docs
»
Index
Edit on GitHub
Index
_
|
A
|
C
|
D
|
F
|
G
|
H
|
I
|
L
|
M
|
N
|
O
|
P
|
R
|
S
|
T
|
U
|
V
|
W
|
Z
_
__init__() (ftw.adders.reverb.MultiSequenceAdder method)
(ftw.adders.reverb.NonOverlappingRNNSequenceAdder method)
(ftw.adders.reverb.SequenceAdder method)
(ftw.agents.tf.ftw.acting.FtwActor method)
(ftw.agents.tf.ftw.agent.FTW method)
(ftw.agents.tf.ftw.learning.FtwLearner method)
(ftw.tf.hyperparameters.ftw_hp.FloatHyperparameter method)
(ftw.tf.hyperparameters.ftw_hp.IntHyperparameter method)
(ftw.tf.internal_reward.InternalRewards method)
(ftw.tf.networks.auxiliary.PixelControl method)
(ftw.tf.networks.auxiliary.RNNPixelControlNetwork method)
(ftw.tf.networks.auxiliary.RewardPrediction method)
(ftw.tf.networks.auxiliary.RewardPredictionNetwork method)
(ftw.tf.networks.distributional.MultivariateNormalDiagHead method)
(ftw.tf.networks.distributional.MultivariateNormalDiagLocScaleHead method)
(ftw.tf.networks.dnc.access.MemoryAccess method)
(ftw.tf.networks.dnc.dnc.DNC method)
(ftw.tf.networks.embedding.OAREmbedding method)
(ftw.tf.networks.policy_value.PolicyValueHead method)
(ftw.tf.networks.recurrence.DNCWrapper method)
(ftw.tf.networks.recurrence.PeriodicVariationalUnit method)
(ftw.tf.networks.recurrence.RPTH method)
,
[1]
(ftw.tf.networks.recurrence.VariationalUnit method)
(ftw.tf.networks.vision.FtwTorso method)
(ftw.wrappers.MarlGymWrapper method)
(ftw.wrappers.ObservationActionRewardMarlWrapper method)
A
access_output (ftw.tf.networks.dnc.dnc.DNCState attribute)
access_state (ftw.tf.networks.dnc.dnc.DNCState attribute)
AccessState (class in ftw.tf.networks.dnc.access)
AdderTestMixin (class in ftw.adders.reverb.test_utils)
append() (ftw.adders.reverb.test_utils.FakeWriter method)
C
close() (ftw.adders.reverb.test_utils.FakeWriter method)
controller_state (ftw.tf.networks.dnc.dnc.DNCState attribute)
core_state (ftw.tf.networks.recurrence.PeriodicRNNState attribute)
(ftw.tf.networks.recurrence.RPTHState attribute)
CosineWeights (class in ftw.tf.networks.dnc.addressing)
create_adders() (in module ftw.agents.tf.ftw.utils)
create_datasets() (in module ftw.agents.tf.ftw.utils)
create_item() (ftw.adders.reverb.test_utils.FakeWriter method)
create_reverb_tables() (in module ftw.agents.tf.ftw.utils)
D
directional_read_weights() (ftw.tf.networks.dnc.addressing.TemporalLinkage method)
distribution_params (ftw.tf.networks.recurrence.RPTHOutput attribute)
DNC (class in ftw.tf.networks.dnc.dnc)
DNCState (class in ftw.tf.networks.dnc.dnc)
DNCWrapper (class in ftw.tf.networks.recurrence)
F
FakeClient (class in ftw.adders.reverb.test_utils)
FakeWriter (class in ftw.adders.reverb.test_utils)
FloatHyperparameter (class in ftw.tf.hyperparameters.ftw_hp)
Freeness (class in ftw.tf.networks.dnc.addressing)
FTW (class in ftw.agents.tf.ftw.agent)
ftw.adders.reverb (module)
ftw.adders.reverb.multi_sequence (module)
ftw.adders.reverb.multi_sequence_test (module)
ftw.adders.reverb.rnn_sequence (module)
ftw.adders.reverb.rnn_sequence_test (module)
ftw.adders.reverb.sequence (module)
ftw.adders.reverb.sequence_test (module)
ftw.adders.reverb.test_utils (module)
ftw.agents.tf.ftw.utils (module)
ftw.datasets.reverb (module)
ftw.tf.networks.dnc.addressing (module)
ftw.tf.networks.ftw_network (module)
ftw.wrappers.atari (module)
FtwActor (class in ftw.agents.tf.ftw.acting)
FtwLearner (class in ftw.agents.tf.ftw.learning)
FtwNetwork (class in ftw.tf.networks.ftw_network)
FtwTorso (class in ftw.tf.networks.vision)
G
get() (ftw.tf.hyperparameters.base.Hyperparameter method)
(ftw.tf.hyperparameters.ftw_hp.FloatHyperparameter method)
(ftw.tf.hyperparameters.ftw_hp.IntHyperparameter method)
(ftw.tf.internal_reward.InternalRewards method)
get_step() (ftw.agents.tf.ftw.learning.FtwLearner method)
get_variables() (ftw.agents.tf.ftw.learning.FtwLearner method)
get_weights() (ftw.agents.tf.ftw.learning.FtwLearner method)
H
Hyperparameter (class in ftw.tf.hyperparameters.base)
I
initial_state() (ftw.tf.networks.auxiliary.RNNPixelControlNetwork method)
(ftw.tf.networks.dnc.access.MemoryAccess method)
(ftw.tf.networks.dnc.addressing.Freeness method)
(ftw.tf.networks.dnc.addressing.TemporalLinkage method)
(ftw.tf.networks.dnc.dnc.DNC method)
(ftw.tf.networks.ftw_network.FtwNetwork method)
(ftw.tf.networks.recurrence.DNCWrapper method)
(ftw.tf.networks.recurrence.PeriodicVariationalUnit method)
(ftw.tf.networks.recurrence.RPTH method)
,
[1]
(ftw.tf.networks.recurrence.VariationalUnit method)
initialize_hypers() (in module ftw.agents.tf.ftw.utils)
initialize_internal_rewards() (in module ftw.agents.tf.ftw.utils)
InternalRewards (class in ftw.tf.internal_reward)
IntHyperparameter (class in ftw.tf.hyperparameters.ftw_hp)
L
link (ftw.tf.networks.dnc.addressing.TemporalLinkageState attribute)
linkage (ftw.tf.networks.dnc.access.AccessState attribute)
loc (ftw.tf.networks.recurrence.LocScaleDistributionParameters attribute)
LocScaleDistributionParameters (class in ftw.tf.networks.recurrence)
M
make_canonical_atari_environment() (in module ftw.wrappers.atari)
make_reverb_fifo_sampler_dataset() (in module ftw.datasets.reverb)
make_reverb_rnn_sequence_fifo_sampler_dataset() (in module ftw.datasets.reverb)
make_rgb_atari_environment_wo_frame_stack() (in module ftw.wrappers.atari)
make_sequence() (in module ftw.adders.reverb.test_utils)
make_trajectory() (in module ftw.adders.reverb.test_utils)
MarlGymWrapper (class in ftw.wrappers)
memory (ftw.tf.networks.dnc.access.AccessState attribute)
MemoryAccess (class in ftw.tf.networks.dnc.access)
MultiSequenceAdder (class in ftw.adders.reverb)
(class in ftw.adders.reverb.multi_sequence)
MultiSequenceAdderTest (class in ftw.adders.reverb.multi_sequence_test)
MultivariateNormalDiagHead (class in ftw.tf.networks.distributional)
MultivariateNormalDiagLocScaleHead (class in ftw.tf.networks.distributional)
N
NonOverlappingRNNSequenceAdder (class in ftw.adders.reverb)
(class in ftw.adders.reverb.rnn_sequence)
O
OAREmbedding (class in ftw.tf.networks.embedding)
observation_spec() (ftw.wrappers.ObservationActionRewardMarlWrapper method)
ObservationActionRewardMarlWrapper (class in ftw.wrappers)
observe() (ftw.agents.tf.ftw.acting.FtwActor method)
(ftw.agents.tf.ftw.agent.FTW method)
observe_first() (ftw.agents.tf.ftw.acting.FtwActor method)
(ftw.agents.tf.ftw.agent.FTW method)
output (ftw.tf.networks.recurrence.PeriodicRNNState attribute)
output_size (ftw.tf.networks.dnc.access.MemoryAccess attribute)
(ftw.tf.networks.dnc.dnc.DNC attribute)
P
PeriodicRNNState (class in ftw.tf.networks.recurrence)
PeriodicVariationalUnit (class in ftw.tf.networks.recurrence)
perturb() (ftw.tf.hyperparameters.base.Hyperparameter method)
(ftw.tf.hyperparameters.ftw_hp.FloatHyperparameter method)
(ftw.tf.hyperparameters.ftw_hp.IntHyperparameter method)
(ftw.tf.internal_reward.InternalRewards method)
PixelControl (class in ftw.tf.networks.auxiliary)
PolicyValueHead (class in ftw.tf.networks.policy_value)
precedence_weights (ftw.tf.networks.dnc.addressing.TemporalLinkageState attribute)
R
read_weights (ftw.tf.networks.dnc.access.AccessState attribute)
reset() (ftw.adders.reverb.multi_sequence.MultiSequenceAdder method)
(ftw.adders.reverb.MultiSequenceAdder method)
(ftw.adders.reverb.NonOverlappingRNNSequenceAdder method)
(ftw.adders.reverb.SequenceAdder method)
,
[1]
(ftw.adders.reverb.rnn_sequence.NonOverlappingRNNSequenceAdder method)
(ftw.adders.reverb.sequence.SequenceAdder method)
reward() (ftw.tf.internal_reward.InternalRewards method)
reward_spec() (ftw.wrappers.MarlGymWrapper method)
RewardPrediction (class in ftw.tf.networks.auxiliary)
RewardPredictionNetwork (class in ftw.tf.networks.auxiliary)
RNNPixelControlNetwork (class in ftw.tf.networks.auxiliary)
RNNSequenceAdderTest (class in ftw.adders.reverb.rnn_sequence_test)
RPTH (class in ftw.tf.networks.recurrence)
,
[1]
RPTHOutput (class in ftw.tf.networks.recurrence)
RPTHState (class in ftw.tf.networks.recurrence)
run() (ftw.agents.tf.ftw.agent.FTW method)
(ftw.agents.tf.ftw.learning.FtwLearner method)
run_test_adder() (ftw.adders.reverb.multi_sequence_test.MultiSequenceAdderTest method)
(ftw.adders.reverb.rnn_sequence_test.RNNSequenceAdderTest method)
(ftw.adders.reverb.test_utils.AdderTestMixin method)
S
scale (ftw.tf.networks.recurrence.LocScaleDistributionParameters attribute)
select_action() (ftw.agents.tf.ftw.acting.FtwActor method)
(ftw.agents.tf.ftw.agent.FTW method)
(ftw.tf.networks.ftw_network.FtwNetwork method)
SequenceAdder (class in ftw.adders.reverb)
,
[1]
(class in ftw.adders.reverb.sequence)
SequenceAdderTest (class in ftw.adders.reverb.sequence_test)
set() (ftw.tf.hyperparameters.base.Hyperparameter method)
(ftw.tf.hyperparameters.ftw_hp.FloatHyperparameter method)
(ftw.tf.hyperparameters.ftw_hp.IntHyperparameter method)
(ftw.tf.internal_reward.InternalRewards method)
set_step() (ftw.agents.tf.ftw.learning.FtwLearner method)
set_weights() (ftw.agents.tf.ftw.learning.FtwLearner method)
setUp() (ftw.adders.reverb.test_utils.AdderTestMixin method)
state (ftw.agents.tf.ftw.learning.FtwLearner attribute)
step (ftw.tf.networks.recurrence.PeriodicRNNState attribute)
(ftw.tf.networks.recurrence.RPTHState attribute)
step() (ftw.agents.tf.ftw.learning.FtwLearner method)
T
TemporalLinkage (class in ftw.tf.networks.dnc.addressing)
TemporalLinkageState (class in ftw.tf.networks.dnc.addressing)
test_adder_EarlyTerminationNoPadding() (ftw.adders.reverb.sequence_test.SequenceAdderTest method)
test_adder_EarlyTerminationPaddingPeriodOne() (ftw.adders.reverb.sequence_test.SequenceAdderTest method)
test_adder_EarlyTerminationPaddingPeriodTwo() (ftw.adders.reverb.sequence_test.SequenceAdderTest method)
test_adder_EarlyTerminationPeriodOne() (ftw.adders.reverb.sequence_test.SequenceAdderTest method)
test_adder_EarlyTerminationPeriodTwo() (ftw.adders.reverb.sequence_test.SequenceAdderTest method)
test_adder_MinimalTest() (ftw.adders.reverb.rnn_sequence_test.RNNSequenceAdderTest method)
test_adder_multi_sequence_MixedSeqLenPeriodsNoPaddingTest() (ftw.adders.reverb.multi_sequence_test.MultiSequenceAdderTest method)
test_adder_multi_sequence_MixedSeqLenPeriodsTest() (ftw.adders.reverb.multi_sequence_test.MultiSequenceAdderTest method)
test_adder_multi_sequence_RewardPredictionTest() (ftw.adders.reverb.multi_sequence_test.MultiSequenceAdderTest method)
test_adder_PaddingPeriodThree() (ftw.adders.reverb.sequence_test.SequenceAdderTest method)
test_adder_PeriodOne() (ftw.adders.reverb.sequence_test.SequenceAdderTest method)
test_adder_PeriodTwo() (ftw.adders.reverb.sequence_test.SequenceAdderTest method)
test_adder_sequence_SequenceEarlyTerminationNoPadding() (ftw.adders.reverb.multi_sequence_test.MultiSequenceAdderTest method)
test_adder_sequence_SequenceEarlyTerminationPaddingPeriodOne() (ftw.adders.reverb.multi_sequence_test.MultiSequenceAdderTest method)
test_adder_sequence_SequenceEarlyTerminationPaddingPeriodTwo() (ftw.adders.reverb.multi_sequence_test.MultiSequenceAdderTest method)
test_adder_sequence_SequenceEarlyTerminationPeriodOne() (ftw.adders.reverb.multi_sequence_test.MultiSequenceAdderTest method)
test_adder_sequence_SequenceEarlyTerminationPeriodTwo() (ftw.adders.reverb.multi_sequence_test.MultiSequenceAdderTest method)
test_adder_sequence_SequencePaddingPeriodThree() (ftw.adders.reverb.multi_sequence_test.MultiSequenceAdderTest method)
test_adder_sequence_SequencePaddingPeriodTwo() (ftw.adders.reverb.multi_sequence_test.MultiSequenceAdderTest method)
test_adder_sequence_SequencePeriodOne() (ftw.adders.reverb.multi_sequence_test.MultiSequenceAdderTest method)
test_adder_sequence_SequencePeriodTwo() (ftw.adders.reverb.multi_sequence_test.MultiSequenceAdderTest method)
U
unroll() (ftw.tf.networks.auxiliary.RNNPixelControlNetwork method)
(ftw.tf.networks.ftw_network.FtwNetwork method)
update() (ftw.agents.tf.ftw.acting.FtwActor method)
(ftw.agents.tf.ftw.agent.FTW method)
usage (ftw.tf.networks.dnc.access.AccessState attribute)
V
variable (ftw.tf.hyperparameters.ftw_hp.FloatHyperparameter attribute)
(ftw.tf.hyperparameters.ftw_hp.IntHyperparameter attribute)
(ftw.tf.internal_reward.InternalRewards attribute)
VariationalUnit (class in ftw.tf.networks.recurrence)
W
weighted_softmax() (in module ftw.tf.networks.dnc.addressing)
write_allocation_weights() (ftw.tf.networks.dnc.addressing.Freeness method)
write_weights (ftw.tf.networks.dnc.access.AccessState attribute)
writer() (ftw.adders.reverb.test_utils.FakeClient method)
Z
z (ftw.tf.networks.recurrence.RPTHOutput attribute)
(ftw.tf.networks.recurrence.RPTHState attribute)