How to convert a pb file into tflite file using python3 or in terminal.
Don't know any of the details of the model.Here is the link of the pb file.
(Edited)
I have done converting pb file to tflite file by the following code :
import tensorflow.compat.v1 as tf
import numpy as np
graph_def_file = "./models/20170512-110547.pb"
def representative_dataset_gen():
for _ in range(num_calibration_steps):
# Get sample input data as a numpy array in a method of your choosing.
yield [input]
converter = tf.lite.TFLiteConverter.from_frozen_graph(graph_def_file,
input_arrays=["input","phase_train"],
output_arrays=["embeddings"],
input_shapes={"input":[1,160,160,3],"phase_train":False})
converter.optimizations = [tf.lite.Optimize.DEFAULT]
converter.representative_dataset = representative_dataset_gen
converter.target_spec.supported_ops = [tf.lite.OpsSet.TFLITE_BUILTINS_INT8]
converter.inference_input_type = tf.uint8
converter.inference_output_type = tf.uint8
tflite_model = converter.convert()
print("converting")
open("./models/converted_model.tflite", "wb").write(tflite_model)
print("Done")
Error :Getting Segmentation fault(core dumped)
2020-01-20 11:42:18.153263: 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
2020-01-20 11:42:18.153363: 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
2020-01-20 11:42:18.153385: 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.
2020-01-20 11:42:18.905028: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcuda.so.1
2020-01-20 11:42:18.906845: E tensorflow/stream_executor/cuda/cuda_driver.cc:351] failed call to cuInit: CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected
2020-01-20 11:42:18.906874: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:156] kernel driver does not appear to be running on this host (kalgudi-GA-78LMT-USB3-6-0): /proc/driver/nvidia/version does not exist
2020-01-20 11:42:18.934144: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 3616020000 Hz
2020-01-20 11:42:18.934849: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x39aa0f0 initialized for platform Host (this does not guarantee that XLA will be used). Devices:
2020-01-20 11:42:18.934910: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version
Segmentation fault (core dumped)
with out any details of the model, you cannot convert it to a .tflite model. I suggest that you go through this document for Post Training Quantization again. As there are too many details that is redundant to show here.
Here is an example for post training quantization of a frozen graph. The model is taken from here (you can see that it's a full tarball of infomation about the model)
import sys, os, glob
import tensorflow as tf
import pathlib
import numpy as np
if len(sys.argv) != 2:
print('Usage: <' + sys.argv[0] + '> <frozen_graph_file> <representative_image_dir>')
exit()
tf.compat.v1.enable_eager_execution()
tf.compat.v1.logging.set_verbosity(tf.compat.v1.logging.DEBUG)
def fake_representative_data_gen():
for _ in range(100):
fake_image = np.random.random((1,192,192,3)).astype(np.float32)
yield [fake_image]
frozen_graph = sys.argv[1]
input_array = ['input']
output_array = ['MobilenetV1/Predictions/Reshape_1']
converter = tf.compat.v1.lite.TFLiteConverter.from_frozen_graph(frozen_graph, input_array, output_array)
converter.optimizations = [tf.lite.Optimize.DEFAULT]
converter.representative_dataset = fake_representative_data_gen
converter.target_spec.supported_ops = [tf.lite.OpsSet.TFLITE_BUILTINS_INT8]
converter.inference_input_type = tf.uint8
converter.inference_output_type = tf.uint8
tflite_model = converter.convert()
quant_dir = pathlib.Path(os.getcwd(), 'output')
quant_dir.mkdir(exist_ok=True, parents=True)
tflite_model_file = quant_dir/'mobilenet_v1_0.25_192_quant.tflite'
tflite_model_file.write_bytes(tflite_model)
Related
i have an S3 was over 130k Json Files which i need to calculate numbers based on data in the json files (for example calculate the number of gender of Speakers). i am currently using s3 Paginator and JSON.load to read each file and extract information form. but it take a very long time to process such a large number of file (2-3 files per second). how can i speed up the process? please provide working code examples if possible. Thank you
here is some of my code:
client = boto3.client('s3')
paginator = client.get_paginator('list_objects_v2')
result = paginator.paginate(Bucket='bucket-name',StartAfter='')
for page in result:
if "Contents" in page:
for key in page[ "Contents" ]:
keyString = key[ "Key" ]
s3 = boto3.resource('s3')
content_object = s3.Bucket('bucket-name').Object(str(keyString))
file_content = content_object.get()['Body'].read().decode('utf-8')
json_content = json.loads(file_content)
x = (json_content['dict-name'])
In order to use the code below, I'm assuming you understand pandas (if not, you may want to get to know it). Also, it's not clear if your 2-3 seconds is on the read or includes part of the number crunching, nonetheless multiprocessing will speed this up dramatically. The gist is to read all the files in (as dataframes), concatenate them, then do your analysis.
To be useful for me, I run this on spot instances that have lots of vCPUs and memory. I've found the instances that are network optimized (like c5n - look for the n) and the inf1 (for machine learning) are much faster at reading/writing than T or M instance types, as examples.
My use case is reading 2000 'directories' with roughly 1200 files in each and analyzing them. The multithreading is orders of magnitude faster than single threading.
File 1: your main script
# create script.py file
import os
from multiprocessing import Pool
from itertools import repeat
import pandas as pd
import json
from utils_file_handling import *
ufh = file_utilities() #instantiate the class functions - see below (second file)
bucket = 'your-bucket'
prefix = 'your-prefix/here/' # if you don't have a prefix pass '' (empty string or function will fail)
#define multiprocessing function - get to know this to use multiple processors to read files simultaneously
def get_dflist_multiprocess(keys_list, num_proc=4):
with Pool(num_proc) as pool:
df_list = pool.starmap(ufh.reader_json, zip(repeat(bucket), keys_list), 15)
pool.close()
pool.join()
return df_list
#create your master keys list upfront; you can loop through all or slice the list to test
keys_list = ufh.get_keys_from_prefix(bucket, prefix)
# keys_list = keys_list[0:2000] # as an exampmle
num_proc = os.cpu_count() #tells you how many processors your machine has; function above defaults to 4 unelss given
df_list = get_dflist_multiprocess(keys_list, num_proc=num_proc) #collect dataframes for each file
df_new = pd.concat(df_list, sort=False)
df_new = df_new.reset_index(drop=True)
# do your analysis on the dataframe
File 2: class functions
#utils_file_handling.py
# create this in a separate file; name as you wish but change the import in the script.py file
import boto3
import json
import pandas as pd
#define client and resource
s3sr = boto3.resource('s3')
s3sc = boto3.client('s3')
class file_utilities:
"""file handling function"""
def get_keys_from_prefix(self, bucket, prefix):
'''gets list of keys and dates for given bucket and prefix'''
keys_list = []
paginator = s3sr.meta.client.get_paginator('list_objects_v2')
# use Delimiter to limit search to that level of hierarchy
for page in paginator.paginate(Bucket=bucket, Prefix=prefix, Delimiter='/'):
keys = [content['Key'] for content in page.get('Contents')]
print('keys in page: ', len(keys))
keys_list.extend(keys)
return keys_list
def read_json_file_from_s3(self, bucket, key):
"""read json file"""
bucket_obj = boto3.resource('s3').Bucket(bucket)
obj = boto3.client('s3').get_object(Bucket=bucket, Key=key)
data = obj['Body'].read().decode('utf-8')
return data
# you may need to tweak this for your ['dict-name'] example; I think I have it correct
def reader_json(self, bucket, key):
'''returns dataframe'''
return pd.DataFrame(json.loads(self.read_json_file_from_s3(bucket, key))['dict-name'])
I have trained the Detectron2 model on the google colab server free server.
from detectron2.engine import DefaultTrainer
from detectron2.config import get_cfg
import os
cfg = get_cfg()
cfg.merge_from_file(model_zoo.get_config_file("COCO-
InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml"))
cfg.DATASETS.TRAIN = ("Dataset_train",)
cfg.DATASETS.TEST = ()
cfg.DATALOADER.NUM_WORKERS = 2
cfg.MODEL.WEIGHTS = model_zoo.get_checkpoint_url("COCO-
InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml")
cfg.SOLVER.IMS_PER_BATCH = 2
cfg.SOLVER.BASE_LR = 0.00025
cfg.SOLVER.MAX_ITER = 2800
cfg.MODEL.ROI_HEADS.NUM_CLASSES = 2
os.makedirs(cfg.OUTPUT_DIR, exist_ok=True)
trainer = DefaultTrainer(cfg)
trainer.resume_or_load(resume=False)
trainer.train()
This above code creates an "output" folder in which I have 4 files:
model_final.pth, metrics.json, last_checkpoint and events.out.file
I can use this model for prediction using
cfg.MODEL.WEIGHTS = os.path.join(cfg.OUTPUT_DIR, "model_final.pth")
cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.5
cfg.DATASETS.TEST = ("Datset_test")
predictor = DefaultPredictor(cfg)
Everything works fine when colab local session is not expired.
Problem:
When I mount this output folder somewhere else than the model not imported.
How to use .pth saved model for predictions?
How to reuse detectron2 trained model for prediction..?
Download your .pth file. When you open a new Colab runtime, upload that file to your local runtime (you can use "!cp" from Google Drive to your runtime to transfer .pth file quickly) and then follow the steps on official tutorial : https://colab.research.google.com/drive/16jcaJoc6bCFAQ96jDe2HwtXj7BMD_-m5
You need to use your config file cfg.merge_from_file("YOUR CONFIG FILE") and your .pth file path: cfg.MODEL.WEIGHTS = ("YOUR .PTH FILE, PROBABLY /CONTENT/MODEL_FINAL.PTH")
I am new to google colab. I am implementing a pretrained vgg16 and resnet50 model using pytorch, but I am unable to load my file and read it as it returns an error of no directory found
I have uploaded the data through file also I have used to upload it using
from google.colab import files
uploaded = files.upload()
The file got uploaded but when I tried to unzip it because it is a zip file using
!unzip content/cropped_months
then it says
no file found
import torch
import torch.nn as nn
import torch.optim as optim
from torchvision.transforms import *
from torch.optim import lr_scheduler
from torch.autograd import Variable
import numpy as np
import torchvision
from torchvision import datasets, models, transforms
import matplotlib.pyplot as plt
import time
import os
import copy
from google.colab import files
uploaded = files.upload()
!unzip content/cropped_months
data_dir = 'content/cropped_months'
#Define transforms for the training data and testing data
train_transforms = transforms.Compose([transforms.RandomRotation(30),transforms.RandomResizedCrop(224),transforms.RandomHorizontalFlip(),transforms.ToTensor(),transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])])
test_transforms = transforms.Compose([transforms.Resize(256),transforms.CenterCrop(224),transforms.ToTensor(),transforms.Normalize([0.485, 0.456, 0.406],[0.229, 0.224, 0.225])])
#pass transform here-in
train_data = datasets.ImageFolder(data_dir + '/train', transform=train_transforms)
test_data = datasets.ImageFolder(data_dir + '/test', transform=test_transforms)
#data loaders
trainloader = torch.utils.data.DataLoader(train_data, batch_size=8, shuffle=True)
testloader = torch.utils.data.DataLoader(test_data, batch_size=8, shuffle=True)
print("Classes: ")
class_names = train_data.classes
print(class_names)
first error
unzip: cannot find or open content/cropped_months,
content/cropped_months.zip or content/cropped_months.ZIP.
second error
--------------------------------------------------------------------------- FileNotFoundError Traceback (most recent call
last) in ()
16
17 #pass transform here-in
---> 18 train_data = datasets.ImageFolder(data_dir + '/train', transform=train_transforms)
19 test_data = datasets.ImageFolder(data_dir + '/test', transform=test_transforms)
20
2 frames
/usr/local/lib/python3.6/dist-packages/torchvision/datasets/folder.py
in _find_classes(self, dir)
114 if sys.version_info >= (3, 5):
115 # Faster and available in Python 3.5 and above
--> 116 classes = [d.name for d in os.scandir(dir) if d.is_dir()]
117 else:
118 classes = [d for d in os.listdir(dir) if os.path.isdir(os.path.join(dir, d))]
FileNotFoundError: [Errno 2] No such file or directory:
'content/cropped_months (1)/train'
You are probably trying to access the wrong path. In my notebook, the file was uploaded to the working directory.
Use google.colab.files to upload the zip.
from google.colab import files
files.upload()
Upload your file. Google Colab will display where it was saved:
Saving dummy.zip to dummy.zip
Then just run !unzip:
!unzip dummy.zip
I think you can use PySurvival library is compatible with Torch , here the link :
https://square.github.io/pysurvival/miscellaneous/save_load.html
I have used the following command for feature extraction:
./build/tools/extract_features.bin models/bvlc_reference_caffenet/bvlc_reference_caffenet.caffemodel examples/_temp/imagenet_val.prototxt fc7 examples/_temp/features 10 leveldb GPU 0
the I used the following code to read leveldb features:
https://15519285443829437810.googlegroups.com/attach/b71d5c99c447fc2a/main.cpp?part=0.1&view=1&vt=ANaJVrHM26ydsY5Z2ognvhBaFtDzXnE_SiVf19DLkrNpf9Q34t5O4bJWy0nHH_HBnBAVx5wZusXd1joX93JBK0_r7XKEIc-5odz9_HPHV1RUo8MD3zNHgoY
everything is OK and I have one .ldb file now.
How can I read this .ldb file? Can I change it to .text?
I have used the following code :
import caffe
import leveldb
import numpy as np
from caffe.proto import caffe_pb2
db = leveldb.LevelDB('/home/deep/rahim/caffe-master/examples/_temp/features')
datum = caffe_pb2.Datum()
for key, value in db.RangeIter():
datum.ParseFromString(value)
label = datum.label
data = caffe.io.datum_to_array(datum)
image = np.transpose(data, (1,2,0))
np.save('feature.txt',image)
Then pass the feature.txt.npy to the following code which convert .npy to .txt:
import struct
import numpy as np
import os
def parseNPY(path, fileJustName):
# load from the file
inputFile = os.path.join(path, fileJustName + ".npy")
matrices = np.load(inputFile)
outputfile = os.path.join(path, fileJustName)
for m in range(matrices.shape[0]):
# file name for this matrix
outFileFull = outputfile + "-" + str(m) + ".txt"
# output matrix to a numbered file
np.savetxt(outFileFull, matrices[m], fmt="%i", delimiter="\t")
mypath = "/home/deep/rahim/caffe-master/python/"
for path, paths, filenames in os.walk(mypath):
# translate all filenames.
for filename in filenames:
fileJustName, fileExtension = os.path.splitext(filename)
if fileExtension == ".npy":
print(os.path.join(path, fileJustName))
parseNPY(path, fileJustName)
I am trying to run the code below for a level set 1D problem (example in fipy webpage). I am getting this error:
Traceback (most recent call last):
File "C:/Users/sgowda/Documents/pde solver code/level set 1D.py", line 20, in
var.calcDistanceFunction()
File "C:\Users\sgowda\AppData\Local\Continuum\Anaconda\lib\site-packages\fipy\variables\distanceVariable.py", line 335, in calcDistanceFunction
raise Exception, "Neither lsmlib nor skfmm can be found on the $PATH"
Exception: Neither lsmlib nor skfmm can be found on the $PATH
Could you please let me know how to fix this. i tried looking into the distancefucntion() but im not sure what this error means?
from fipy import Grid1D, CellVariable, TransientTerm, DiffusionTerm, Viewer, DistanceVariable
import matplotlib.pyplot as plt
velocity = 1.
dx = 1.
nx = 10
timeStepDuration = 1.
steps = 2
L = nx * dx
interfacePosition = L / 5.
from fipy.tools import serialComm
mesh = Grid1D(dx=dx, nx=nx, communicator=serialComm)
var = DistanceVariable(name='level set variable',
mesh=mesh,
value=-1.,
hasOld=1)
var.setValue(1., where=mesh.cellCenters[0] > interfacePosition)
var.calcDistanceFunction()
advEqn = TransientTerm() + FirstOrderAdvectionTerm(velocity)
viewer = Viewer(vars=var, datamin=-10., datamax=10.)
viewer.plot()
for step in range(steps):
var.updateOld()
advEqn.solve(var, dt=timeStepDuration)
viewer.plot()
plt.show()
FiPy doesn't have a native level set implementation so uses either LSMLIB or Scikit-fmm to provide the level set / fast marching method functionality.
To see whether you have them installed correctly, use either
$ python -c “import pylsmlib; pylsmlib.test()”
or
$ python -c “import skfmm; skfmm.test()”
to test.
The requirement is outlined in the FiPy documentation, see http://www.ctcms.nist.gov/fipy/INSTALLATION.html#level-set-packages
It is probably easier to install Scikit-fmm initially, see https://pythonhosted.org/scikit-fmm/, but
$ pip install scikit-fmm
should work.