Language Models

posted Dec 5, 2018, 2:42 AM by MUHAMMAD MUN`IM AHMAD ZABIDI   [ updated Dec 5, 2018, 2:51 AM ]

Language modeling is key to many interesting problems such as speech recognition, machine translation, or image captioning.  The goal of the problem is to fit a probabilistic model which assigns probabilities to sentences. It does so by predicting next words in a text given a history of previous words.

Language modeling is key to many interestin

g problems such as speech recognition, machine translation, or image captioning.


posted Dec 3, 2018, 10:22 PM by MUHAMMAD MUN`IM AHMAD ZABIDI   [ updated Dec 3, 2018, 10:28 PM ]

clinfo is a simple command-line application that enumerates all possible (known) properties of the OpenCL platform and devices available on the system.
Let's install and see what  I have.

$ cat /proc/cpuinfo  | grep 'name'| uniq
model name : AMD FX(tm)-8350 Eight-Core Processor

$ sudo apt install clinfo

$ clinfo Number of platforms 1 Platform Name NVIDIA CUDA Platform Vendor NVIDIA Corporation Platform Version OpenCL 1.2 CUDA 9.1.84 Platform Profile FULL_PROFILE Platform Extensions cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_fp64 cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll cl_nv_copy_opts cl_nv_create_buffer Platform Extensions function suffix NV Platform Name NVIDIA CUDA Number of devices 1 Device Name GeForce GTX 1080 Ti Device Vendor NVIDIA Corporation Device Vendor ID 0x10de Device Version OpenCL 1.2 CUDA Driver Version 390.87 Device OpenCL C Version OpenCL C 1.2 Device Type GPU Device Topology (NV) PCI-E, 01:00.0 Device Profile FULL_PROFILE Device Available Yes Compiler Available Yes Linker Available Yes Max compute units 28 Max clock frequency 1582MHz Compute Capability (NV) 6.1 Device Partition (core) Max number of sub-devices 1 Supported partition types None Max work item dimensions 3 Max work item sizes 1024x1024x64 Max work group size 1024 Preferred work group size multiple 32 Warp size (NV) 32

10 Best Links: Word Embeddings

posted Nov 2, 2018, 8:32 AM by MUHAMMAD MUN`IM AHMAD ZABIDI   [ updated Dec 3, 2018, 8:55 PM ]

A popular idea in modern machine learning is to represent words by vectors. These vectors capture hidden information about a language, like word analogies or semantic.
  1. An introduction to word embeddings
  2. Introduction to Word Embeddings: Problems and Theory
  3. [Hamilton 2016] Hamilton, William L., et al. “Inducing domain-specific sentiment lexicons from unlabeled corpora.” arXiv preprint arXiv:1606.02820 (2016).
  4. [Kusner 2015] Kusner, Matt, et al. “From word embeddings to document distances.” International Conference on Machine Learning. 2015.
  5. [Mikolov 2013a] Mikolov, Tomas, Wen-tau Yih, and Geoffrey Zweig. “Linguistic regularities in continuous space word representations.” hlt-Naacl. Vol. 13. 2013.
  6. [Mikolov 2013b] Mikolov, Tomas, et al. “Efficient estimation of word representations in vector space.” arXiv preprint arXiv:1301.3781 (2013).
  7. [Mikolov 2013c] Mikolov, Tomas, et al. “Distributed representations of words and phrases and their compositionality.” Advances in neural information processing systems. 2013.
  8. [Mikolov 2013d] Mikolov, Tomas, Quoc V. Le, and Ilya Sutskever. “Exploiting similarities among languages for machine translation.” arXiv preprint arXiv:1309.4168 (2013).

10 Best Links: Face Recognition

posted Nov 2, 2018, 7:51 AM by MUHAMMAD MUN`IM AHMAD ZABIDI   [ updated Nov 2, 2018, 7:52 AM ]

10 Best Links: DL for Speech Recognition

posted Oct 30, 2018, 1:05 AM by MUHAMMAD MUN`IM AHMAD ZABIDI   [ updated Oct 30, 2018, 1:15 AM ]

The task of speech recognition is to map an acoustic signal containing a spoken natural language utterance into the corresponding sequence of words intended by the speaker.
— Page 458, Deep Learning, 2016.

Anaconda Navigator Launcher Icon

posted Oct 20, 2018, 7:21 PM by MUHAMMAD MUN`IM AHMAD ZABIDI   [ updated Oct 20, 2018, 7:23 PM ]

First get a suitable icon. It's already in your drive but it's located so deep in thes system it's easier just to Google for it. For example get it from here:

$ wget

There you go!

Next move it to the pixmaps directory

# mv anaconda-icon-1024x1024.png /usr/share/pixmaps/anaconda-navigator.png

Enter the following text in your favorite text editor. The command:

$ atom Anaconda.desktop

The text follows. The Exec option is the location of the executable. The Icon option is the filename in the pixmaps directory.

[Desktop Entry]

The move the file in to the Applications 'registry':

# mv Anaconda.desktop /usr/share/applications

Test it by hitting the Windows key on your Ubuntu keyboard:

Deep NLP

posted Oct 20, 2018, 5:08 PM by MUHAMMAD MUN`IM AHMAD ZABIDI   [ updated Oct 20, 2018, 5:09 PM ]

curl & wget

posted Oct 20, 2018, 3:20 AM by MUHAMMAD MUN`IM AHMAD ZABIDI   [ updated Oct 20, 2018, 5:11 PM ]

curl is a tool to transfer data from or to a server, using one of thesupported protocols (DICT, FILE, FTP, FTPS, GOPHER, HTTP, HTTPS, IMAP,IMAPS, LDAP, LDAPS, POP3, POP3S, RTMP, RTSP, SCP, SFTP, SMTP, SMTPS,TELNET and TFTP). The command is designed to work without user inter?action.

  • Basic usage is fairly simple - just pass the URL as input to the curl command, and redirect the output to a file.

curl > test.torrent

  • force curl to use the name of the file being downloaded as the local file name. This can be done using the -O command line option.

curl -O

  • If you want curl to follow the redirect, use the -L command line option instead.

curl -L

  • Resume a download from point of interruption?

curl -C - -O

wget is similar to curl. On Linux, wget is more common than curl. Whereas curl is the equivalent in MacOS.

  • To download the whole directory, wget is better
wget -r

1-10 of 92