def evaluate(args):
if args.model_prefix:
model_paths = sorted(
glob.glob(
os.path.join(args.output_dir, "Models",
f"{args.model_prefix}*.pt")),
key=lambda p: int(re.search(r"_step_(\d+)", p).group(1)))
model_scores = {}
for model_path in model_paths:
translate(model_path, "valid", args)
processData(args, False)
scores = calculate_scores(args, "valid")
model_scores[os.path.basename(model_path)] = scores
write_scores(
model_scores,
os.path.join(
args.output_dir, "Reports",
f"{args.model_prefix}.valid.{args.src_lang}2{args.tgt_lang}.log"
))
if args.eval_model:
model_scores = {}
translate(args.eval_model, "test", args)
processData(args, False)
scores = calculate_scores(args, "test")
model_scores[os.path.basename(args.eval_model)] = scores
write_scores(
model_scores,
os.path.join(
args.output_dir, "Reports",
f"{os.path.basename(args.eval_model)}.test.{args.src_lang}2{args.tgt_lang}.log"
))
def main(args):
processData(args, True)
if args.do_preprocess:
preprocess(args)
if args.do_train:
train(args)
if args.model_prefix and args.average_last:
average_models(args)
if args.do_eval:
evaluate(args)
def selectNegativeWindowsFromPositiveImages(groundTruths,featuresDir,featuresExt,maxVectors,overlap,model=False):
gtb = dict()
for x in groundTruths:
im,bx = x[0],map(float,x[1:])
try:
gtb[im].append(bx)
except:
gtb[im] = [bx]
task = NWFPIFilter(gtb,featuresDir,featuresExt,maxVectors/len(gtb.keys()),overlap,model)
result = dp.processData(gtb.keys(),featuresDir,featuresExt,task)
posIdx,posFeat,negIdx,negFeat = [],[],[],[]
for r in result:
posIdx += r[0]
posFeat += r[1]
negIdx += r[2]
negFeat += r[3]
Xp = emptyMatrix( (len(posIdx),posFeat[0].shape[1]) )
Xn = emptyMatrix( (len(negIdx),negFeat[0].shape[1]) )
k = 0
for i in range(len(posFeat)):
Xp[k:k+posFeat[i].shape[0],:] = posFeat[i]
k = k + posFeat[i].shape[0]
k = 0
for i in range(len(negFeat)):
Xn[k:k+negFeat[i].shape[0],:] = negFeat[i]
k + k + negFeat[i].shape[0]
print 'NegFromPos ready:',len(negIdx)
return {'posIdx':posIdx, 'posFeat':Xp, 'negIdx':negIdx, 'negFeat':Xn}
def detectObjects(imageList, featuresDir, indexType, groundTruthDir,
outputDir):
maxOverlap = 0.3
categories, catIndex = bse.categoryIndex(indexType)
task = SoftmaxDetector(maxOverlap, catIndex)
result = processData(imageList, featuresDir, 'prob', task)
# Collect detection results after NMS
detections = dict([(c, []) for c in catIndex])
for res in result:
for idx in catIndex:
img, filteredBoxes, filteredScores = res[idx]
for j in range(len(filteredBoxes)):
detections[idx].append([img, filteredScores[j]] +
filteredBoxes[j])
# Evaluate results for each category independently
for idx in catIndex:
groundTruthFile = groundTruthDir + '/' + categories[
idx] + '_test_bboxes.txt'
output = outputDir + '/' + categories[idx] + '.out'
detections[idx].sort(key=lambda x: x[1], reverse=True)
gtBoxes = [x.split() for x in open(groundTruthFile)]
numPositives = len(gtBoxes)
groundTruth = eval.loadGroundTruthAnnotations(gtBoxes)
results = eval.evaluateDetections(groundTruth, detections[idx], 0.5)
prec, recall = eval.computePrecisionRecall(numPositives, results['tp'],
results['fp'], output)
def detectObjects(model,
imageList,
featuresDir,
featuresExt,
maxOverlap,
threshold,
outputFile=None):
task = Detector(model, threshold, maxOverlap)
result = processData(imageList, featuresDir, featuresExt, task)
if outputFile != None:
outf = open(outputFile, 'w')
writeF = lambda x, y, b: outf.write(
x + ' {:.8f} {:.0f} {:.0f} {:.0f} {:.0f} {:.0f}\n'.format(
y, b[0], b[1], b[2], b[3], b[4]))
else:
writeF = lambda x, y, b: x
detectionsList = []
for data in result:
img, filteredBoxes, filteredScores = data
for i in range(len(filteredBoxes)):
b = filteredBoxes[i]
writeF(img, filteredScores[i], b)
detectionsList.append(
[img, filteredScores[i], b[0], b[1], b[2], b[3], b[4]])
if outputFile != None:
outf.close()
return detectionsList
def detectObjects(model,
imageList,
featuresDir,
featuresExt,
threshold,
projector=None):
task = MaskDetector(model, threshold, projector)
result = processData(imageList, featuresDir, featuresExt, task)
if projector == None:
totalDetections = reduce(lambda x, y: x + y,
[d[1].shape[0] for d in result])
detections = np.zeros((totalDetections, 5))
images = {}
imgId = 0
i = 0
for data in result:
img, cells = data
cells[:, 0] = cells[:, 0] * imgId
detections[i:i + cells.shape[0], :] = cells
images[img] = imgId
imgId += 1
i = i + cells.shape[0]
return (images, detections[0:i, :])
else:
resultsList = []
for data in result:
resultsList += data
return resultsList
def loadHardNegativesFromMatrix(featuresDir,imagesIdx,detMatrix,featuresExt,numFeatures,totalNegatives):
i = 0
task = LoadHardNegatives(imagesIdx,detMatrix,numFeatures)
result = dp.processData(imagesIdx.keys(),featuresDir,featuresExt,task)
hardng = cu.emptyMatrix([totalNegatives,numFeatures])
while len(result) > 0:
data = result.pop(0)
hardng[i:i+data.shape[0],:] = data
i = i + data.shape[0]
return hardng[0:i,:]
def loadHardNegativesFromMatrix(featuresDir, imagesIdx, detMatrix, featuresExt,
numFeatures, totalNegatives):
i = 0
task = LoadHardNegatives(imagesIdx, detMatrix, numFeatures)
result = dp.processData(imagesIdx.keys(), featuresDir, featuresExt, task)
hardng = cu.emptyMatrix([totalNegatives, numFeatures])
while len(result) > 0:
data = result.pop(0)
hardng[i:i + data.shape[0], :] = data
i = i + data.shape[0]
return hardng[0:i, :]
def loadHardNegativesFromList(featuresDir,negativesInfo,featuresExt,numFeatures,totalNegatives,idx=False):
i = 0
task = LoadHardNegatives(negativesInfo)
result = dp.processData(negativesInfo.keys(),featuresDir,featuresExt,task)
hardng = emptyMatrix([totalNegatives,numFeatures])
hardNames = []
boxes = []
while len(result) > 0:
data,imgs,box = result.pop(0)
hardng[i:i+data.shape[0],:] = data
hardNames += imgs
boxes += box
i = i + data.shape[0]
return (hardng[0:i,:],boxes)
def getRandomNegs(featuresDir,negativeList,featuresExt,numFeatures,maxVectors,maxNegativeImages):
randomBoxes = maxVectors/maxNegativeImages
cu.rnd.shuffle(negativeList)
task = RandomNegativesFilter(numFeatures,randomBoxes)
negatives = [negativeList.pop(0) for i in range(maxNegativeImages)]
result = dp.processData(negatives,featuresDir,featuresExt,task)
neg = emptyMatrix([maxVectors,numFeatures])
boxes = []
n = 0
while len(result) > 0:
mat,box = result.pop()
neg[n:n+mat.shape[0]] = mat
n = n + mat.shape[0]
boxes += box
return (neg[0:n],boxes[0:n])
def getHardNegatives(negativesDir,negativesList,featuresExt,numFeatures,maxVectors,currentModel):
maxVectorsPerImage = maxVectors/len(negativesList)
i = 0
task = HardNegativeMining(currentModel,maxVectorsPerImage)
result = dp.processData(negativesList,negativesDir,featuresExt,task)
hardng = emptyMatrix([2*maxVectors,numFeatures])
boxes = []
while len(result) > 0:
data = result.pop(0)
if data[0].shape[0]+i > hardng.shape[0]:
print 'Not enough matrix space'
hardng = np.concatenate( (hardng,emptyMatrix([maxVectors,numFeatures])) )
hardng[i:i+data[0].shape[0],:] = data[0]
boxes += data[2]
i = i + data[0].shape[0]
return hardng[0:i,:],boxes[0:i]
def detectObjects(model,imageList,featuresDir,featuresExt,maxOverlap,threshold,outputFile=None):
task = Detector(model,threshold,maxOverlap)
result = processData(imageList,featuresDir,featuresExt,task)
if outputFile != None:
outf = open(outputFile,'w')
writeF = lambda x,y,b: outf.write(x + ' {:.8f} {:.0f} {:.0f} {:.0f} {:.0f} {:.0f}\n'.format(y,b[0],b[1],b[2],b[3],b[4]))
else:
writeF = lambda x,y,b: x
detectionsList = []
for data in result:
img,filteredBoxes,filteredScores = data
for i in range(len(filteredBoxes)):
b = filteredBoxes[i]
writeF(img,filteredScores[i],b)
detectionsList.append( [img,filteredScores[i],b[0],b[1],b[2],b[3],b[4]] )
if outputFile != None:
outf.close()
return detectionsList
def selectRegions(imageList, featuresDir, groundTruths, outputDir, featExt, category, operator):
task = RegionSelector(groundTruths, operator)
result = processData(imageList, featuresDir, featExt, task)
nBoxes,nFeat = 0,0
for r in result:
nBoxes += r[0].shape[0]
nFeat = r[0].shape[1]
featureMatrix = np.zeros( (nBoxes,nFeat) )
i = 0
outputFile = open(outputDir + '/' + category + '.idx','w')
for r in result:
featureMatrix[i:i+r[0].shape[0]] = r[0]
for box in r[1]:
outputFile.write(box[0] + ' ' + ' '.join(map(str,map(int,box[1:]))) + '\n')
i += r[0].shape[0]
outputFile.close()
cu.saveMatrix(featureMatrix,outputDir + '/' + category + '.' + featExt)
print 'Total of',nBoxes,'positive examples collected for',category
def selectRegions(imageList, featuresDir, groundTruths, outputDir, featExt,
category, operator):
task = RegionSelector(groundTruths, operator)
result = processData(imageList, featuresDir, featExt, task)
nBoxes, nFeat = 0, 0
for r in result:
nBoxes += r[0].shape[0]
nFeat = r[0].shape[1]
featureMatrix = np.zeros((nBoxes, nFeat))
i = 0
outputFile = open(outputDir + '/' + category + '.idx', 'w')
for r in result:
featureMatrix[i:i + r[0].shape[0]] = r[0]
for box in r[1]:
outputFile.write(box[0] + ' ' +
' '.join(map(str, map(int, box[1:]))) + '\n')
i += r[0].shape[0]
outputFile.close()
cu.saveMatrix(featureMatrix, outputDir + '/' + category + '.' + featExt)
print 'Total of', nBoxes, 'positive examples collected for', category
def detectObjects(imageList, featuresDir, indexType, groundTruthDir, outputDir):
maxOverlap = 0.3
categories, catIndex = bse.categoryIndex(indexType)
task = SoftmaxDetector(maxOverlap, catIndex)
result = processData(imageList, featuresDir, 'prob', task)
# Collect detection results after NMS
detections = dict([ (c,[]) for c in catIndex])
for res in result:
for idx in catIndex:
img, filteredBoxes, filteredScores = res[idx]
for j in range(len(filteredBoxes)):
detections[idx].append( [img, filteredScores[j]] + filteredBoxes[j] )
# Evaluate results for each category independently
for idx in catIndex:
groundTruthFile = groundTruthDir + '/' + categories[idx] + '_test_bboxes.txt'
output = outputDir + '/' + categories[idx] + '.out'
detections[idx].sort(key=lambda x:x[1], reverse=True)
gtBoxes = [x.split() for x in open(groundTruthFile)]
numPositives = len(gtBoxes)
groundTruth = eval.loadGroundTruthAnnotations(gtBoxes)
results = eval.evaluateDetections(groundTruth, detections[idx], 0.5)
prec, recall = eval.computePrecisionRecall(numPositives, results['tp'], results['fp'], output)
def detectObjects(model,imageList,featuresDir,featuresExt,threshold,projector=None):
task = MaskDetector(model,threshold,projector)
result = processData(imageList,featuresDir,featuresExt,task)
if projector == None:
totalDetections = reduce(lambda x,y:x+y,[d[1].shape[0] for d in result])
detections = np.zeros( (totalDetections,5) )
images = {}
imgId = 0
i = 0
for data in result:
img,cells = data
cells[:,0] = cells[:,0]*imgId
detections[i:i+cells.shape[0],:] = cells
images[img] = imgId
imgId += 1
i = i + cells.shape[0]
return (images,detections[0:i,:])
else:
resultsList = []
for data in result:
resultsList += data
return resultsList
import utils as cu
import libDetection as det
from dataProcessor import processData
class Checker():
def __init__(self):
print 'Starting checker'
def run(self,img,features,bboxes):
return img,features.shape[0] == len(bboxes)
## Main Program Parameters
params = cu.loadParams("testImageList featuresDir featuresExt")
imageList = [x.replace('\n','') for x in open(params['testImageList'])]
## Run Detector
task = Checker()
start = cu.tic()
result = processData(imageList,params['featuresDir'],params['featuresExt'],task)
cu.toc('All images checked',start)
totalP = 0
for data in result:
img,r = data
if not r:
print 'Problems with',img
totalP += 1
print 'Total problems:',totalP
def extractFeatures(model,imageList,featuresDir,featuresExt): task = CategoryScores(model,featuresDir) result = processData(imageList,featuresDir,featuresExt,task)
def extractFeatures(model, imageList, featuresDir, featuresExt):
task = CategoryScores(model, featuresDir)
result = processData(imageList, featuresDir, featuresExt, task)
modelClass = svm.SVMDetector
else:
import sys
print 'Model not supported'
sys.exit()
for c in categories.keys():
filename = params['modelDir'] + '/' + c + params['modelSuffix']
categories[c] = modelClass()
categories[c].load(filename)
imageList = [x.replace('\n', '') for x in open(params['testImageList'])]
maxOverlap = float(params['maxOverlap'])
threshold = float(params['threshold'])
## Run Detector
task = Detector(categories, threshold, maxOverlap)
result = processData(imageList, params['featuresDir'], params['featuresExt'],
task)
# Prepare output files
for c in categories.keys():
categories[c] = open(c + params['outputFile'], 'w')
for data in result:
for c in data.keys():
img, filteredBoxes, filteredScores = data[c]
for i in range(len(filteredBoxes)):
b = filteredBoxes[i]
categories[c].write(img + ' {:.8f} {:} {:} {:} {:}\n'.format(
filteredScores[i], b[0], b[1], b[2], b[3]))
for c in categories.keys():
categories[c].close()