exp/lm_train_lm_en_bpe500/config.yaml

config: conf/train_lm.yaml
print_config: false
log_level: INFO
dry_run: false
iterator_type: sequence
output_dir: exp/lm_train_lm_en_bpe500
ngpu: 1
seed: 0
num_workers: 1
num_att_plot: 3
dist_backend: nccl
dist_init_method: env://
dist_world_size: null
dist_rank: null
local_rank: 0
dist_master_addr: null
dist_master_port: null
dist_launcher: null
multiprocessing_distributed: false
unused_parameters: false
sharded_ddp: false
cudnn_enabled: true
cudnn_benchmark: false
cudnn_deterministic: true
collect_stats: false
write_collected_feats: false
max_epoch: 20
patience: 3
val_scheduler_criterion:
- valid
- loss
early_stopping_criterion:
- valid
- loss
- min
best_model_criterion:
-   - valid
    - loss
    - min
keep_nbest_models: 1
grad_clip: 5.0
grad_clip_type: 2.0
grad_noise: false
accum_grad: 1
no_forward_run: false
resume: true
train_dtype: float32
use_amp: false
log_interval: null
use_tensorboard: true
use_wandb: false
wandb_project: null
wandb_id: null
detect_anomaly: false
pretrain_path: null
init_param: []
freeze_param: []
num_iters_per_epoch: null
batch_size: 64
valid_batch_size: null
batch_bins: 1000000
valid_batch_bins: null
train_shape_file:
- exp/lm_stats_en_bpe500/train/text_shape.bpe
valid_shape_file:
- exp/lm_stats_en_bpe500/valid/text_shape.bpe
batch_type: folded
valid_batch_type: null
fold_length:
- 150
sort_in_batch: descending
sort_batch: descending
multiple_iterator: false
chunk_length: 500
chunk_shift_ratio: 0.5
num_cache_chunks: 1024
train_data_path_and_name_and_type:
-   - dump/raw/lm_train.txt
    - text
    - text
valid_data_path_and_name_and_type:
-   - dump/raw/dev/text
    - text
    - text
allow_variable_data_keys: false
max_cache_size: 0.0
max_cache_fd: 32
valid_max_cache_size: null
optim: sgd
optim_conf: {}
scheduler: null
scheduler_conf: {}
token_list:
- <blank>
- <unk>
- s
- ▁the
- t
- ▁a
- e
- ▁and
- d
- ▁to
- ▁of
- ''''
- ing
- ▁in
- ▁that
- ▁i
- m
- p
- a
- y
- n
- i
- c
- ▁it
- ▁we
- ▁you
- ed
- o
- l
- ▁is
- u
- re
- ▁
- g
- r
- ▁this
- ar
- er
- b
- ▁so
- al
- in
- or
- ▁s
- h
- ▁for
- ▁f
- ▁be
- k
- ▁c
- ▁do
- le
- ▁t
- ▁re
- ▁w
- ve
- f
- es
- it
- w
- ▁was
- ▁e
- ▁they
- ic
- ▁on
- ly
- ▁are
- ra
- ▁have
- ▁p
- ▁what
- ▁with
- ▁m
- ent
- ▁can
- 'on'
- an
- ▁but
- en
- ll
- v
- at
- ri
- ▁st
- th
- ur
- ▁one
- ▁he
- il
- ce
- ▁about
- ▁me
- ▁de
- ▁all
- ▁not
- ▁g
- ▁ma
- ch
- ▁there
- ter
- ▁at
- ation
- ▁my
- ▁as
- ▁n
- ▁mo
- el
- ▁like
- ▁people
- ▁con
- ▁an
- ▁or
- ▁if
- ro
- ▁from
- ▁co
- ate
- ▁d
- ver
- ▁these
- ▁now
- ▁out
- ▁our
- ▁know
- ▁see
- ▁su
- ion
- ▁just
- ▁ch
- us
- ol
- ▁ex
- ect
- ne
- ▁when
- ▁o
- ul
- ▁b
- te
- ▁very
- ▁go
- ies
- ▁think
- ▁because
- om
- ▁who
- ▁up
- un
- tion
- ▁us
- ▁le
- ir
- ity
- ▁get
- ▁pa
- ▁how
- as
- ive
- ▁by
- et
- ▁bo
- ▁se
- ▁ca
- ▁which
- ▁some
- ▁no
- ▁going
- ▁pro
- ▁po
- lo
- ng
- ▁look
- ▁time
- ▁more
- ▁had
- ▁would
- ▁here
- am
- ment
- ▁really
- la
- is
- ▁your
- x
- ▁ho
- ▁k
- ge
- ist
- ▁them
- ry
- ▁two
- id
- ck
- ▁fa
- ▁world
- ▁sp
- ure
- ▁way
- ide
- ▁don
- ▁want
- ▁work
- ▁lo
- ▁were
- ru
- est
- ers
- ▁their
- age
- ▁li
- ut
- ▁ba
- ▁en
- ▁other
- ine
- ow
- ▁actually
- ▁sh
- able
- ▁make
- ▁un
- ac
- ▁la
- op
- ▁into
- ▁where
- ▁could
- ▁has
- z
- vi
- ous
- ▁will
- ▁di
- ight
- ▁things
- ally
- ▁ro
- ▁say
- ▁vi
- ▁years
- ▁ra
- ▁than
- ▁br
- ther
- ical
- ▁ha
- ▁comp
- ▁did
- ▁then
- ok
- ▁new
- ▁any
- ound
- ▁well
- ▁she
- um
- ▁been
- ▁need
- ▁right
- ▁thousand
- ▁take
- ▁start
- ▁something
- ta
- ▁over
- he
- ▁hundred
- ▁com
- ▁kind
- j
- ▁first
- ▁those
- ▁pre
- ▁said
- ize
- tic
- ▁thing
- ▁even
- ▁fi
- ia
- ▁back
- ▁every
- ▁little
- nce
- ▁car
- ▁sc
- ▁much
- ▁life
- ▁also
- ▁hi
- ▁bi
- if
- ▁most
- os
- ▁three
- ▁lot
- ▁got
- ▁sa
- ▁come
- ▁talk
- iv
- ▁te
- ▁inter
- ▁per
- ▁use
- ant
- ▁bu
- ▁part
- ▁his
- ▁happen
- ance
- ▁mean
- ish
- ▁dr
- ▁day
- ▁only
- ▁many
- ▁different
- ten
- ▁show
- ▁down
- ▁dis
- ful
- ▁good
- ated
- ▁idea
- ▁around
- ▁ga
- ▁human
- ▁five
- ▁put
- ▁through
- ▁why
- ▁real
- ▁change
- ff
- ▁let
- ible
- ▁fact
- ▁same
- ▁jo
- ▁her
- ▁problem
- ▁live
- ▁four
- ▁year
- ▁give
- ▁ri
- ick
- ▁though
- ▁big
- ▁tell
- ▁great
- ▁ta
- ▁va
- ition
- ian
- ▁system
- ▁six
- ▁plan
- ▁place
- ▁build
- ▁called
- ▁point
- ▁again
- ▁twenty
- ▁percent
- port
- ▁ten
- ▁find
- ▁after
- ▁app
- ▁nine
- ▁qu
- ▁fl
- ▁long
- ▁eight
- ▁gene
- ▁design
- ▁today
- ▁made
- ▁imp
- ▁should
- ▁came
- ▁last
- ▁learn
- ▁own
- ▁turn
- ▁seven
- ▁high
- ▁question
- ▁person
- ▁another
- ▁brain
- ▁important
- ▁create
- ▁act
- ▁trans
- ▁hu
- ▁power
- ness
- land
- ▁play
- ▁sort
- ▁old
- ▁course
- ▁before
- ▁feel
- ▁understand
- q
- ▁each
- ▁million
- ▁together
- ▁better
- ▁ago
- ▁example
- ▁help
- ▁story
- ▁next
- ▁hand
- ▁school
- ▁develop
- ▁technology
- ▁under
- form
- ▁second
- que
- ▁grow
- ▁still
- ▁believe
- ▁cell
- ▁light
- ▁number
- ▁small
- ▁between
- ▁become
- ▁america
- ▁maybe
- ▁space
- ▁project
- ▁organ
- ▁vo
- ▁home
- ▁children
- graph
- ▁picture
- ▁fifty
- ▁africa
- ▁thirty
- ▁reason
- ▁health
- ▁large
- ▁hard
- ▁sense
- ▁computer
- ▁always
- ▁women
- ▁money
- ▁everything
- ▁information
- ▁country
- ▁teach
- ▁energy
- ▁run
- ▁experience
- ▁food
- ▁process
- ▁quite
- qua
- ▁future
- ▁interesting
- ization
- ▁enough
- ▁science
- ▁stuff
- ▁social
- ▁probably
- ▁friend
- ▁ja
- ▁half
- ability
- ▁group
- ▁animal
- ▁wonder
- ▁product
- ▁close
- ▁simple
- ▁piece
- ▁billion
- ▁control
- ▁sign
- ▁environment
- ▁global
- ▁imagine
- ▁basically
- ▁answer
- ō
- ă
- ť
- <sos/eos>
init: null
model_conf:
    ignore_id: 0
use_preprocessor: true
token_type: bpe
bpemodel: data/en_token_list/bpe_unigram500/bpe.model
non_linguistic_symbols: null
cleaner: null
g2p: null
lm: seq_rnn
lm_conf:
    unit: 650
    nlayers: 2
required:
- output_dir
- token_list
version: 0.9.8
distributed: false