How to deal with IOError: [Errno 32] Broken pipe

[tuyunbin]

I download coco caption two days ago. Unfortunately，I meet a trouble when I run metricsl.py.It is normal when computed bleu, but when computed the meteor,I meet a error as followed: computing METEOR score... Traceback (most recent call last): File "metrics.py", line 203, in test_cocoeval() File "metrics.py", line 199, in test_cocoeval valid_score, test_score = score_with_cocoeval(samples_valid, samples_test, engine) File "metrics.py", line 92, in score_with_cocoeval valid_score = scorer.score(gts_valid, samples_valid, engine.valid_ids) File "/home/tuyunbin/vedio caption/arctic-capgen-vid-master/cocoeval.py", line 42, in score score, scores = scorer.compute_score(gts, res) File "/home/tuyunbin/vedio caption/arctic-capgen-vid-master/pycocoevalcap/meteor/meteor.py", line 37, in compute_score stat = self._stat(res[i][0], gts[i]) File "/home/tuyunbin/vedio caption/arctic-capgen-vid-master/pycocoevalcap/meteor/meteor.py", line 55, in _stat self.meteor_p.stdin.write('{}\n'.format(score_line)) IOError: [Errno 32] Broken pipe Finally,I followed the Li Yao's readme way to add this 'self.meteor_p.kill()' at 45 line of meteor.py ,but the result is still the same, and now I am very anxious.

[endernewton]

It was later found out that there are white spaces in naming the folder, which caused the problem.

-Xinlei

On Sat, Sep 24, 2016 at 9:14 AM, tuynbin notifications@github.com wrote:

hello,Two days ago, I downloaded your code ,It is normal when computed bleu,but when computed meteor,I meet the trouble as followed: computing METEOR score... Traceback (most recent call last): File "/home/tuyunbin/vedio caption/arctic-capgen-vid-master/metrics.py", line 203, in test_cocoeval() File "/home/tuyunbin/vedio caption/arctic-capgen-vid-master/metrics.py", line 199, in test_cocoeval valid_score, test_score = score_with_cocoeval(samples_valid, samples_test, engine) File "/home/tuyunbin/vedio caption/arctic-capgen-vid-master/metrics.py", line 92, in score_with_cocoeval valid_score = scorer.score(gts_valid, samples_valid, engine.valid_ids) File "/home/tuyunbin/vedio caption/arctic-capgen-vid-master/cocoeval.py", line 42, in score score, scores = scorer.compute_score(gts, res) File "/home/tuyunbin/vedio caption/arctic-capgen-vid- master/pycocoevalcap/meteor/meteor.py", line 37, in compute_score stat = self._stat(res[i][0], gts[i]) File "/home/tuyunbin/vedio caption/arctic-capgen-vid- master/pycocoevalcap/meteor/meteor.py", line 55, in _stat self.meteor_p.stdin.write('{}\n'.format(score_line)) IOError: [Errno 32] Broken pipe I follow the readme way to add that line to 45-line position, but the result is still the same, and now I am very anxious.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/tylin/coco-caption/issues/17, or mute the thread https://github.com/notifications/unsubscribe-auth/ACTwEZkBXLutNpW3iIJApnR5RC2j-t-kks5qtSJQgaJpZM4KFo9y .

[thisray]

I don't use the space in the path, but the problem still happened...

{'reflen': 42989, 'guess': [43141, 39089, 35037, 30985], 'testlen': 43141, 'correct': [24003, 7902, 2445, 759]}
ratio: 1.00353578822
Traceback (most recent call last):
  File "train_new.py", line 59, in <module>
    main()
  File "train_new.py", line 37, in main
    solver.train()
  File "/home/thisray/Code/lab7/Lab7_run_HERE/core/solver.py", line 172, in train
    scores = evaluate(data_path='./data', split='val', get_scores=True)
  File "/home/thisray/Code/lab7/Lab7_run_HERE/core/bleu.py", line 55, in evaluate
    final_scores = score(ref, hypo)
  File "/home/thisray/Code/lab7/Lab7_run_HERE/core/bleu.py", line 20, in score
    score,scores = scorer.compute_score(ref,hypo)
  File "../coco_caption/pycocoevalcap/meteor/meteor.py", line 38, in compute_score
    stat = self._stat(res[i][0], gts[i])
  File "../coco_caption/pycocoevalcap/meteor/meteor.py", line 57, in _stat
    self.meteor_p.stdin.write('{}\n'.format(score_line))
IOError: [Errno 32] Broken pipe

---

and if I don't add shell=True in subprocess.Popen(), it would show another error.

self.meteor_p = subprocess.Popen(self.meteor_cmd, \
cwd=os.path.dirname(os.path.abspath(__file__)), stdin=subprocess.PIPE, \
stdout=subprocess.PIPE, stderr=subprocess.PIPE, shell=True)

[endernewton]

Can you provide a small piece of examples that you used for evaluation?

---

Ender X. Chen

Language Technologies Institute, School of Computer Science, Carnegie Mellon University, Pittsburgh, PA

On Fri, Apr 21, 2017 at 10:45 AM, thisray notifications@github.com wrote:

I don't use the space in the path, but the problem still happened...

{'reflen': 42989, 'guess': [43141, 39089, 35037, 30985], 'testlen': 43141, 'correct': [24003, 7902, 2445, 759]} ratio: 1.00353578822 Traceback (most recent call last): File "train_new.py", line 59, in main() File "train_new.py", line 37, in main solver.train() File "/home/thisray/Code/lab7/Lab7_run_HERE/core/solver.py", line 172, in train scores = evaluate(data_path='./data', split='val', get_scores=True) File "/home/thisray/Code/lab7/Lab7_run_HERE/core/bleu.py", line 55, in evaluate final_scores = score(ref, hypo) File "/home/thisray/Code/lab7/Lab7_run_HERE/core/bleu.py", line 20, in score score,scores = scorer.compute_score(ref,hypo) File "../coco_caption/pycocoevalcap/meteor/meteor.py", line 38, in compute_score stat = self._stat(res[i][0], gts[i]) File "../coco_caption/pycocoevalcap/meteor/meteor.py", line 57, in _stat self.meteor_p.stdin.write('{}\n'.format(score_line)) IOError: [Errno 32] Broken pipe

---

and if I don't add shell=True in subprocess.Popen(), it would show another error.

self.meteor_p = subprocess.Popen(self.meteor_cmd, \ cwd=os.path.dirname(os.path.abspath(file)), stdin=subprocess.PIPE, \ stdout=subprocess.PIPE, stderr=subprocess.PIPE, shell=True)

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/tylin/coco-caption/issues/17#issuecomment-296210722, or mute the thread https://github.com/notifications/unsubscribe-auth/ACTwEa_KAjFKAWfflREWKq0CmNr5brB6ks5ryMDtgaJpZM4KFo9y .

[thisray]

final_scores = score(ref, hypo)

ref and hypo are both dict {}

ref: {0: [u'a shoe rack with some shoes and a dog sleeping on them .', 
        u'a small dog is curled up on top of the shoes .', 
        u'various slides and other footwear rest in a metal basket outdoors .', 
        u'a dog sleeping on a show rack in the shoes .', 
        u'this wire metal rack holds several pairs of shoes and sandals .'], 
      1: [...], ...  }

hypo: {0: [u'a large white plate with a white plate with a white plate .'], 
       1: [u'a clock tower with a clock on a table .'], 
       2: [u'a man is standing in a skateboard in a city .'], ... }

I upload them Here thanks!

[endernewton]

Wait. are these coco captions? What is the score function you have? Are you using COCOEval?

[thisray]

Sorry, I don't explain clearly. The final_scores = score(ref, hypo) is from Here.

The score function: (/show-attend-and-tell/core/bleu.py)

sys.path.append('../coco_caption')
from pycocoevalcap.bleu.bleu import Bleu
from pycocoevalcap.rouge.rouge import Rouge
from pycocoevalcap.cider.cider import Cider
from pycocoevalcap.meteor.meteor import Meteor

def score(ref, hypo):
    scorers = [
        (Bleu(4),["Bleu_1","Bleu_2","Bleu_3","Bleu_4"]),
        (Meteor(),"METEOR"),
        (Rouge(),"ROUGE_L"),
        (Cider(),"CIDEr")
    ]
    final_scores = {}
    for scorer,method in scorers:
        score,scores = scorer.compute_score(ref,hypo)  # problem here
        if type(score)==list:
            for m,s in zip(method,score):
                final_scores[m] = s
        else:
            final_scores[method] = score

    return final_scores

And in meteor.py: (/coco_caption/pycocoevalcap/meteor/meteor.py)

def compute_score(self, gts, res):
    assert(gts.keys() == res.keys())
    imgIds = gts.keys()
    scores = []

    eval_line = 'EVAL'
    self.lock.acquire()
    for i in imgIds:
        assert(len(res[i]) == 1)
        stat = self._stat(res[i][0], gts[i])  # problem here
        eval_line += ' ||| {}'.format(stat)

    self.meteor_p.stdin.write('{}\n'.format(eval_line))
    for i in range(0,len(imgIds)):
        scores.append(float(self.meteor_p.stdout.readline().strip()))
    score = float(self.meteor_p.stdout.readline().strip())
    self.lock.release()

    return score, scores

def _stat(self, hypothesis_str, reference_list):
    # SCORE ||| reference 1 words ||| reference n words ||| hypothesis words
    hypothesis_str = hypothesis_str.replace('|||','').replace('  ',' ')
    score_line = ' ||| '.join(('SCORE', ' ||| '.join(reference_list), hypothesis_str))
    self.meteor_p.stdin.write('{}\n'.format(score_line))  # problem here
    return self.meteor_p.stdout.readline().strip()

The IOError: [Errno 32] Broken pipe problem happened on meteor.py, so I asked my problem here.. If its my misunderstanding, I'm very sorry that I wasted your time. thanks!

[endernewton]

You might want to compare the code here and the code there. You may also want to replace the jar file to debug.

[bj2016]

It might be a problem of java. Try this:

1. Install java on your linux system, if it wasn't installed.
2. In the initialization function of meteor.py:

   def __init__(self):
   self.meteor_cmd = ['java', '-jar', '-Xmx2G', METEOR_JAR, '-', '-', '-stdio', '-l', 'en', '-norm']
   self.meteor_p = subprocess.Popen(' '.join(self.meteor_cmd),\     # change this line
       cwd=os.path.dirname(os.path.abspath(__file__)), \
       stdin=subprocess.PIPE, \
       stdout=subprocess.PIPE, \
       stderr=subprocess.PIPE,\
       shell=True)
       # Used to guarantee thread safety
       self.lock = threading.Lock()

[thisray]

Yes, I tried some commands about java (but I forgot is it about install or not) and it can work now. Thanks a lot!

[Bharry1995]

@BruceJing2017 i tried doing what you said. But still I am getting a Broken Pipe error. self.meteor_p.stdin.write('{}\n'.format(score_line))

At this line.

[Noahsark]

I have exactly same problem and finally find the reason. For my case, it simply because I process meteor.compute_score twice. After the first run meteor_p has been killed so the second run is fail. To solve this I just delete the line "self.meteor_p.kill()". This works for me. Or you may delete meteor scorer first and init a new one in second time.

[qlwang25]

I also have the problem, who can help me? Thanks very much.

File "train.py", line 179, in train
    score = eval(epoch)
  File "train.py", line 245, in eval
    result = utils.eval_meteor(reference, candidate)
  File "/home/wangqianlong/model/WEAN-master/data/utils.py", line 99, in eval_meteor
    score, scores = Meteor().compute_score(ref, hypo)
  File "/home/wangqianlong/model/WEAN-master/data/script/meteor.py", line 38, in compute_score
    stat = self._stat(res[i][0], gts[i])
  File "/home/wangqianlong/model/WEAN-master/data/script/meteor.py", line 59, in _stat
    self.meteor_p.stdin.flush()
BrokenPipeError: [Errno 32] Broken pipe

[fuzihaofzh]

I have also encountered with this problem, solution are as follows:

remove all temp files at "pycocoevalcap/tokenizer". It's something like this:

pycocoevalcap/tokenizer/tmpB9Jv0J
pycocoevalcap/tokenizer/tmpoo7WGG

[akira-l]

meet the same problem i installed java, removed temp file, and changed codes mentioned above, but still can not run a simple demo

[akira-l]

the code mainly comes from the example in this repo `from pycocotools.coco import COCO

from pycocoevalcap.eval import COCOEvalCap

dataDir='.'

dataType='val2014'

algName = 'fakecap'

annFile='%s/annotations/captions_%s.json'%(dataDir,dataType)

subtypes=['results', 'evalImgs', 'eval']

[resFile, evalImgsFile, evalFile]= \

['%s/results/captions%s%s_%s.json'%(dataDir,dataType,algName,subtype) for subtype in subtypes]

coco = COCO(annFile)

cocoRes = coco.loadRes(resFile)

cocoEval = COCOEvalCap(coco, cocoRes)

cocoEval.params['image_id'] = cocoRes.getImgIds()

cocoEval.evaluate()

cocoEval = COCOEvalCap(coco, cocoRes)

for metric, score in cocoEval.eval.items():

 print('%s: %.3f'%(metric, score))`

and always has the error: `Traceback (most recent call last):

File "cap_eval.py", line 20, in

cocoEval.evaluate()

File "/home/Up-Down-Captioner/external/coco-caption/pycocoevalcap/eval.py", line 52, in evaluate

score, scores = scorer.compute_score(gts, res)

File "/home/Up-Down-Captioner/external/coco-caption/pycocoevalcap/meteor/meteor.py", line 39, in compute_score

stat = self._stat(res[i][0], gts[i])

File "/home/Up-Down-Captioner/external/coco-caption/pycocoevalcap/meteor/meteor.py", line 58, in _stat

self.meteor_p.stdin.write('{}\n'.format(score_line))

IOError: [Errno 32] Broken pipe`

[dahuang37]

I had the same problem before as well (but I was using Python 3 version from this repo). My fix was adding the data file when running java cmd. Hope this helps.

self.meteor_cmd = ['java', '-jar', '-Xmx2G', METEOR_JAR, '-', '-', '-stdio', '-l', 'en', '-norm', '-a', 'data/paraphrase-en.gz']

[helloMickey]

I encountered same bug，and try to delete parameter 'shell = True' in the function self.meteor_p = subprocess.Popen(.....), fixed this bugs！！

I use https://github.com/salaniz/pycocoevalcap, a python3 version coco-caption evaluation code.

[ltp1995]

I have meet the same error, solutions

1 First we should install the java environment with following codes: sudo apt-get install default-jre sudo apt-get install default-jdk 2 Then ['java', '-jar', '-Xmx2G', METEOR_JAR, '-', '-', '-stdio', '-l', 'en', '-norm', '-a', 'data/paraphrase-en.gz'] can be changed into ['/usr/bin/java', '-jar', '-Xmx2G', METEOR_JAR, '-', '-', '-stdio', '-l', 'en', '-norm', '-a', 'data/paraphrase-en.gz'] where '/usr/bin/java' is our installed Java library file in the first stage. 3 Last, delete parameter 'shell=True' Fixed the error!

[wanghao14]

@ltp1995 Thanks for sharing your solution! It helps me a lot!!!

[XYers]

' '.join(self.meteor_cmd),\     # change this line

changing the second line of init might be essential to this problem.

[Strke]

another way is to download paraphrase-en.gz this file https://github.com/salaniz/pycocoevalcap/blob/ad63453cfab57a81a02b2949b17a91fab1c3df77/meteor/data/paraphrase-en.gz and change the cmd like @dahuang37, Mainly adding these two parameters'-a', 'data/paraphrase-en.gz'