AdamStelmaszczyk

import           Control.Monad            (when, forM_)
import qualified Data.Judy                as J
import           Data.Maybe               (fromJust)
import qualified Data.Vector.Unboxed      as V
import           Control.Applicative

p10 n = do
    let r  = floor (sqrt (fromIntegral n))
    let n' = n `div` r - 1
    let v1 = V.generate r  $ \i -> n `div` (i + 1)  -- [n `div` i | i <- [1..r]]

AdamStelmaszczyk

def P10(n):
    r = int(n**0.5)
    assert r*r <= n and (r+1)**2 > n
    V = [n//i for i in range(1,r+1)]
    V += list(range(V[-1]-1,0,-1))
    S = {i:i*(i+1)//2-1 for i in V}
    for p in range(2,r+1):
        if S[p] > S[p-1]:  # p is prime
            sp = S[p-1]  # sum of primes smaller than p
            p2 = p*p

AdamStelmaszczyk

import qualified Control.Monad as ControlMonad
import qualified Data.Array as Array
import qualified Data.Array.ST as ArrayST
import qualified Data.Array.Base as ArrayBase
import           Control.DeepSeq
import qualified Data.Map as Map

-- From "The Genuine Sieve of Eratosthenes" by Melissa O'Neill.
-- O(n * log n * log log n)
sieve xs = sieve' xs Map.empty

AdamStelmaszczyk

// $ g++ sparse.cpp --std=c++11
// $ time ./a.out
// sum(2000000000) = 95673602693282040
//
// real	0m0.150s
// user	0m0.147s
// sys	0m0.003s


#include <iostream>

AdamStelmaszczyk

$ python examples/quickstart.py 
[2017-07-24 23:13:32,771] Making new env: CartPole-v0
Finished episode 1 after 40 timesteps (reward: 40.0)
Finished episode 2 after 25 timesteps (reward: 25.0)
Finished episode 3 after 20 timesteps (reward: 20.0)
Finished episode 4 after 36 timesteps (reward: 36.0)
Finished episode 5 after 21 timesteps (reward: 21.0)
Finished episode 6 after 44 timesteps (reward: 44.0)
Finished episode 7 after 12 timesteps (reward: 12.0)
Finished episode 8 after 16 timesteps (reward: 16.0)

AdamStelmaszczyk

$ python examples/quickstart.py 
[2017-07-25 01:30:48,333] Making new env: CartPole-v0
{max_kl_divergence=0.005, states={shape=(4,), type=float}, cg_damping=0.01, network=<function network_builder at 0x7f7e7a055f50>, normalize_advantage=False, actions={continuous=False, num_actions=2}, gae_lambda=0.97, line_search_steps=20, loglevel=info, override_line_search=False, batch_size=100, generalized_advantage_estimation=True, cg_iterations=20, baseline={type=mlp, repeat_update=100, size=32}}
Finished episode 1 after 18 timesteps (reward: 18.0)
Finished episode 2 after 19 timesteps (reward: 19.0)
Finished episode 3 after 44 timesteps (reward: 44.0)
Finished episode 4 after 13 timesteps (reward: 13.0)
Finished episode 5 after 17 timesteps (reward: 17.0)
Finished episode 6 after 22 timesteps (reward: 22.0)
Finished episode 7 after 25 timesteps (reward: 25.0)

AdamStelmaszczyk

# Copyright 2017 reinforce.io. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,

AdamStelmaszczyk

==47003== HEAP SUMMARY:
==47003==     in use at exit: 10,310,600 bytes in 10,626 blocks
==47003==   total heap usage: 34,382,917 allocs, 34,372,291 frees, 6,408,626,531 bytes allocated
==47003== 
==47003== 48 bytes in 1 blocks are definitely lost in loss record 1,256 of 1,984
==47003==    at 0x4C2BBAF: malloc (vg_replace_malloc.c:299)
==47003==    by 0x6EA8FBD: SwigPyClientData_New (in /home/astelma/miniconda3/envs/py2/lib/python2.7/site-packages/opensim/_simbody.so)
==47003==    by 0x6EB4B73: SwigPyIterator_swigregister (in /home/astelma/miniconda3/envs/py2/lib/python2.7/site-packages/opensim/_simbody.so)
==47003==    by 0x4F351E4: call_function (ceval.c:4352)
==47003==    by 0x4F351E4: PyEval_EvalFrameEx (ceval.c:2989)

AdamStelmaszczyk

$ python run_job.py -n 5 -g 60 -c 12 --use_sync --name neptune_job_name
args.offline:  False
('bash command: ', 'srun -A luna -N 5 -n 5 -c 12 -t 6:00:00 distributed_tensorpack_mkl.sh 35083 45394 Breakout-v0 adam 1 "3nodes 12cores" "neptune_job_name_1517179219.79" 0.00015 128 60 0 None 256 100 1 uniform normal False . True 1 False /net/archive/groups/plggluna/intel_2/logs/ 1e-08 0.9 0.999 0 False False False False 120 False')
SLURM_JOB_ID  9486553 ; SLURM_JOB_NAME  distributed_tensorpack_mkl.sh ; SLURM_JOB_NODELIST  p[1340-1341,1343,1347-1348] ; SLURMD_NODENAME  p1341 ; SLURM_JOB_NUM_NODES  5
SLURM_JOB_ID  9486553 ; SLURM_JOB_NAME  distributed_tensorpack_mkl.sh ; SLURM_JOB_NODELIST  p[1340-1341,1343,1347-1348] ; SLURMD_NODENAME  p1343 ; SLURM_JOB_NUM_NODES  5
SLURM_JOB_ID  9486553 ; SLURM_JOB_NAME  distributed_tensorpack_mkl.sh ; SLURM_JOB_NODELIST  p[1340-1341,1343,1347-1348] ; SLURMD_NODENAME  p1347 ; SLURM_JOB_NUM_NODES  5
SLURM_JOB_ID  9486553 ; SLURM_JOB_NAME  distributed_tensorpack_mkl.sh ; SLURM_JOB_NOD

AdamStelmaszczyk

$ python run_job.py -n 5 -g 60 -c 12 --use_sync --name neptune_job_name                    clear^C
[prometheus][plghenrykm@login01 src]$ vim OpenAIGym/train.py 
[prometheus][plghenrykm@login01 src]$ python run_job.py -n 5 -g 60 -c 12 --use_sync --name neptune_job_name
args.offline:  False
('bash command: ', 'srun -A luna -N 5 -n 5 -c 12 -t 6:00:00 distributed_tensorpack_mkl.sh 17351 9236 Breakout-v0 adam 1 "3nodes 12cores" "neptune_job_name_1517240471.8" 0.00015 128 60 0 None 256 100 1 uniform normal False . True 1 False /net/archive/groups/plggluna/intel_2/logs/ 1e-08 0.9 0.999 0 False False False False 120 False')
SLURM_JOB_ID  9495449 ; SLURM_JOB_NAME  distributed_tensorpack_mkl.sh ; SLURM_JOB_NODELIST  p[1567-1568,1577,1580,1584] ; SLURMD_NODENAME  p1584 ; SLURM_JOB_NUM_NODES  5
SLURM_JOB_ID  9495449 ; SLURM_JOB_NAME  distributed_tensorpack_mkl.sh ; SLURM_JOB_NODELIST  p[1567-1568,1577,1580,1584] ; SLURMD_NODENAME  p1580 ; SLURM_JOB_NUM_NODES  5
mkdir: cannot create directory ‘/net/archive/groups/plggluna