# ray.tune.error.TuneError: ('Trials did not complete', \[A3C\_A3C\_four\_way\_train-v0\_00000\])

**URL:** <https://discuss.ray.io/t/ray-tune-error-tuneerror-trials-did-not-complete-a3c-a3c-four-way-train-v0-00000/15215>\
**Category:** Uncategorized\
**Created:** [July 13, 2024, 7:02am UTC](https://discuss.ray.io/t/ray-tune-error-tuneerror-trials-did-not-complete-a3c-a3c-four-way-train-v0-00000/15215 "2024-07-13T07:02:19Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![SExpert12](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/sexpert12/32/6311_2.png) [@SExpert12](https://discuss.ray.io/u/SExpert12)\
**Post date:** [July 13, 2024, 7:02am UTC](https://discuss.ray.io/t/ray-tune-error-tuneerror-trials-did-not-complete-a3c-a3c-four-way-train-v0-00000/15215/1 "2024-07-13T07:02:19Z")

</div>

I am using ray for A3C algorithm. My system is Ubuntu with 1 GPU.This is my code:

import gym  
from gym.envs.registration import register  
import macad\_gym # noqa F401  
import argparse,logging  
import os  
from pprint import pprint

import cv2  
import ray,sys,os  
import ray.tune as tune  
from gym.spaces import Box, Discrete  
from macad\_agents.rllib.env\_wrappers import wrap\_deepmind  
from macad\_agents.rllib.models import register\_mnih15\_net

from ray.rllib.agents.a3c.a3c\_tf\_policy import A3CTFPolicy #0.8.5  
from ray.rllib.models.catalog import ModelCatalog  
from ray.rllib.models.preprocessors import Preprocessor  
from ray.tune import register\_env

import time  
import tensorflow as tf  
from tensorboardX import SummaryWriter  
from ray.tune.schedulers import PopulationBasedTraining

parser = argparse.ArgumentParser()  
parser.add\_argument(  
“–env”,  
default=“PongNoFrameskip-v4”,  
help=“Name Gym env. Used only in debug mode. Default=PongNoFrameskip-v4”)  
parser.add\_argument(  
“–disable-comet”,  
action=“store\_true”,  
help=“Disables comet logging. Used for local smoke tests”)  
parser.add\_argument(  
“–num-workers”,  
default=2, #2 #fix  
type=int,  
help=“Num workers (CPU cores) to use”)  
parser.add\_argument(  
“–num-gpus”, default=1, type=int, help=“Number of gpus to use. Default=2”)  
parser.add\_argument(  
“–sample-bs-per-worker”, #one iteration  
default=1024,  
type=int,  
help=“Number of samples in a batch per worker. Default=50”)  
parser.add\_argument(  
“–train-bs”,  
default=128,  
type=int,  
help=“Train batch size. Use as per available GPU mem. Default=500”)  
parser.add\_argument(  
“–envs-per-worker”,  
default=1,  
type=int,  
help=“Number of env instances per worker. Default=10”)  
parser.add\_argument(  
“–notes”,  
default=None,  
help=“Custom experiment description to be added to comet logs”)  
parser.add\_argument(  
“–model-arch”,  
default=“mnih15”,  
help=“Model architecture to use. Default=mnih15”)  
parser.add\_argument(  
“–num-steps”,  
default=4000000,  
type=int,  
help=“Number of steps to train. Default=20M”)  
parser.add\_argument(  
“–num-iters”,  
default=50,  
type=int,  
help=“Number of training iterations. Default=20”)  
parser.add\_argument(  
“–log-graph”,  
action=“store\_true”,  
help=“Write TF graph on Tensorboard for debugging”,default=True)  
parser.add\_argument(  
“–num-framestack”,  
type=int,  
default=4,  
help=“Number of obs frames to stack”)  
parser.add\_argument(  
“–debug”, action=“store\_true”, help=“Run in debug-friendly mode”, default=False)  
parser.add\_argument(  
“–redis-address”,  
default=None,  
help=“Address of ray head node. Be sure to start ray with”  
“ray start --redis-address \<…\> --num-gpus\<.\> before running this script”)  
parser.add\_argument(  
“–use-lstm”, action=“store\_true”, help=“Append a LSTM cell to the model”,default=True)

args = parser.parse\_args()

model\_name = args.model\_arch  
if model\_name == “mnih15”:  
register\_mnih15\_net() # Registers mnih15  
else:  
print(“Unsupported model arch. Using default”)  
register\_mnih15\_net()  
model\_name = “mnih15”  
env\_name=‘A3C\_four\_way\_train-v0’  
env = gym.make(env\_name)  
print (env.spec.max\_episode\_steps,“-±±±±±±±±±±±±±±±±±+”)  
env.spec.max\_episode\_steps=1024  
print (env.spec.max\_episode\_steps,“-±±±±±±±±±±±±±±±±±+”)

env\_actor\_configs = env.configs  
num\_framestack = args.num\_framestack

# env\_config[“env”][“render”] = False

def env\_creator(env\_config):

```
import macad_gym
env = gym.make("A3C_four_way_train-v0")

# Apply wrappers to: convert to Grayscale, resize to 84 x 84,
# stack frames & some more op
env = wrap_deepmind(env, dim=84, num_framestack=num_framestack)
return env

```

env = wrap\_deepmind(env, dim=84, num\_framestack=num\_framestack)

register\_env(env\_name, lambda config: env\_creator(config))

# # Placeholder to enable use of a custom pre-processor

class ImagePreproc(Preprocessor):  
def \_init\_shape(self, obs\_space, options):  
self.shape = (84, 84, 3) # Adjust third dim if stacking frames  
return self.shape

```
def transform(self, observation):
    observation = cv2.resize(observation, (self.shape[0], self.shape[1]))
    return observation

```

def transform(self, observation):  
observation = cv2.resize(observation, (self.shape[0], self.shape[1]))  
return observation

ModelCatalog.register\_custom\_preprocessor(“sq\_im\_84”, ImagePreproc)

if args.redis\_address is not None:  
# num\_gpus (& num\_cpus) must not be provided when connecting to an  
# existing cluster  
ray.init(redis\_address=args.redis\_address,lru\_evict=True, log\_to\_driver=False)  
else:  
ray.init(num\_gpus=args.num\_gpus,lru\_evict=True, log\_to\_driver=False)

config = {  
# Model and preprocessor options.  
“model”: {  
“custom\_model”: model\_name,  
“custom\_options”: {  
# Custom notes for the experiment  
“notes”: {  
“args”: vars(args)  
},  
},  
# NOTE:Wrappers are applied by RLlib if custom\_preproc is NOT specified  
“custom\_preprocessor”: “sq\_im\_84”,  
“dim”: 84,  
“free\_log\_std”: False, # if args.discrete\_actions else True,  
“grayscale”: True,  
# conv\_filters to be used with the custom CNN model.  
# “conv\_filters”: [[16, [4, 4], 2], [32, [3, 3], 2], [16, [3, 3], 2]]  
},  
# preproc\_pref is ignored if custom\_preproc is specified  
# “preprocessor\_pref”: “deepmind”,

```
# env_config to be passed to env_creator

"env_config": env_actor_configs

```

}

def default\_policy():  
env\_actor\_configs[“env”][“render”] = False

```
config = {
# Model and preprocessor options.
"model": {
    "custom_model": model_name,
    "custom_options": {
        # Custom notes for the experiment
        "notes": {
            "args": vars(args)
        },
    },
    # NOTE:Wrappers are applied by RLlib if custom_preproc is NOT specified
    "custom_preprocessor": "sq_im_84",
    "dim": 84,
    "free_log_std": False, # if args.discrete_actions else True,
    "grayscale": True,
    # conv_filters to be used with the custom CNN model.
    # "conv_filters": [[16, [4, 4], 2], [32, [3, 3], 2], [16, [3, 3], 2]]
},

# Should use a critic as a baseline (otherwise don't use value baseline;
# required for using GAE).
"use_critic": True,
# If true, use the Generalized Advantage Estimator (GAE)
# with a value function, see https://arxiv.org/pdf/1506.02438.pdf.
"use_gae": True,
# Size of rollout batch
"rollout_fragment_length": 10,
# GAE(gamma) parameter
"lambda": 1.0,
# Max global norm for each gradient calculated by worker
"grad_clip": 40.0,
"epsilon":
0.1,
# Learning rate
"lr": 0.0001,
# Learning rate schedule
"lr_schedule": None,
# Value Function Loss coefficient
"vf_loss_coeff": 0.5,
# Entropy coefficient
"entropy_coeff": 0.01,
# Min time per iteration
"min_iter_time_s": 5,
# Workers sample async. Note that this increases the effective
# rollout_fragment_length by up to 5x due to async buffering of batches.
"sample_async": True,

# Discount factor of the MDP.
"gamma": 0.9,
# Number of steps after which the episode is forced to terminate. Defaults
# to `env.spec.max_episode_steps` (if present) for Gym envs.
"horizon": 1024,
# Calculate rewards but don't reset the environment when the horizon is
# hit. This allows value estimation and RNN state to span across logical
# episodes denoted by horizon. This only has an effect if horizon != inf.
"soft_horizon": True,
# Don't set 'done' at the end of the episode. Note that you still need to
# set this if soft_horizon=True, unless your env is actually running
# forever without returning done=True.
"no_done_at_end": True,
"monitor": True,

# System params.
# Should be divisible by num_envs_per_worker
"sample_batch_size":
 args.sample_bs_per_worker,
"train_batch_size":
args.train_bs,
# "rollout_fragment_length": 128,
"num_workers":
args.num_workers,
# Number of environments to evaluate vectorwise per worker.
"num_envs_per_worker":
args.envs_per_worker,
"num_cpus_per_worker":
1,
"num_gpus_per_worker":
1,
# "eager_tracing": True,

# # Learning params.
# "grad_clip":
# 40.0,
# "clip_rewards":
# True,
# either "adam" or "rmsprop"
"opt_type":
"adam",
# "lr":
# 0.003,
"lr_schedule": [
    [0, 0.0006],
    [20000000, 0.000000000001], # Anneal linearly to 0 from start 2 end
],
# rmsprop considered
"decay":
0.5,
"momentum":
0.0,

# # balancing the three losses
# "vf_loss_coeff":
# 0.5, # Baseline loss scaling
# "entropy_coeff":
# -0.01,

# preproc_pref is ignored if custom_preproc is specified
# "preprocessor_pref": "deepmind",

```

# “gamma”: 0.99,

```
"use_lstm": args.use_lstm,
# env_config to be passed to env_creator
"env":{
    "render": False
},
# "in_evaluation": True,
# "evaluation_num_episodes": 1,
"env_config": env_actor_configs
}

# pprint (config)
return (A3CTFPolicy, Box(0.0, 255.0, shape=(84, 84, 3)), Discrete(9),config)

```

# pprint (args.checkpoint\_path)

# pprint(os.path.isfile(args.checkpoint\_path))

if args.debug:  
# For checkpoint loading and retraining (not used in this script)  
experiment\_spec = tune.Experiment(  
“multi-carla/” + args.model\_arch,  
“A3C”,  
# restore=args.checkpoint\_path,  
# timesteps\_total is init with None (not 0) which causes issue  
# stop={“timesteps\_total”: args.num\_steps},  
stop={“timesteps\_since\_restore”: args.num\_steps},  
config=config,  
# checkpoint\_freq=1000, #1000  
# checkpoint\_at\_end=True,  
resources\_per\_trial={  
“cpu”: 1,  
“gpu”: 1  
})

```
experiment_spec = tune.run_experiments({
        "MA-Inde-A3C-SUI1B2C1PCARLA": {
            "run": "A3C",
            "env": env_name,
            "stop": {
                
                "training_iteration": args.num_iters,
                "timesteps_total": args.num_steps,
                "episodes_total": 1024,
            },
            # "restore":args.checkpoint_path,   
            "config": {

                "log_level": "DEBUG",
               # "num_sgd_iter": 10, # Enables Experience Replay
                "multiagent": {
                    "policies": {
                        id: default_policy()
                        for id in env_actor_configs["actors"].keys()
                    },
                    "policy_mapping_fn":
                    tune.function(lambda agent_id: agent_id),
                    "policies_to_train": ["car2","car3"],
                },
                "env_config": env_actor_configs,
                "num_workers": args.num_workers,
                "num_envs_per_worker": args.envs_per_worker,
                "sample_batch_size": args.sample_bs_per_worker,
                "train_batch_size": args.train_bs,
                "horizon": 512,

            },
            "checkpoint_freq": 5,
            "checkpoint_at_end": True,

        }
    })

```

else:

```
pbt = PopulationBasedTraining(
time_attr=args.num_iters,
metric ='episode_reward_mean',
mode = 'max',
# reward_attr='car2PPO/policy_reward_mean',
perturbation_interval=2,
resample_probability=0.5,
quantile_fraction=0.5, # copy bottom % with top %
# Specifies the search space for these hyperparams
hyperparam_mutations={
    # "lambda": [0.9, 1.0],
    # "clip_param": [0.1, 0.5],
    "lr":[1e-3, 1e-5],
},
log_config=True,)
# custom_explore_fn=explore)

analysis = tune.run(
    "A3C",
    name="A3C_Four_way",
    scheduler=pbt,
    verbose=1,
    reuse_actors=True,
    # num_samples=args.num_samples,
    stop={
            # "timesteps_since_restore": args.num_steps,
            "training_iteration": args.num_iters,
            "timesteps_total": args.num_steps,
            "episodes_total": 500,},

    config= {
                "env": env_name,
                "log_level": "DEBUG",
            # "num_sgd_iter": 4, # Enables Experience Replay
                "multiagent": {
                    "policies": {
                        id: default_policy()
                        for id in env_actor_configs["actors"].keys()
                    },
                    "policy_mapping_fn":
                    tune.function(lambda agent_id: agent_id),
                    "policies_to_train": ["carA3C"], #car2PPO is Autonomous driving models
                },
                "env_config": env_actor_configs,
                "num_workers": args.num_workers,
                "num_envs_per_worker": args.envs_per_worker,
                "sample_batch_size": args.sample_bs_per_worker,
                "train_batch_size": args.train_bs,
                #"horizon": 512, #yet to be fixed

            },
        checkpoint_freq = 5,
        checkpoint_at_end = True,    
    )

```

ray.shutdown()

Training get started but later I get error like this:

2024-07-13 04:25:08.938498: W tensorflow/stream\_executor/platform/default/dso\_loader.cc:55] Could not load dynamic library ‘libnvinfer.so.6’; dlerror: libnvinfer.so.6: cannot open shared object file: No such file or directory  
2024-07-13 04:25:08.938568: W tensorflow/stream\_executor/platform/default/dso\_loader.cc:55] Could not load dynamic library ‘libnvinfer\_plugin.so.6’; dlerror: libnvinfer\_plugin.so.6: cannot open shared object file: No such file or directory  
2024-07-13 04:25:08.938577: W tensorflow/compiler/tf2tensorrt/utils/py\_utils.cc:30] Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.  
None -±±±±±±±±±±±±±±±±±+  
1024 -±±±±±±±±±±±±±±±±±+  
2024-07-13 04:25:09,896 INFO resource\_spec.py:212 – Starting Ray with 13.48 GiB memory available for workers and up to 6.74 GiB for objects. You can adjust these settings with ray.init(memory=, object\_store\_memory=).  
2024-07-13 04:25:10,257 INFO services.py:1148 – View the Ray dashboard at localhost:8267  
2024-07-13 04:25:10,386 WARNING sample.py:27 – DeprecationWarning: wrapping \<function at 0x7f23fb88dd08\> with tune.function() is no longer needed  
2024-07-13 04:25:10,519 ERROR trial\_executor.py:64 – Trial A3C\_A3C\_four\_way\_train-v0\_00000: Error checkpointing trial metadata.  
Traceback (most recent call last):  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/tune/trial\_executor.py”, line 61, in try\_checkpoint\_metadata  
self.\_cached\_trial\_state[trial.trial\_id] = trial. **getstate** ()  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/tune/trial.py”, line 551, in **getstate**  
self.result\_logger.flush(sync\_down=False)  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/tune/logger.py”, line 330, in flush  
\_logger.flush()  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/tune/logger.py”, line 237, in flush  
self.\_file\_writer.flush()  
AttributeError: ‘SummaryWriter’ object has no attribute ‘flush’  
== Status ==  
Memory usage on this node: 9.8/31.2 GiB  
PopulationBasedTraining: 0 checkpoints, 0 perturbs  
Resources requested: 3/16 CPUs, 0/1 GPUs, 0.0/13.48 GiB heap, 0.0/4.64 GiB objects  
Result logdir: /home/ray\_results/A3C\_Four\_way  
Number of trials: 1 (1 RUNNING)  
±--------------------------------±---------±------+  
| Trial name | status | loc |  
|---------------------------------±---------±------|  
| A3C\_A3C\_four\_way\_train-v0\_00000 | RUNNING | |  
±--------------------------------±---------±------+

2024-07-13 04:35:16,915 ERROR trial\_runner.py:521 – Trial A3C\_A3C\_four\_way\_train-v0\_00000: Error processing event.  
Traceback (most recent call last):  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/tune/trial\_runner.py”, line 467, in \_process\_trial  
result = self.trial\_executor.fetch\_result(trial)  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/tune/ray\_trial\_executor.py”, line 381, in fetch\_result  
result = ray.get(trial\_future[0], DEFAULT\_GET\_TIMEOUT)  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/worker.py”, line 1513, in get  
raise value.as\_instanceof\_cause()  
ray.exceptions.RayTaskError(Empty): ray::A3C.train() (pid=28802, ip=192.168.15.93)  
File “python/ray/\_raylet.pyx”, line 452, in ray.\_raylet.execute\_task  
File “python/ray/\_raylet.pyx”, line 407, in ray.\_raylet.execute\_task.function\_executor  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/rllib/agents/trainer.py”, line 502, in train  
raise e  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/rllib/agents/trainer.py”, line 491, in train  
result = Trainable.train(self)  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/tune/trainable.py”, line 261, in train  
result = self.\_train()  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/rllib/agents/trainer\_template.py”, line 142, in \_train  
return self.\_train\_exec\_impl()  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/rllib/agents/trainer\_template.py”, line 174, in \_train\_exec\_impl  
res = next(self.train\_exec\_impl)  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/util/iter.py”, line 634, in **next**  
return next(self.built\_iterator)  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/util/iter.py”, line 644, in apply\_foreach  
for item in it:  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/util/iter.py”, line 685, in apply\_filter  
for item in it:  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/util/iter.py”, line 644, in apply\_foreach  
for item in it:  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/util/iter.py”, line 670, in add\_wait\_hooks  
item = next(it)  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/util/iter.py”, line 644, in apply\_foreach  
for item in it:  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/util/iter.py”, line 470, in base\_iterator  
yield ray.get(obj\_id)  
ray.exceptions.RayTaskError(Empty): ray::RolloutWorker.par\_iter\_next() (pid=28805, ip=192.168.15.93)  
File “python/ray/\_raylet.pyx”, line 452, in ray.\_raylet.execute\_task  
File “python/ray/\_raylet.pyx”, line 407, in ray.\_raylet.execute\_task.function\_executor  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/util/iter.py”, line 961, in par\_iter\_next  
return next(self.local\_it)  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/util/iter.py”, line 644, in apply\_foreach  
for item in it:  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/rllib/evaluation/rollout\_worker.py”, line 251, in gen\_rollouts  
yield self.sample()  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/rllib/evaluation/rollout\_worker.py”, line 492, in sample  
batches = [self.input\_reader.next()]  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/rllib/evaluation/sampler.py”, line 53, in next  
batches = [self.get\_data()]  
File “/home/miniconda3/envs/newbench/lib/python3.6/site-packages/ray/rllib/evaluation/sampler.py”, line 199, in get\_data  
rollout = self.queue.get(timeout=600.0)  
File “/home/miniconda3/envs/newbench/lib/python3.6/queue.py”, line 172, in get  
raise Empty  
queue.Empty  
== Status ==  
Memory usage on this node: 11.6/31.2 GiB  
PopulationBasedTraining: 0 checkpoints, 0 perturbs  
Resources requested: 0/16 CPUs, 0/1 GPUs, 0.0/13.48 GiB heap, 0.0/4.64 GiB objects  
Result logdir: /home/ray\_results/A3C\_Four\_way  
Number of trials: 1 (1 ERROR)

 ![Screenshot from 2024-07-13 05-02-50](https://us1.discourse-cdn.com/flex020/uploads/ray/original/2X/e/ebae10caecaba7dcbecb693c45f6a7b629e2bde7.png)

How to solve this?