def setup_bot(bot):
# Argument Handling
bot.debug = any("debug" in arg.lower() for arg in sys.argv)
# Logging
init_sentry(bot)
bot.sentry = sentry
discord_log = logging.getLogger("discord")
discord_log.setLevel(logging.CRITICAL if not bot.debug else logging.INFO)
log = logging.getLogger("bot")
bot.log = log
log.info(f"\n{get_banner()}\nLoading....")
# Load modules
bot.session = aiohttp.ClientSession(loop=bot.loop)
bot.load_extension("modules.Events.Ready")
# Database
credentials = {
"user": os.getenv("POSTGRES_USER"),
"password": os.getenv("POSTGRES_PASS"),
"database": os.getenv("POSTGRES_DATABASE"),
"host": os.getenv("POSTGRES_HOST"),
"init": init_connection
}
bot.pool = bot.loop.run_until_complete(asyncpg.create_pool(**credentials))
bot.log.info(
f"Postgres connected to database ({bot.pool._working_params.database})"
f" under the ({bot.pool._working_params.user}) user")
# Config
bot.config = get_config
bot.kclient = KClient(api_key=os.getenv("KSOFT"))
bot.spotify = spotipy.Spotify(
client_credentials_manager=SpotifyClientCredentials(
client_id=os.getenv("SPOTIFY_ID"),
client_secret=os.getenv("SPOTIFY_SECRET")))
bot.uptime = datetime.datetime.utcnow()
bot.version = {
"bot": bot.config()["version"],
"python": sys.version.split(" ")[0],
"discord.py": discord.__version__
}
bot.counter = Counter()
bot.commands_used = Counter()
bot.process = psutil.Process()
bot.color = bot.config()["colors"]["main"]
bot.error_color = bot.config()["colors"]["error"]
def count_text(words):
"""
This function will count the words in the list given
:param words: a list of words
:return: a dict where the key is the word and the value is the frequency of the word
"""
return Counter(words)
def build_vocab(self, *args, **kwargs):
"""Add unaligned_token to the list of special symbols."""
counter = Counter()
sources = []
for arg in args: # arg是QEDataset类，里面包括examples和fields
if isinstance(arg, data.Dataset):
sources += [
getattr(arg, name) for name, field in arg.fields.items()
if field is self
] # source是列表，列表中元素是迭代器
else:
sources.append(arg)
for sample in sources:
for x in sample: # 每次循环读取一个样本，将样本处理成list形式，然后更新counter
if not self.sequential:
x = [x]
try:
counter.update(x)
except TypeError:
counter.update(chain.from_iterable(x))
specials = list(
OrderedDict.fromkeys(tok for tok in [
self.unk_token, self.pad_token, self.init_token,
self.eos_token, self.unaligned_token
] if tok is not None)) # ['<unk>', '<pad>', '<unaligned>']
self.vocab = self.vocab_cls(counter, specials=specials, **kwargs)
def build_vocab(self, *args, **kwargs):
"""Add unaligned_token to the list of special symbols."""
counter = Counter()
sources = []
for arg in args:
if isinstance(arg, data.Dataset):
sources += [
getattr(arg, name) for name, field in arg.fields.items()
if field is self
]
else:
sources.append(arg)
for sample in sources:
for x in sample:
if not self.sequential:
x = [x]
try:
counter.update(x)
except TypeError:
counter.update(chain.from_iterable(x))
specials = list(
OrderedDict.fromkeys(tok for tok in [
self.unk_token,
self.pad_token,
self.init_token,
self.eos_token,
self.unaligned_token,
] if tok is not None))
self.vocab = self.vocab_cls(counter, specials=specials, **kwargs)
def reset(self):
self.tok2id = dict()
self.id2tok = dict()
self._add_special_tokens()
self._set_special_token_ids()
self.vocab = Counter()
def get_product_count_dict_from_prokka(ffn_file: Path) -> dict:
""" Reads Prokka fasta file and returns a dict of counts of each product name extracted from the headers """
product_list = []
with open(str(ffn_file)) as f:
for line in f.readlines():
if line.startswith(">"):
line = line.strip()
product = get_product_from_prokka_fasta_header(line)
product_list.append(product)
product_count_dict = dict(Counter(product_list))
return product_count_dict
def get_seperate_list(raw_result):
"""
seperate a n-dimension list into n list with group by each item
ex: gen_seperate_list([[1,'b'],[2,'a']]) >>> [1,2],['b','a']
:param raw_result:
:return:
"""
#get raw_result's item's count
count = Counter(raw_result[0]).__len__()
all = Counter(raw_result).__len__()
i = 0
j = 0
while j < count:
new_list = []
while i < all:
new_list.append(raw_result[i][j])
i += 1
yield (new_list)
i = 0
j += 1
def load_from_vocab_file(self, file):
"""
Load vocabulary from a .vocab file
"""
self.tok2id = dict()
self.id2tok = dict()
self._add_special_tokens()
self._set_special_token_ids()
self.vocab = Counter()
for line in open(file, encoding="utf-8").readlines():
token, count = line.split("\t")
self.vocab[token] = float(count)
self.add_token(token)
def __init__(self,
pad="<pad>",
sos="<sos>",
eos="<eos>",
unk="<unk>",
oovs=0):
self.PAD = pad
self.SOS = sos
self.EOS = eos
self.UNK = unk
self.oovs = oovs
self.vocab = Counter()
self.tok2id = dict()
self.id2tok = dict()
self.size = 0
self.subword = None
def __init__(self,
pad="<pad>",
sos="<sos>",
eos="<eos>",
unk="<unk>",
oovs=0,
file=None,
preprocess=None,
subword=None,
lang=None):
self.PAD = pad
self.SOS = sos
self.EOS = eos
self.UNK = unk
self.PAD_id = 0
self.SOS_id = None
self.EOS_id = None
self.UNK_id = None
self.lang = lang
self.oovs = oovs
self.vocab = Counter()
self.tok2id = dict()
self.id2tok = dict()
self.freqs = dict()
self.gpt2_tokenizer = None
self.is_gpt2 = False
self.subword = subword
if file is not None:
assert preprocess is not None, "Need preprocess() to build vocab!"
self.build(file, preprocess)
def fit_vocab(
self,
samples,
vocab_size=None,
vocab_min_freq=0,
embeddings_name=None,
keep_rare_words_with_embeddings=False,
add_embeddings_vocab=False,
):
tokens = Counter()
for sample in samples:
# TODO: subtokenize?
tokens.update(self.tokenize(sample))
# We use our own Vocabulary class
specials = list(
OrderedDict.fromkeys(
tok for tok in [self.unaligned_token] if tok is not None
)
)
# TODO: handle embeddings/vectors
self.vocab = Vocabulary(
tokens,
max_size=vocab_size,
min_freq=vocab_min_freq,
unk_token=self.unk_token,
pad_token=self.pad_token,
bos_token=self.bos_token,
eos_token=self.eos_token,
specials=specials,
specials_first=self.specials_first,
# TODO: missing vectors, etc.
vectors=None,
rare_with_vectors=keep_rare_words_with_embeddings,
add_vectors_vocab=add_embeddings_vocab,
)
def setup_bot(bot):
bot.config = bot.loop.run_until_complete(get_config())
bot.remove_command('help')
discord_log = logging.getLogger('discord')
log = logging.getLogger("main")
discord_log.setLevel(logging.CRITICAL)
bot.log, bot.logger, = log, log
log.info(f"{get_icon()}\nLoading....\n")
bot.formatter = EmbededHelp()
bot.debug = any('debug' in arg.lower() for arg in sys.argv)
bot.uptime = datetime.datetime.utcnow()
bot.commands_used = Counter()
bot.messages_sent = 0
bot.process = psutil.Process()
bot.at_everyone_seen = 0
bot.session = aiohttp.ClientSession(loop=bot.loop)
bot.keys = bot.loop.run_until_complete(get_keys())
bot.config = bot.loop.run_until_complete(get_config())
bot.conn = bot.loop.run_until_complete(
r.connect("localhost", db=bot.config['db'], port=28015))
bot.loop.run_until_complete(module_loader(bot, "handlers"))
bot.loop.run_until_complete(module_loader(bot, "commands"))
if bot.debug:
discord_log.setLevel(logging.INFO)
def populate_product_occurence_dictionary(self):
""" Counts the number of times any product occurs in self.total_gene_list and stores in dict """
self.product_occurence_dictionary = Counter(self.total_product_list)
def get_member_counts(df: pd.DataFrame) -> dict:
# Get a complete list of all cluster_rep values including duplicates, do total counts
raw_cluster_rep_list = list(df['cluster_rep'])
member_counts = Counter(raw_cluster_rep_list)
return member_counts
def filter_core_cluster_tsv(cluster_tsv: Path, outdir: Path):
"""
Filters/repairs the core cluster .tsv output file to ensure there are no cases where a single sample is represented
in a cluster more than once. Previous versions of MMSeqs2 would sometimes encounter minor issues with clustering,
and this method attempts to fix this. TODO: Verify this is still necessary
:param cluster_tsv: path to cluster .tsv file
:param outdir: path desired output directory
:return: path to filtered/repaired cluster .tsv
"""
# Read TSV
df = cluster_tsv_to_df(cluster_tsv=cluster_tsv)
# TODO: Refactor some of the logic below into clean functions
# Get the cluster member dict
cluster_member_dict = get_cluster_member_dict(df)
# Determine which clusters contain duplicate samples, create a dictionary to store them
to_repair_dict = {}
for cluster_rep, members in cluster_member_dict.items():
sample_id_list = []
for m in members:
sample_id = m.rsplit("_", 1)[0]
sample_id_list.append(sample_id)
counter_dict = Counter(sample_id_list)
problematic_sample_id_dict = {
x: counter_dict[x]
for x in counter_dict if counter_dict[x] > 1
}
if len(problematic_sample_id_dict) >= 1:
log.debug(
f"Warning: Detected problematic samples: {problematic_sample_id_dict}"
)
to_repair_dict[cluster_rep] = problematic_sample_id_dict
else:
continue
# Write a file containing all clusters/members flagged as problematic
debug_dict = {}
for cluster_rep, members in cluster_member_dict.items():
sample_id_list = []
for m in members:
sample_id = m.rsplit("_", 1)[0]
sample_id_list.append(sample_id)
counter_dict = Counter(sample_id_list)
problematic_sample_id_dict = {
x: counter_dict[x]
for x in counter_dict if counter_dict[x] > 1
}
if len(problematic_sample_id_dict) >= 1:
debug_list = []
for s in list(problematic_sample_id_dict.keys()):
for m in members:
if s in m:
debug_list.append(m)
debug_dict[cluster_rep] = debug_list
debug_file = Path(outdir / 'master_genome_DB.cluster.debug.tsv')
with open(str(debug_file), 'w') as f:
for cluster_rep, members in debug_dict.items():
for m in members:
f.write(f"{cluster_rep}\t{m}\n")
# Repair summary tsv by removing rows til only 1 gene per sample is represented within a cluster
# Determine index of rows to filter out
row_indexes_to_remove = []
for i, row in enumerate(df.itertuples()):
cluster_rep = row.cluster_rep
cluster_member = row.cluster_member
if cluster_rep in to_repair_dict.keys():
# Member sample_IDs that need to be reduced
members_dict = to_repair_dict[cluster_rep]
members_list = list(members_dict.keys())
# Detect if this is a row we want to remove
for m in members_list:
if m in cluster_member:
# Check counter dict to see if there is > 1 member for cluster
if to_repair_dict[cluster_rep][m] > 1:
log.debug(f"Repairing {cluster_rep}")
row_indexes_to_remove.append(i)
# Decrement the counter dict
to_repair_dict[cluster_rep][m] -= 1
else:
continue
# Create filtered version of df with duplicate rows removed
filtered_df = df.drop(df.index[row_indexes_to_remove])
# Export to TSV file
outpath = outdir / cluster_tsv.with_suffix(".filtered.tsv").name
filtered_df.to_csv(outpath, sep="\t", header=None, index=None)
get_difference_between_cluster_tsvs(cluster_tsv=cluster_tsv,
filtered_cluster_tsv=outpath)
return outpath
