{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"2021-06-19-recsys20-tutorial-feature-engineering-part-3.ipynb","provenance":[{"file_id":"1bnZOC6wVT95f8hS9bWb6x03D_FnxaYWL","timestamp":1624098368206}],"collapsed_sections":[],"mount_file_id":"1Jf1ZWLsC8YZ1tdYKWEoJ2LV8DqEh0tY5","authorship_tag":"ABX9TyOQqrmtHSIk86ZF/weoFT8k"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"},"accelerator":"GPU"},"cells":[{"cell_type":"markdown","metadata":{"id":"Qp9CyeCjAP6I"},"source":["# Recsys'20 Feature Engineering Tutorial Part 3\n","> RecSys'20 tutorial on feature engineering on a large retail dataset part 3\n","\n","- toc: true\n","- badges: true\n","- comments: true\n","- categories: [features, recsys, cudf, retail, bigdata, nvtabular, xgboost]\n","- image: "]},{"cell_type":"markdown","metadata":{"id":"qU7bKl-xHQi3"},"source":["### Install RAPIDS"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"A477KsIzC8yR","executionInfo":{"status":"ok","timestamp":1624098504001,"user_tz":-330,"elapsed":730,"user":{"displayName":"Sparsh Agarwal","photoUrl":"","userId":"13037694610922482904"}},"outputId":"9f84a159-bdc7-43a9-8033-6a890ed6a9b7"},"source":["# Check Python Version\n","!python --version\n","\n","# Check Ubuntu Version\n","!lsb_release -a\n","\n","# Check CUDA/cuDNN Version\n","!nvcc -V && which nvcc\n","\n","# Check GPU\n","!nvidia-smi"],"execution_count":1,"outputs":[{"output_type":"stream","text":["Python 3.7.10\n","No LSB modules are available.\n","Distributor ID:\tUbuntu\n","Description:\tUbuntu 18.04.5 LTS\n","Release:\t18.04\n","Codename:\tbionic\n","nvcc: NVIDIA (R) Cuda compiler driver\n","Copyright (c) 2005-2020 NVIDIA Corporation\n","Built on Wed_Jul_22_19:09:09_PDT_2020\n","Cuda compilation tools, release 11.0, V11.0.221\n","Build cuda_11.0_bu.TC445_37.28845127_0\n","/usr/local/cuda/bin/nvcc\n","Sat Jun 19 10:28:21 2021 \n","+-----------------------------------------------------------------------------+\n","| NVIDIA-SMI 465.27 Driver Version: 460.32.03 CUDA Version: 11.2 |\n","|-------------------------------+----------------------+----------------------+\n","| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |\n","| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |\n","| | | MIG M. |\n","|===============================+======================+======================|\n","| 0 Tesla T4 Off | 00000000:00:04.0 Off | 0 |\n","| N/A 45C P8 10W / 70W | 0MiB / 15109MiB | 0% Default |\n","| | | N/A |\n","+-------------------------------+----------------------+----------------------+\n"," \n","+-----------------------------------------------------------------------------+\n","| Processes: |\n","| GPU GI CI PID Type Process name GPU Memory |\n","| ID ID Usage |\n","|=============================================================================|\n","| No running processes found |\n","+-----------------------------------------------------------------------------+\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"mMqV93ivDa4q","executionInfo":{"status":"ok","timestamp":1624098528171,"user_tz":-330,"elapsed":1296,"user":{"displayName":"Sparsh Agarwal","photoUrl":"","userId":"13037694610922482904"}},"outputId":"1a9b0002-e588-4748-d0e6-c19b14dadb4d"},"source":["# This get the RAPIDS-Colab install files and test check your GPU. Run this and the next cell only.\n","# Please read the output of this cell. If your Colab Instance is not RAPIDS compatible, it will warn you and give you remediation steps.\n","!git clone https://github.com/rapidsai/rapidsai-csp-utils.git\n","!python rapidsai-csp-utils/colab/env-check.py"],"execution_count":2,"outputs":[{"output_type":"stream","text":["Cloning into 'rapidsai-csp-utils'...\n","remote: Enumerating objects: 272, done.\u001b[K\n","remote: Counting objects: 100% (101/101), done.\u001b[K\n","remote: Compressing objects: 100% (82/82), done.\u001b[K\n","remote: Total 272 (delta 56), reused 38 (delta 19), pack-reused 171\u001b[K\n","Receiving objects: 100% (272/272), 79.66 KiB | 1.33 MiB/s, done.\n","Resolving deltas: 100% (118/118), done.\n","***********************************************************************\n","Woo! Your instance has the right kind of GPU, a Tesla T4!\n","***********************************************************************\n","\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"wOzXb8sdE69d","outputId":"5d039904-3555-405a-f55c-3f70d44ef968"},"source":["# This will update the Colab environment and restart the kernel. Don't run the next cell until you see the session crash.\n","!bash rapidsai-csp-utils/colab/update_gcc.sh\n","import os\n","os._exit(00)"],"execution_count":null,"outputs":[{"output_type":"stream","text":["Updating your Colab environment. This will restart your kernel. Don't Panic!\n","Ign:1 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64 InRelease\n","Hit:2 https://cloud.r-project.org/bin/linux/ubuntu bionic-cran40/ InRelease\n","Ign:3 https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64 InRelease\n","Get:4 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64 Release [697 B]\n","Hit:5 https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64 Release\n","Get:6 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64 Release.gpg [836 B]\n","Hit:8 http://ppa.launchpad.net/c2d4u.team/c2d4u4.0+/ubuntu bionic InRelease\n","Get:9 http://security.ubuntu.com/ubuntu bionic-security InRelease [88.7 kB]\n","Ign:10 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64 Packages\n","Get:10 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64 Packages [599 kB]\n","Hit:11 http://archive.ubuntu.com/ubuntu bionic InRelease\n","Get:12 http://archive.ubuntu.com/ubuntu bionic-updates InRelease [88.7 kB]\n","Hit:13 http://ppa.launchpad.net/cran/libgit2/ubuntu bionic InRelease\n","Get:14 http://security.ubuntu.com/ubuntu bionic-security/universe amd64 Packages [1,415 kB]\n","Hit:15 http://ppa.launchpad.net/deadsnakes/ppa/ubuntu bionic InRelease\n","Get:16 http://archive.ubuntu.com/ubuntu bionic-backports InRelease [74.6 kB]\n","Hit:17 http://ppa.launchpad.net/graphics-drivers/ppa/ubuntu bionic InRelease\n","Get:18 http://security.ubuntu.com/ubuntu bionic-security/main amd64 Packages [2,185 kB]\n","Get:19 http://archive.ubuntu.com/ubuntu bionic-updates/main amd64 Packages [2,619 kB]\n","Get:20 http://ppa.launchpad.net/ubuntu-toolchain-r/test/ubuntu bionic InRelease [20.8 kB]\n","Get:21 http://ppa.launchpad.net/ubuntu-toolchain-r/test/ubuntu bionic/main amd64 Packages [50.4 kB]\n","Get:22 http://archive.ubuntu.com/ubuntu bionic-updates/universe amd64 Packages [2,185 kB]\n","Fetched 9,326 kB in 2s (4,126 kB/s)\n","Reading package lists... Done\n","Added repo\n","Ign:1 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64 InRelease\n","Hit:2 https://cloud.r-project.org/bin/linux/ubuntu bionic-cran40/ InRelease\n","Ign:3 https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64 InRelease\n","Hit:4 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64 Release\n","Hit:5 https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64 Release\n","Hit:6 http://security.ubuntu.com/ubuntu bionic-security InRelease\n","Hit:8 http://ppa.launchpad.net/c2d4u.team/c2d4u4.0+/ubuntu bionic InRelease\n","Hit:10 http://archive.ubuntu.com/ubuntu bionic InRelease\n","Hit:11 http://archive.ubuntu.com/ubuntu bionic-updates InRelease\n","Hit:12 http://ppa.launchpad.net/cran/libgit2/ubuntu bionic InRelease\n","Hit:13 http://archive.ubuntu.com/ubuntu bionic-backports InRelease\n","Hit:14 http://ppa.launchpad.net/deadsnakes/ppa/ubuntu bionic InRelease\n","Hit:15 http://ppa.launchpad.net/graphics-drivers/ppa/ubuntu bionic InRelease\n","Hit:16 http://ppa.launchpad.net/ubuntu-toolchain-r/test/ubuntu bionic InRelease\n","Reading package lists... Done\n","Installing libstdc++\n","Reading package lists... Done\n","Building dependency tree \n","Reading state information... Done\n","Selected version '11.1.0-1ubuntu1~18.04.1' (Toolchain test builds:18.04/bionic [amd64]) for 'libstdc++6'\n","The following additional packages will be installed:\n"," gcc-11-base libgcc-s1\n","The following NEW packages will be installed:\n"," gcc-11-base libgcc-s1\n","The following packages will be upgraded:\n"," libstdc++6\n","1 upgraded, 2 newly installed, 0 to remove and 58 not upgraded.\n","Need to get 641 kB of archives.\n","After this operation, 981 kB of additional disk space will be used.\n","Get:1 http://ppa.launchpad.net/ubuntu-toolchain-r/test/ubuntu bionic/main amd64 gcc-11-base amd64 11.1.0-1ubuntu1~18.04.1 [19.0 kB]\n","Get:2 http://ppa.launchpad.net/ubuntu-toolchain-r/test/ubuntu bionic/main amd64 libgcc-s1 amd64 11.1.0-1ubuntu1~18.04.1 [41.8 kB]\n","Get:3 http://ppa.launchpad.net/ubuntu-toolchain-r/test/ubuntu bionic/main amd64 libstdc++6 amd64 11.1.0-1ubuntu1~18.04.1 [580 kB]\n","Fetched 641 kB in 2s (332 kB/s)\n","debconf: unable to initialize frontend: Dialog\n","debconf: (No usable dialog-like program is installed, so the dialog based frontend cannot be used. at /usr/share/perl5/Debconf/FrontEnd/Dialog.pm line 76, <> line 3.)\n","debconf: falling back to frontend: Readline\n","debconf: unable to initialize frontend: Readline\n","debconf: (This frontend requires a controlling tty.)\n","debconf: falling back to frontend: Teletype\n","dpkg-preconfigure: unable to re-open stdin: \n","Selecting previously unselected package gcc-11-base:amd64.\n","(Reading database ... 160772 files and directories currently installed.)\n","Preparing to unpack .../gcc-11-base_11.1.0-1ubuntu1~18.04.1_amd64.deb ...\n","Unpacking gcc-11-base:amd64 (11.1.0-1ubuntu1~18.04.1) ...\n","Setting up gcc-11-base:amd64 (11.1.0-1ubuntu1~18.04.1) ...\n","Selecting previously unselected package libgcc-s1:amd64.\n","(Reading database ... 160777 files and directories currently installed.)\n","Preparing to unpack .../libgcc-s1_11.1.0-1ubuntu1~18.04.1_amd64.deb ...\n","Unpacking libgcc-s1:amd64 (11.1.0-1ubuntu1~18.04.1) ...\n","Replacing files in old package libgcc1:amd64 (1:8.4.0-1ubuntu1~18.04) ...\n","Setting up libgcc-s1:amd64 (11.1.0-1ubuntu1~18.04.1) ...\n","(Reading database ... 160779 files and directories currently installed.)\n","Preparing to unpack .../libstdc++6_11.1.0-1ubuntu1~18.04.1_amd64.deb ...\n","Unpacking libstdc++6:amd64 (11.1.0-1ubuntu1~18.04.1) over (8.4.0-1ubuntu1~18.04) ...\n","Setting up libstdc++6:amd64 (11.1.0-1ubuntu1~18.04.1) ...\n","Processing triggers for libc-bin (2.27-3ubuntu1.2) ...\n","/sbin/ldconfig.real: /usr/local/lib/python3.7/dist-packages/ideep4py/lib/libmkldnn.so.0 is not a symbolic link\n","\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"uajvUZkpEqSe","executionInfo":{"status":"ok","timestamp":1624098593588,"user_tz":-330,"elapsed":34502,"user":{"displayName":"Sparsh Agarwal","photoUrl":"","userId":"13037694610922482904"}},"outputId":"09ba29a8-9b5c-46c7-c66a-c518c40f1a9e"},"source":["# This will install CondaColab. This will restart your kernel one last time. Run this cell by itself and only run the next cell once you see the session crash.\n","import condacolab\n","condacolab.install()"],"execution_count":1,"outputs":[{"output_type":"stream","text":["⏬ Downloading https://github.com/jaimergp/miniforge/releases/latest/download/Mambaforge-colab-Linux-x86_64.sh...\n","📦 Installing...\n","📌 Adjusting configuration...\n","🩹 Patching environment...\n","⏲ Done in 0:00:34\n","🔁 Restarting kernel...\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"HckAjLPjENGz","executionInfo":{"status":"ok","timestamp":1624098602267,"user_tz":-330,"elapsed":631,"user":{"displayName":"Sparsh Agarwal","photoUrl":"","userId":"13037694610922482904"}},"outputId":"6d610243-4e48-4de4-9c24-5d09d7920241"},"source":["# you can now run the rest of the cells as normal\n","import condacolab\n","condacolab.check()"],"execution_count":1,"outputs":[{"output_type":"stream","text":["✨🍰✨ Everything looks OK!\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"7_Ip6zuYG3u-","executionInfo":{"status":"ok","timestamp":1624099394341,"user_tz":-330,"elapsed":785647,"user":{"displayName":"Sparsh Agarwal","photoUrl":"","userId":"13037694610922482904"}},"outputId":"b8dc828f-9f12-4c60-c70c-f201f9ca3780"},"source":["# Installing RAPIDS is now 'python rapidsai-csp-utils/colab/install_rapids.py '\n","# The options are 'stable' and 'nightly'. Leaving it blank or adding any other words will default to stable.\n","# The option are default blank or 'core'. By default, we install RAPIDSAI and BlazingSQL. The 'core' option will install only RAPIDSAI and not include BlazingSQL, \n","!python rapidsai-csp-utils/colab/install_rapids.py stable"],"execution_count":2,"outputs":[{"output_type":"stream","text":["Installing RAPIDS Stable 21.06\n","Starting the RAPIDS+BlazingSQL install on Colab. This will take about 15 minutes.\n","Collecting package metadata (current_repodata.json): ...working... done\n","Solving environment: ...working... failed with initial frozen solve. Retrying with flexible solve.\n","Solving environment: ...working... failed with repodata from current_repodata.json, will retry with next repodata source.\n","Collecting package metadata (repodata.json): ...working... done\n","Solving environment: ...working... done\n","\n","## Package Plan ##\n","\n"," environment location: /usr/local\n","\n"," added / updated specs:\n"," - cudatoolkit=11.0\n"," - gcsfs\n"," - llvmlite\n"," - openssl\n"," - python=3.7\n"," - rapids-blazing=21.06\n","\n","\n","The following packages will be downloaded:\n","\n"," package | build\n"," ---------------------------|-----------------\n"," abseil-cpp-20210324.1 | h9c3ff4c_0 1015 KB conda-forge\n"," aiohttp-3.7.4.post0 | py37h5e8e339_0 625 KB conda-forge\n"," anyio-3.2.0 | py37h89c1867_0 138 KB conda-forge\n"," appdirs-1.4.4 | pyh9f0ad1d_0 13 KB conda-forge\n"," argon2-cffi-20.1.0 | py37h5e8e339_2 47 KB conda-forge\n"," arrow-cpp-1.0.1 |py37haa335b2_40_cuda 21.1 MB conda-forge\n"," arrow-cpp-proc-3.0.0 | cuda 24 KB conda-forge\n"," async-timeout-3.0.1 | py_1000 11 KB conda-forge\n"," async_generator-1.10 | py_0 18 KB conda-forge\n"," attrs-21.2.0 | pyhd8ed1ab_0 44 KB conda-forge\n"," aws-c-cal-0.5.11 | h95a6274_0 37 KB conda-forge\n"," aws-c-common-0.6.2 | h7f98852_0 168 KB conda-forge\n"," aws-c-event-stream-0.2.7 | h211b232_12 47 KB conda-forge\n"," aws-c-io-0.10.4 | hfb6a706_1 121 KB conda-forge\n"," aws-checksums-0.1.11 | ha31a3da_7 50 KB conda-forge\n"," aws-sdk-cpp-1.8.186 | hb4091e7_3 4.6 MB conda-forge\n"," backcall-0.2.0 | pyh9f0ad1d_0 13 KB conda-forge\n"," backports-1.0 | py_2 4 KB conda-forge\n"," backports.functools_lru_cache-1.6.4| pyhd8ed1ab_0 9 KB conda-forge\n"," blazingsql-21.06.00 |cuda_11.0_py37_g95ff589f8_0 190.2 MB rapidsai\n"," bleach-3.3.0 | pyh44b312d_0 111 KB conda-forge\n"," blinker-1.4 | py_1 13 KB conda-forge\n"," bokeh-2.2.3 | py37h89c1867_0 7.0 MB conda-forge\n"," boost-1.72.0 | py37h48f8a5e_1 339 KB conda-forge\n"," boost-cpp-1.72.0 | h9d3c048_4 16.3 MB conda-forge\n"," brotli-1.0.9 | h9c3ff4c_4 389 KB conda-forge\n"," ca-certificates-2021.5.30 | ha878542_0 136 KB conda-forge\n"," cachetools-4.2.2 | pyhd8ed1ab_0 12 KB conda-forge\n"," cairo-1.16.0 | h6cf1ce9_1008 1.5 MB conda-forge\n"," certifi-2021.5.30 | py37h89c1867_0 141 KB conda-forge\n"," cfitsio-3.470 | hb418390_7 1.3 MB conda-forge\n"," click-7.1.2 | pyh9f0ad1d_0 64 KB conda-forge\n"," click-plugins-1.1.1 | py_0 9 KB conda-forge\n"," cligj-0.7.2 | pyhd8ed1ab_0 10 KB conda-forge\n"," cloudpickle-1.6.0 | py_0 22 KB conda-forge\n"," colorcet-2.0.6 | pyhd8ed1ab_0 1.5 MB conda-forge\n"," conda-4.10.1 | py37h89c1867_0 3.1 MB conda-forge\n"," cudatoolkit-11.0.221 | h6bb024c_0 953.0 MB nvidia\n"," cudf-21.06.01 |cuda_11.0_py37_g101fc0fda4_2 108.4 MB rapidsai\n"," cudf_kafka-21.06.01 |py37_g101fc0fda4_2 1.7 MB rapidsai\n"," cugraph-21.06.00 | py37_gf9ffd2de_0 65.0 MB rapidsai\n"," cuml-21.06.02 |cuda11.0_py37_g7dfbf8d9e_0 78.9 MB rapidsai\n"," cupy-9.0.0 | py37h4fdb0f7_0 50.3 MB conda-forge\n"," curl-7.77.0 | hea6ffbf_0 149 KB conda-forge\n"," cusignal-21.06.00 | py38_ga78207b_0 1.0 MB rapidsai\n"," cuspatial-21.06.00 | py37_g37798cd_0 15.2 MB rapidsai\n"," custreamz-21.06.01 |py37_g101fc0fda4_2 32 KB rapidsai\n"," cuxfilter-21.06.00 | py37_g9459467_0 136 KB rapidsai\n"," cycler-0.10.0 | py_2 9 KB conda-forge\n"," cyrus-sasl-2.1.27 | h230043b_2 224 KB conda-forge\n"," cytoolz-0.11.0 | py37h5e8e339_3 403 KB conda-forge\n"," dask-2021.5.0 | pyhd8ed1ab_0 4 KB conda-forge\n"," dask-core-2021.5.0 | pyhd8ed1ab_0 735 KB conda-forge\n"," dask-cuda-21.06.00 | py37_0 110 KB rapidsai\n"," dask-cudf-21.06.01 |py37_g101fc0fda4_2 103 KB rapidsai\n"," datashader-0.11.1 | pyh9f0ad1d_0 14.0 MB conda-forge\n"," datashape-0.5.4 | py_1 49 KB conda-forge\n"," decorator-4.4.2 | py_0 11 KB conda-forge\n"," defusedxml-0.7.1 | pyhd8ed1ab_0 23 KB conda-forge\n"," distributed-2021.5.0 | py37h89c1867_0 1.1 MB conda-forge\n"," dlpack-0.5 | h9c3ff4c_0 12 KB conda-forge\n"," entrypoints-0.3 | pyhd8ed1ab_1003 8 KB conda-forge\n"," expat-2.4.1 | h9c3ff4c_0 182 KB conda-forge\n"," faiss-proc-1.0.0 | cuda 24 KB rapidsai\n"," fastavro-1.4.1 | py37h5e8e339_0 496 KB conda-forge\n"," fastrlock-0.6 | py37hcd2ae1e_0 31 KB conda-forge\n"," fiona-1.8.20 | py37ha0cc35a_0 1.1 MB conda-forge\n"," fontconfig-2.13.1 | hba837de_1005 357 KB conda-forge\n"," freetype-2.10.4 | h0708190_1 890 KB conda-forge\n"," freexl-1.0.6 | h7f98852_0 48 KB conda-forge\n"," fsspec-2021.6.0 | pyhd8ed1ab_0 79 KB conda-forge\n"," future-0.18.2 | py37h89c1867_3 714 KB conda-forge\n"," gcsfs-2021.6.0 | pyhd8ed1ab_0 23 KB conda-forge\n"," gdal-3.2.2 | py37hb0e9ad2_0 1.5 MB conda-forge\n"," geopandas-0.9.0 | pyhd8ed1ab_1 5 KB conda-forge\n"," geopandas-base-0.9.0 | pyhd8ed1ab_1 950 KB conda-forge\n"," geos-3.9.1 | h9c3ff4c_2 1.1 MB conda-forge\n"," geotiff-1.6.0 | hcf90da6_5 296 KB conda-forge\n"," gettext-0.19.8.1 | h0b5b191_1005 3.6 MB conda-forge\n"," gflags-2.2.2 | he1b5a44_1004 114 KB conda-forge\n"," giflib-5.2.1 | h36c2ea0_2 77 KB conda-forge\n"," glog-0.5.0 | h48cff8f_0 104 KB conda-forge\n"," google-auth-1.30.2 | pyh6c4a22f_0 77 KB conda-forge\n"," google-auth-oauthlib-0.4.4 | pyhd8ed1ab_0 19 KB conda-forge\n"," google-cloud-cpp-1.28.0 | hbd34f9f_0 9.3 MB conda-forge\n"," greenlet-1.1.0 | py37hcd2ae1e_0 83 KB conda-forge\n"," grpc-cpp-1.38.0 | h2519f57_0 3.6 MB conda-forge\n"," hdf4-4.2.15 | h10796ff_3 950 KB conda-forge\n"," hdf5-1.10.6 |nompi_h6a2412b_1114 3.1 MB conda-forge\n"," heapdict-1.0.1 | py_0 7 KB conda-forge\n"," importlib-metadata-4.5.0 | py37h89c1867_0 31 KB conda-forge\n"," ipykernel-5.5.5 | py37h085eea5_0 167 KB conda-forge\n"," ipython-7.24.1 | py37h085eea5_0 1.1 MB conda-forge\n"," ipython_genutils-0.2.0 | py_1 21 KB conda-forge\n"," ipywidgets-7.6.3 | pyhd3deb0d_0 101 KB conda-forge\n"," jedi-0.18.0 | py37h89c1867_2 923 KB conda-forge\n"," jinja2-3.0.1 | pyhd8ed1ab_0 99 KB conda-forge\n"," joblib-1.0.1 | pyhd8ed1ab_0 206 KB conda-forge\n"," jpeg-9d | h36c2ea0_0 264 KB conda-forge\n"," jpype1-1.3.0 | py37h2527ec5_0 482 KB conda-forge\n"," json-c-0.15 | h98cffda_0 274 KB conda-forge\n"," jsonschema-3.2.0 | pyhd8ed1ab_3 45 KB conda-forge\n"," jupyter-server-proxy-3.0.2 | pyhd8ed1ab_0 27 KB conda-forge\n"," jupyter_client-6.1.12 | pyhd8ed1ab_0 79 KB conda-forge\n"," jupyter_core-4.7.1 | py37h89c1867_0 72 KB conda-forge\n"," jupyter_server-1.8.0 | pyhd8ed1ab_0 255 KB conda-forge\n"," jupyterlab_pygments-0.1.2 | pyh9f0ad1d_0 8 KB conda-forge\n"," jupyterlab_widgets-1.0.0 | pyhd8ed1ab_1 130 KB conda-forge\n"," kealib-1.4.14 | hcc255d8_2 186 KB conda-forge\n"," kiwisolver-1.3.1 | py37h2527ec5_1 78 KB conda-forge\n"," krb5-1.19.1 | hcc1bbae_0 1.4 MB conda-forge\n"," lcms2-2.12 | hddcbb42_0 443 KB conda-forge\n"," libblas-3.9.0 | 9_openblas 11 KB conda-forge\n"," libcblas-3.9.0 | 9_openblas 11 KB conda-forge\n"," libcrc32c-1.1.1 | h9c3ff4c_2 20 KB conda-forge\n"," libcudf-21.06.01 |cuda11.0_g101fc0fda4_2 187.7 MB rapidsai\n"," libcudf_kafka-21.06.01 | g101fc0fda4_2 125 KB rapidsai\n"," libcugraph-21.06.00 |cuda11.0_gf9ffd2de_0 213.6 MB rapidsai\n"," libcuml-21.06.02 |cuda11.0_g7dfbf8d9e_0 95.2 MB rapidsai\n"," libcumlprims-21.06.00 |cuda11.0_gfda2e6c_0 1.1 MB nvidia\n"," libcurl-7.77.0 | h2574ce0_0 334 KB conda-forge\n"," libcuspatial-21.06.00 |cuda11.0_g37798cd_0 7.6 MB rapidsai\n"," libdap4-3.20.6 | hd7c4107_2 11.3 MB conda-forge\n"," libevent-2.1.10 | hcdb4288_3 1.1 MB conda-forge\n"," libfaiss-1.7.0 |cuda110h8045045_8_cuda 67.0 MB conda-forge\n"," libgcrypt-1.9.3 | h7f98852_1 677 KB conda-forge\n"," libgdal-3.2.2 | h804b7da_0 13.2 MB conda-forge\n"," libgfortran-ng-9.3.0 | hff62375_19 22 KB conda-forge\n"," libgfortran5-9.3.0 | hff62375_19 2.0 MB conda-forge\n"," libglib-2.68.3 | h3e27bee_0 3.1 MB conda-forge\n"," libgpg-error-1.42 | h9c3ff4c_0 278 KB conda-forge\n"," libgsasl-1.8.0 | 2 125 KB conda-forge\n"," libhwloc-2.3.0 | h5e5b7d1_1 2.7 MB conda-forge\n"," libkml-1.3.0 | hd79254b_1012 640 KB conda-forge\n"," liblapack-3.9.0 | 9_openblas 11 KB conda-forge\n"," libllvm10-10.0.1 | he513fc3_3 26.4 MB conda-forge\n"," libnetcdf-4.7.4 |nompi_h56d31a8_107 1.3 MB conda-forge\n"," libntlm-1.4 | h7f98852_1002 32 KB conda-forge\n"," libopenblas-0.3.15 |pthreads_h8fe5266_1 9.2 MB conda-forge\n"," libpng-1.6.37 | h21135ba_2 306 KB conda-forge\n"," libpq-13.3 | hd57d9b9_0 2.7 MB conda-forge\n"," libprotobuf-3.16.0 | h780b84a_0 2.5 MB conda-forge\n"," librdkafka-1.5.3 | hc49e61c_1 11.2 MB conda-forge\n"," librmm-21.06.00 |cuda11.0_gee432a0_0 57 KB rapidsai\n"," librttopo-1.1.0 | h1185371_6 235 KB conda-forge\n"," libsodium-1.0.18 | h36c2ea0_1 366 KB conda-forge\n"," libspatialindex-1.9.3 | h9c3ff4c_3 4.6 MB conda-forge\n"," libspatialite-5.0.1 | h20cb978_4 4.4 MB conda-forge\n"," libthrift-0.14.1 | he6d91bd_1 4.5 MB conda-forge\n"," libtiff-4.2.0 | hbd63e13_2 639 KB conda-forge\n"," libutf8proc-2.6.1 | h7f98852_0 95 KB conda-forge\n"," libuuid-2.32.1 | h7f98852_1000 28 KB conda-forge\n"," libuv-1.41.0 | h7f98852_0 1.0 MB conda-forge\n"," libwebp-1.2.0 | h3452ae3_0 85 KB conda-forge\n"," libwebp-base-1.2.0 | h7f98852_2 815 KB conda-forge\n"," libxcb-1.13 | h7f98852_1003 395 KB conda-forge\n"," libxgboost-1.4.2dev.rapidsai21.06| cuda11.0_0 115.3 MB rapidsai\n"," libxml2-2.9.12 | h72842e0_0 772 KB conda-forge\n"," llvmlite-0.36.0 | py37h9d7f4d0_0 2.7 MB conda-forge\n"," locket-0.2.0 | py_2 6 KB conda-forge\n"," mapclassify-2.4.2 | pyhd8ed1ab_0 36 KB conda-forge\n"," markdown-3.3.4 | pyhd8ed1ab_0 67 KB conda-forge\n"," markupsafe-2.0.1 | py37h5e8e339_0 22 KB conda-forge\n"," matplotlib-base-3.4.2 | py37hdd32ed1_0 7.2 MB conda-forge\n"," matplotlib-inline-0.1.2 | pyhd8ed1ab_2 11 KB conda-forge\n"," mistune-0.8.4 |py37h5e8e339_1003 54 KB conda-forge\n"," msgpack-python-1.0.2 | py37h2527ec5_1 91 KB conda-forge\n"," multidict-5.1.0 | py37h5e8e339_1 67 KB conda-forge\n"," multipledispatch-0.6.0 | py_0 12 KB conda-forge\n"," munch-2.5.0 | py_0 12 KB conda-forge\n"," nbclient-0.5.3 | pyhd8ed1ab_0 67 KB conda-forge\n"," nbconvert-6.0.7 | py37h89c1867_3 535 KB conda-forge\n"," nbformat-5.1.3 | pyhd8ed1ab_0 47 KB conda-forge\n"," nccl-2.9.9.1 | h96e36e3_0 82.3 MB conda-forge\n"," nest-asyncio-1.5.1 | pyhd8ed1ab_0 9 KB conda-forge\n"," netifaces-0.10.9 |py37h5e8e339_1003 17 KB conda-forge\n"," networkx-2.5.1 | pyhd8ed1ab_0 1.2 MB conda-forge\n"," nlohmann_json-3.9.1 | h9c3ff4c_1 122 KB conda-forge\n"," nodejs-14.15.4 | h92b4a50_1 15.7 MB conda-forge\n"," notebook-6.4.0 | pyha770c72_0 6.1 MB conda-forge\n"," numba-0.53.1 | py37hb11d6e1_1 3.7 MB conda-forge\n"," numpy-1.20.3 | py37h038b26d_1 5.7 MB conda-forge\n"," nvtx-0.2.3 | py37h5e8e339_0 55 KB conda-forge\n"," oauthlib-3.1.1 | pyhd8ed1ab_0 87 KB conda-forge\n"," olefile-0.46 | pyh9f0ad1d_1 32 KB conda-forge\n"," openjdk-8.0.282 | h7f98852_0 99.3 MB conda-forge\n"," openjpeg-2.4.0 | hb52868f_1 444 KB conda-forge\n"," openssl-1.1.1k | h7f98852_0 2.1 MB conda-forge\n"," orc-1.6.7 | h89a63ab_2 751 KB conda-forge\n"," packaging-20.9 | pyh44b312d_0 35 KB conda-forge\n"," pandas-1.2.4 | py37h219a48f_0 11.8 MB conda-forge\n"," pandoc-2.14.0.2 | h7f98852_0 12.0 MB conda-forge\n"," pandocfilters-1.4.2 | py_1 9 KB conda-forge\n"," panel-0.10.3 | pyhd8ed1ab_0 6.1 MB conda-forge\n"," param-1.10.1 | pyhd3deb0d_0 64 KB conda-forge\n"," parquet-cpp-1.5.1 | 2 3 KB conda-forge\n"," parso-0.8.2 | pyhd8ed1ab_0 68 KB conda-forge\n"," partd-1.2.0 | pyhd8ed1ab_0 18 KB conda-forge\n"," pcre-8.45 | h9c3ff4c_0 253 KB conda-forge\n"," pexpect-4.8.0 | pyh9f0ad1d_2 47 KB conda-forge\n"," pickle5-0.0.11 | py37h5e8e339_0 173 KB conda-forge\n"," pickleshare-0.7.5 | py_1003 9 KB conda-forge\n"," pillow-8.2.0 | py37h4600e1f_1 684 KB conda-forge\n"," pixman-0.40.0 | h36c2ea0_0 627 KB conda-forge\n"," poppler-21.03.0 | h93df280_0 15.9 MB conda-forge\n"," poppler-data-0.4.10 | 0 3.8 MB conda-forge\n"," postgresql-13.3 | h2510834_0 5.3 MB conda-forge\n"," proj-8.0.0 | h277dcde_0 3.1 MB conda-forge\n"," prometheus_client-0.11.0 | pyhd8ed1ab_0 46 KB conda-forge\n"," prompt-toolkit-3.0.19 | pyha770c72_0 244 KB conda-forge\n"," protobuf-3.16.0 | py37hcd2ae1e_0 342 KB conda-forge\n"," psutil-5.8.0 | py37h5e8e339_1 342 KB conda-forge\n"," pthread-stubs-0.4 | h36c2ea0_1001 5 KB conda-forge\n"," ptyprocess-0.7.0 | pyhd3deb0d_0 16 KB conda-forge\n"," py-xgboost-1.4.2dev.rapidsai21.06| cuda11.0py37_0 151 KB rapidsai\n"," pyarrow-1.0.1 |py37hb63ea2f_40_cuda 2.4 MB conda-forge\n"," pyasn1-0.4.8 | py_0 53 KB conda-forge\n"," pyasn1-modules-0.2.7 | py_0 60 KB conda-forge\n"," pyct-0.4.6 | py_0 3 KB conda-forge\n"," pyct-core-0.4.6 | py_0 13 KB conda-forge\n"," pydeck-0.5.0 | pyh9f0ad1d_0 3.6 MB conda-forge\n"," pyee-7.0.4 | pyh9f0ad1d_0 14 KB conda-forge\n"," pygments-2.9.0 | pyhd8ed1ab_0 754 KB conda-forge\n"," pyhive-0.6.4 | pyhd8ed1ab_0 39 KB conda-forge\n"," pyjwt-2.1.0 | pyhd8ed1ab_0 17 KB conda-forge\n"," pynvml-11.0.0 | pyhd8ed1ab_0 39 KB conda-forge\n"," pyparsing-2.4.7 | pyh9f0ad1d_0 60 KB conda-forge\n"," pyppeteer-0.2.2 | py_1 104 KB conda-forge\n"," pyproj-3.0.1 | py37h2bb2a07_1 484 KB conda-forge\n"," pyrsistent-0.17.3 | py37h5e8e339_2 89 KB conda-forge\n"," python-confluent-kafka-1.5.0| py37h8f50634_0 122 KB conda-forge\n"," python-dateutil-2.8.1 | py_0 220 KB conda-forge\n"," pytz-2021.1 | pyhd8ed1ab_0 239 KB conda-forge\n"," pyu2f-0.1.5 | pyhd8ed1ab_0 31 KB conda-forge\n"," pyviz_comms-2.0.2 | pyhd8ed1ab_0 25 KB conda-forge\n"," pyyaml-5.4.1 | py37h5e8e339_0 189 KB conda-forge\n"," pyzmq-22.1.0 | py37h336d617_0 500 KB conda-forge\n"," rapids-21.06.00 |cuda11.0_py37_ge3c8282_427 5 KB rapidsai\n"," rapids-blazing-21.06.00 |cuda11.0_py37_ge3c8282_427 5 KB rapidsai\n"," rapids-xgboost-21.06.00 |cuda11.0_py37_ge3c8282_427 4 KB rapidsai\n"," re2-2021.04.01 | h9c3ff4c_0 218 KB conda-forge\n"," readline-8.1 | h46c0cb4_0 295 KB conda-forge\n"," requests-oauthlib-1.3.0 | pyh9f0ad1d_0 21 KB conda-forge\n"," rmm-21.06.00 |cuda_11.0_py37_gee432a0_0 7.0 MB rapidsai\n"," rsa-4.7.2 | pyh44b312d_0 28 KB conda-forge\n"," rtree-0.9.7 | py37h0b55af0_1 45 KB conda-forge\n"," s2n-1.0.10 | h9b69904_0 442 KB conda-forge\n"," sasl-0.3a1 | py37hcd2ae1e_0 74 KB conda-forge\n"," scikit-learn-0.24.2 | py37h18a542f_0 7.5 MB conda-forge\n"," scipy-1.6.3 | py37h29e03ee_0 20.5 MB conda-forge\n"," send2trash-1.5.0 | py_0 12 KB conda-forge\n"," shapely-1.7.1 | py37h2d1e849_5 438 KB conda-forge\n"," simpervisor-0.4 | pyhd8ed1ab_0 9 KB conda-forge\n"," snappy-1.1.8 | he1b5a44_3 32 KB conda-forge\n"," sniffio-1.2.0 | py37h89c1867_1 15 KB conda-forge\n"," sortedcontainers-2.4.0 | pyhd8ed1ab_0 26 KB conda-forge\n"," spdlog-1.8.5 | h4bd325d_0 353 KB conda-forge\n"," sqlalchemy-1.4.18 | py37h5e8e339_0 2.2 MB conda-forge\n"," streamz-0.6.2 | pyh44b312d_0 59 KB conda-forge\n"," tblib-1.7.0 | pyhd8ed1ab_0 15 KB conda-forge\n"," terminado-0.10.1 | py37h89c1867_0 26 KB conda-forge\n"," testpath-0.5.0 | pyhd8ed1ab_0 86 KB conda-forge\n"," threadpoolctl-2.1.0 | pyh5ca1d4c_0 15 KB conda-forge\n"," thrift-0.13.0 | py37hcd2ae1e_2 120 KB conda-forge\n"," thrift_sasl-0.4.2 | py37h8f50634_0 14 KB conda-forge\n"," tiledb-2.2.9 | h91fcb0e_0 4.0 MB conda-forge\n"," toolz-0.11.1 | py_0 46 KB conda-forge\n"," tornado-6.1 | py37h5e8e339_1 646 KB conda-forge\n"," traitlets-5.0.5 | py_0 81 KB conda-forge\n"," treelite-1.3.0 | py37hfdac9b6_0 2.7 MB conda-forge\n"," typing-extensions-3.10.0.0 | hd8ed1ab_0 8 KB conda-forge\n"," typing_extensions-3.10.0.0 | pyha770c72_0 28 KB conda-forge\n"," tzcode-2021a | h7f98852_1 68 KB conda-forge\n"," tzdata-2021a | he74cb21_0 121 KB conda-forge\n"," ucx-1.9.0+gcd9efd3 | cuda11.0_0 8.2 MB rapidsai\n"," ucx-proc-1.0.0 | gpu 9 KB rapidsai\n"," ucx-py-0.20.0 | py37_gcd9efd3_0 294 KB rapidsai\n"," wcwidth-0.2.5 | pyh9f0ad1d_2 33 KB conda-forge\n"," webencodings-0.5.1 | py_1 12 KB conda-forge\n"," websocket-client-0.57.0 | py37h89c1867_4 59 KB conda-forge\n"," websockets-8.1 | py37h5e8e339_3 90 KB conda-forge\n"," widgetsnbextension-3.5.1 | py37h89c1867_4 1.8 MB conda-forge\n"," xarray-0.18.2 | pyhd8ed1ab_0 599 KB conda-forge\n"," xerces-c-3.2.3 | h9d8b166_2 1.8 MB conda-forge\n"," xgboost-1.4.2dev.rapidsai21.06| cuda11.0py37_0 17 KB rapidsai\n"," xorg-kbproto-1.0.7 | h7f98852_1002 27 KB conda-forge\n"," xorg-libice-1.0.10 | h7f98852_0 58 KB conda-forge\n"," xorg-libsm-1.2.3 | hd9c2040_1000 26 KB conda-forge\n"," xorg-libx11-1.7.2 | h7f98852_0 941 KB conda-forge\n"," xorg-libxau-1.0.9 | h7f98852_0 13 KB conda-forge\n"," xorg-libxdmcp-1.1.3 | h7f98852_0 19 KB conda-forge\n"," xorg-libxext-1.3.4 | h7f98852_1 54 KB conda-forge\n"," xorg-libxrender-0.9.10 | h7f98852_1003 32 KB conda-forge\n"," xorg-renderproto-0.11.1 | h7f98852_1002 9 KB conda-forge\n"," xorg-xextproto-7.3.0 | h7f98852_1002 28 KB conda-forge\n"," xorg-xproto-7.0.31 | h7f98852_1007 73 KB conda-forge\n"," yarl-1.6.3 | py37h5e8e339_1 141 KB conda-forge\n"," zeromq-4.3.4 | h9c3ff4c_0 352 KB conda-forge\n"," zict-2.0.0 | py_0 10 KB conda-forge\n"," zipp-3.4.1 | pyhd8ed1ab_0 11 KB conda-forge\n"," ------------------------------------------------------------\n"," Total: 2.67 GB\n","\n","The following NEW packages will be INSTALLED:\n","\n"," abseil-cpp conda-forge/linux-64::abseil-cpp-20210324.1-h9c3ff4c_0\n"," aiohttp conda-forge/linux-64::aiohttp-3.7.4.post0-py37h5e8e339_0\n"," anyio conda-forge/linux-64::anyio-3.2.0-py37h89c1867_0\n"," appdirs conda-forge/noarch::appdirs-1.4.4-pyh9f0ad1d_0\n"," argon2-cffi conda-forge/linux-64::argon2-cffi-20.1.0-py37h5e8e339_2\n"," arrow-cpp conda-forge/linux-64::arrow-cpp-1.0.1-py37haa335b2_40_cuda\n"," arrow-cpp-proc conda-forge/linux-64::arrow-cpp-proc-3.0.0-cuda\n"," async-timeout conda-forge/noarch::async-timeout-3.0.1-py_1000\n"," async_generator conda-forge/noarch::async_generator-1.10-py_0\n"," attrs conda-forge/noarch::attrs-21.2.0-pyhd8ed1ab_0\n"," aws-c-cal conda-forge/linux-64::aws-c-cal-0.5.11-h95a6274_0\n"," aws-c-common conda-forge/linux-64::aws-c-common-0.6.2-h7f98852_0\n"," aws-c-event-stream conda-forge/linux-64::aws-c-event-stream-0.2.7-h211b232_12\n"," aws-c-io conda-forge/linux-64::aws-c-io-0.10.4-hfb6a706_1\n"," aws-checksums conda-forge/linux-64::aws-checksums-0.1.11-ha31a3da_7\n"," aws-sdk-cpp conda-forge/linux-64::aws-sdk-cpp-1.8.186-hb4091e7_3\n"," backcall conda-forge/noarch::backcall-0.2.0-pyh9f0ad1d_0\n"," backports conda-forge/noarch::backports-1.0-py_2\n"," backports.functoo~ conda-forge/noarch::backports.functools_lru_cache-1.6.4-pyhd8ed1ab_0\n"," blazingsql rapidsai/linux-64::blazingsql-21.06.00-cuda_11.0_py37_g95ff589f8_0\n"," bleach conda-forge/noarch::bleach-3.3.0-pyh44b312d_0\n"," blinker conda-forge/noarch::blinker-1.4-py_1\n"," bokeh conda-forge/linux-64::bokeh-2.2.3-py37h89c1867_0\n"," boost conda-forge/linux-64::boost-1.72.0-py37h48f8a5e_1\n"," boost-cpp conda-forge/linux-64::boost-cpp-1.72.0-h9d3c048_4\n"," brotli conda-forge/linux-64::brotli-1.0.9-h9c3ff4c_4\n"," cachetools conda-forge/noarch::cachetools-4.2.2-pyhd8ed1ab_0\n"," cairo conda-forge/linux-64::cairo-1.16.0-h6cf1ce9_1008\n"," cfitsio conda-forge/linux-64::cfitsio-3.470-hb418390_7\n"," click conda-forge/noarch::click-7.1.2-pyh9f0ad1d_0\n"," click-plugins conda-forge/noarch::click-plugins-1.1.1-py_0\n"," cligj conda-forge/noarch::cligj-0.7.2-pyhd8ed1ab_0\n"," cloudpickle conda-forge/noarch::cloudpickle-1.6.0-py_0\n"," colorcet conda-forge/noarch::colorcet-2.0.6-pyhd8ed1ab_0\n"," cudatoolkit nvidia/linux-64::cudatoolkit-11.0.221-h6bb024c_0\n"," cudf rapidsai/linux-64::cudf-21.06.01-cuda_11.0_py37_g101fc0fda4_2\n"," cudf_kafka rapidsai/linux-64::cudf_kafka-21.06.01-py37_g101fc0fda4_2\n"," cugraph rapidsai/linux-64::cugraph-21.06.00-py37_gf9ffd2de_0\n"," cuml rapidsai/linux-64::cuml-21.06.02-cuda11.0_py37_g7dfbf8d9e_0\n"," cupy conda-forge/linux-64::cupy-9.0.0-py37h4fdb0f7_0\n"," curl conda-forge/linux-64::curl-7.77.0-hea6ffbf_0\n"," cusignal rapidsai/noarch::cusignal-21.06.00-py38_ga78207b_0\n"," cuspatial rapidsai/linux-64::cuspatial-21.06.00-py37_g37798cd_0\n"," custreamz rapidsai/linux-64::custreamz-21.06.01-py37_g101fc0fda4_2\n"," cuxfilter rapidsai/linux-64::cuxfilter-21.06.00-py37_g9459467_0\n"," cycler conda-forge/noarch::cycler-0.10.0-py_2\n"," cyrus-sasl conda-forge/linux-64::cyrus-sasl-2.1.27-h230043b_2\n"," cytoolz conda-forge/linux-64::cytoolz-0.11.0-py37h5e8e339_3\n"," dask conda-forge/noarch::dask-2021.5.0-pyhd8ed1ab_0\n"," dask-core conda-forge/noarch::dask-core-2021.5.0-pyhd8ed1ab_0\n"," dask-cuda rapidsai/linux-64::dask-cuda-21.06.00-py37_0\n"," dask-cudf rapidsai/linux-64::dask-cudf-21.06.01-py37_g101fc0fda4_2\n"," datashader conda-forge/noarch::datashader-0.11.1-pyh9f0ad1d_0\n"," datashape conda-forge/noarch::datashape-0.5.4-py_1\n"," decorator conda-forge/noarch::decorator-4.4.2-py_0\n"," defusedxml conda-forge/noarch::defusedxml-0.7.1-pyhd8ed1ab_0\n"," distributed conda-forge/linux-64::distributed-2021.5.0-py37h89c1867_0\n"," dlpack conda-forge/linux-64::dlpack-0.5-h9c3ff4c_0\n"," entrypoints conda-forge/noarch::entrypoints-0.3-pyhd8ed1ab_1003\n"," expat conda-forge/linux-64::expat-2.4.1-h9c3ff4c_0\n"," faiss-proc rapidsai/linux-64::faiss-proc-1.0.0-cuda\n"," fastavro conda-forge/linux-64::fastavro-1.4.1-py37h5e8e339_0\n"," fastrlock conda-forge/linux-64::fastrlock-0.6-py37hcd2ae1e_0\n"," fiona conda-forge/linux-64::fiona-1.8.20-py37ha0cc35a_0\n"," fontconfig conda-forge/linux-64::fontconfig-2.13.1-hba837de_1005\n"," freetype conda-forge/linux-64::freetype-2.10.4-h0708190_1\n"," freexl conda-forge/linux-64::freexl-1.0.6-h7f98852_0\n"," fsspec conda-forge/noarch::fsspec-2021.6.0-pyhd8ed1ab_0\n"," future conda-forge/linux-64::future-0.18.2-py37h89c1867_3\n"," gcsfs conda-forge/noarch::gcsfs-2021.6.0-pyhd8ed1ab_0\n"," gdal conda-forge/linux-64::gdal-3.2.2-py37hb0e9ad2_0\n"," geopandas conda-forge/noarch::geopandas-0.9.0-pyhd8ed1ab_1\n"," geopandas-base conda-forge/noarch::geopandas-base-0.9.0-pyhd8ed1ab_1\n"," geos conda-forge/linux-64::geos-3.9.1-h9c3ff4c_2\n"," geotiff conda-forge/linux-64::geotiff-1.6.0-hcf90da6_5\n"," gettext conda-forge/linux-64::gettext-0.19.8.1-h0b5b191_1005\n"," gflags conda-forge/linux-64::gflags-2.2.2-he1b5a44_1004\n"," giflib conda-forge/linux-64::giflib-5.2.1-h36c2ea0_2\n"," glog conda-forge/linux-64::glog-0.5.0-h48cff8f_0\n"," google-auth conda-forge/noarch::google-auth-1.30.2-pyh6c4a22f_0\n"," google-auth-oauth~ conda-forge/noarch::google-auth-oauthlib-0.4.4-pyhd8ed1ab_0\n"," google-cloud-cpp conda-forge/linux-64::google-cloud-cpp-1.28.0-hbd34f9f_0\n"," greenlet conda-forge/linux-64::greenlet-1.1.0-py37hcd2ae1e_0\n"," grpc-cpp conda-forge/linux-64::grpc-cpp-1.38.0-h2519f57_0\n"," hdf4 conda-forge/linux-64::hdf4-4.2.15-h10796ff_3\n"," hdf5 conda-forge/linux-64::hdf5-1.10.6-nompi_h6a2412b_1114\n"," heapdict conda-forge/noarch::heapdict-1.0.1-py_0\n"," importlib-metadata conda-forge/linux-64::importlib-metadata-4.5.0-py37h89c1867_0\n"," ipykernel conda-forge/linux-64::ipykernel-5.5.5-py37h085eea5_0\n"," ipython conda-forge/linux-64::ipython-7.24.1-py37h085eea5_0\n"," ipython_genutils conda-forge/noarch::ipython_genutils-0.2.0-py_1\n"," ipywidgets conda-forge/noarch::ipywidgets-7.6.3-pyhd3deb0d_0\n"," jedi conda-forge/linux-64::jedi-0.18.0-py37h89c1867_2\n"," jinja2 conda-forge/noarch::jinja2-3.0.1-pyhd8ed1ab_0\n"," joblib conda-forge/noarch::joblib-1.0.1-pyhd8ed1ab_0\n"," jpeg conda-forge/linux-64::jpeg-9d-h36c2ea0_0\n"," jpype1 conda-forge/linux-64::jpype1-1.3.0-py37h2527ec5_0\n"," json-c conda-forge/linux-64::json-c-0.15-h98cffda_0\n"," jsonschema conda-forge/noarch::jsonschema-3.2.0-pyhd8ed1ab_3\n"," jupyter-server-pr~ conda-forge/noarch::jupyter-server-proxy-3.0.2-pyhd8ed1ab_0\n"," jupyter_client conda-forge/noarch::jupyter_client-6.1.12-pyhd8ed1ab_0\n"," jupyter_core conda-forge/linux-64::jupyter_core-4.7.1-py37h89c1867_0\n"," jupyter_server conda-forge/noarch::jupyter_server-1.8.0-pyhd8ed1ab_0\n"," jupyterlab_pygmen~ conda-forge/noarch::jupyterlab_pygments-0.1.2-pyh9f0ad1d_0\n"," jupyterlab_widgets conda-forge/noarch::jupyterlab_widgets-1.0.0-pyhd8ed1ab_1\n"," kealib conda-forge/linux-64::kealib-1.4.14-hcc255d8_2\n"," kiwisolver conda-forge/linux-64::kiwisolver-1.3.1-py37h2527ec5_1\n"," lcms2 conda-forge/linux-64::lcms2-2.12-hddcbb42_0\n"," libblas conda-forge/linux-64::libblas-3.9.0-9_openblas\n"," libcblas conda-forge/linux-64::libcblas-3.9.0-9_openblas\n"," libcrc32c conda-forge/linux-64::libcrc32c-1.1.1-h9c3ff4c_2\n"," libcudf rapidsai/linux-64::libcudf-21.06.01-cuda11.0_g101fc0fda4_2\n"," libcudf_kafka rapidsai/linux-64::libcudf_kafka-21.06.01-g101fc0fda4_2\n"," libcugraph rapidsai/linux-64::libcugraph-21.06.00-cuda11.0_gf9ffd2de_0\n"," libcuml rapidsai/linux-64::libcuml-21.06.02-cuda11.0_g7dfbf8d9e_0\n"," libcumlprims nvidia/linux-64::libcumlprims-21.06.00-cuda11.0_gfda2e6c_0\n"," libcuspatial rapidsai/linux-64::libcuspatial-21.06.00-cuda11.0_g37798cd_0\n"," libdap4 conda-forge/linux-64::libdap4-3.20.6-hd7c4107_2\n"," libevent conda-forge/linux-64::libevent-2.1.10-hcdb4288_3\n"," libfaiss conda-forge/linux-64::libfaiss-1.7.0-cuda110h8045045_8_cuda\n"," libgcrypt conda-forge/linux-64::libgcrypt-1.9.3-h7f98852_1\n"," libgdal conda-forge/linux-64::libgdal-3.2.2-h804b7da_0\n"," libgfortran-ng conda-forge/linux-64::libgfortran-ng-9.3.0-hff62375_19\n"," libgfortran5 conda-forge/linux-64::libgfortran5-9.3.0-hff62375_19\n"," libglib conda-forge/linux-64::libglib-2.68.3-h3e27bee_0\n"," libgpg-error conda-forge/linux-64::libgpg-error-1.42-h9c3ff4c_0\n"," libgsasl conda-forge/linux-64::libgsasl-1.8.0-2\n"," libhwloc conda-forge/linux-64::libhwloc-2.3.0-h5e5b7d1_1\n"," libkml conda-forge/linux-64::libkml-1.3.0-hd79254b_1012\n"," liblapack conda-forge/linux-64::liblapack-3.9.0-9_openblas\n"," libllvm10 conda-forge/linux-64::libllvm10-10.0.1-he513fc3_3\n"," libnetcdf conda-forge/linux-64::libnetcdf-4.7.4-nompi_h56d31a8_107\n"," libntlm conda-forge/linux-64::libntlm-1.4-h7f98852_1002\n"," libopenblas conda-forge/linux-64::libopenblas-0.3.15-pthreads_h8fe5266_1\n"," libpng conda-forge/linux-64::libpng-1.6.37-h21135ba_2\n"," libpq conda-forge/linux-64::libpq-13.3-hd57d9b9_0\n"," libprotobuf conda-forge/linux-64::libprotobuf-3.16.0-h780b84a_0\n"," librdkafka conda-forge/linux-64::librdkafka-1.5.3-hc49e61c_1\n"," librmm rapidsai/linux-64::librmm-21.06.00-cuda11.0_gee432a0_0\n"," librttopo conda-forge/linux-64::librttopo-1.1.0-h1185371_6\n"," libsodium conda-forge/linux-64::libsodium-1.0.18-h36c2ea0_1\n"," libspatialindex conda-forge/linux-64::libspatialindex-1.9.3-h9c3ff4c_3\n"," libspatialite conda-forge/linux-64::libspatialite-5.0.1-h20cb978_4\n"," libthrift conda-forge/linux-64::libthrift-0.14.1-he6d91bd_1\n"," libtiff conda-forge/linux-64::libtiff-4.2.0-hbd63e13_2\n"," libutf8proc conda-forge/linux-64::libutf8proc-2.6.1-h7f98852_0\n"," libuuid conda-forge/linux-64::libuuid-2.32.1-h7f98852_1000\n"," libuv conda-forge/linux-64::libuv-1.41.0-h7f98852_0\n"," libwebp conda-forge/linux-64::libwebp-1.2.0-h3452ae3_0\n"," libwebp-base conda-forge/linux-64::libwebp-base-1.2.0-h7f98852_2\n"," libxcb conda-forge/linux-64::libxcb-1.13-h7f98852_1003\n"," libxgboost rapidsai/linux-64::libxgboost-1.4.2dev.rapidsai21.06-cuda11.0_0\n"," llvmlite conda-forge/linux-64::llvmlite-0.36.0-py37h9d7f4d0_0\n"," locket conda-forge/noarch::locket-0.2.0-py_2\n"," mapclassify conda-forge/noarch::mapclassify-2.4.2-pyhd8ed1ab_0\n"," markdown conda-forge/noarch::markdown-3.3.4-pyhd8ed1ab_0\n"," markupsafe conda-forge/linux-64::markupsafe-2.0.1-py37h5e8e339_0\n"," matplotlib-base conda-forge/linux-64::matplotlib-base-3.4.2-py37hdd32ed1_0\n"," matplotlib-inline conda-forge/noarch::matplotlib-inline-0.1.2-pyhd8ed1ab_2\n"," mistune conda-forge/linux-64::mistune-0.8.4-py37h5e8e339_1003\n"," msgpack-python conda-forge/linux-64::msgpack-python-1.0.2-py37h2527ec5_1\n"," multidict conda-forge/linux-64::multidict-5.1.0-py37h5e8e339_1\n"," multipledispatch conda-forge/noarch::multipledispatch-0.6.0-py_0\n"," munch conda-forge/noarch::munch-2.5.0-py_0\n"," nbclient conda-forge/noarch::nbclient-0.5.3-pyhd8ed1ab_0\n"," nbconvert conda-forge/linux-64::nbconvert-6.0.7-py37h89c1867_3\n"," nbformat conda-forge/noarch::nbformat-5.1.3-pyhd8ed1ab_0\n"," nccl conda-forge/linux-64::nccl-2.9.9.1-h96e36e3_0\n"," nest-asyncio conda-forge/noarch::nest-asyncio-1.5.1-pyhd8ed1ab_0\n"," netifaces conda-forge/linux-64::netifaces-0.10.9-py37h5e8e339_1003\n"," networkx conda-forge/noarch::networkx-2.5.1-pyhd8ed1ab_0\n"," nlohmann_json conda-forge/linux-64::nlohmann_json-3.9.1-h9c3ff4c_1\n"," nodejs conda-forge/linux-64::nodejs-14.15.4-h92b4a50_1\n"," notebook conda-forge/noarch::notebook-6.4.0-pyha770c72_0\n"," numba conda-forge/linux-64::numba-0.53.1-py37hb11d6e1_1\n"," numpy conda-forge/linux-64::numpy-1.20.3-py37h038b26d_1\n"," nvtx conda-forge/linux-64::nvtx-0.2.3-py37h5e8e339_0\n"," oauthlib conda-forge/noarch::oauthlib-3.1.1-pyhd8ed1ab_0\n"," olefile conda-forge/noarch::olefile-0.46-pyh9f0ad1d_1\n"," openjdk conda-forge/linux-64::openjdk-8.0.282-h7f98852_0\n"," openjpeg conda-forge/linux-64::openjpeg-2.4.0-hb52868f_1\n"," orc conda-forge/linux-64::orc-1.6.7-h89a63ab_2\n"," packaging conda-forge/noarch::packaging-20.9-pyh44b312d_0\n"," pandas conda-forge/linux-64::pandas-1.2.4-py37h219a48f_0\n"," pandoc conda-forge/linux-64::pandoc-2.14.0.2-h7f98852_0\n"," pandocfilters conda-forge/noarch::pandocfilters-1.4.2-py_1\n"," panel conda-forge/noarch::panel-0.10.3-pyhd8ed1ab_0\n"," param conda-forge/noarch::param-1.10.1-pyhd3deb0d_0\n"," parquet-cpp conda-forge/noarch::parquet-cpp-1.5.1-2\n"," parso conda-forge/noarch::parso-0.8.2-pyhd8ed1ab_0\n"," partd conda-forge/noarch::partd-1.2.0-pyhd8ed1ab_0\n"," pcre conda-forge/linux-64::pcre-8.45-h9c3ff4c_0\n"," pexpect conda-forge/noarch::pexpect-4.8.0-pyh9f0ad1d_2\n"," pickle5 conda-forge/linux-64::pickle5-0.0.11-py37h5e8e339_0\n"," pickleshare conda-forge/noarch::pickleshare-0.7.5-py_1003\n"," pillow conda-forge/linux-64::pillow-8.2.0-py37h4600e1f_1\n"," pixman conda-forge/linux-64::pixman-0.40.0-h36c2ea0_0\n"," poppler conda-forge/linux-64::poppler-21.03.0-h93df280_0\n"," poppler-data conda-forge/noarch::poppler-data-0.4.10-0\n"," postgresql conda-forge/linux-64::postgresql-13.3-h2510834_0\n"," proj conda-forge/linux-64::proj-8.0.0-h277dcde_0\n"," prometheus_client conda-forge/noarch::prometheus_client-0.11.0-pyhd8ed1ab_0\n"," prompt-toolkit conda-forge/noarch::prompt-toolkit-3.0.19-pyha770c72_0\n"," protobuf conda-forge/linux-64::protobuf-3.16.0-py37hcd2ae1e_0\n"," psutil conda-forge/linux-64::psutil-5.8.0-py37h5e8e339_1\n"," pthread-stubs conda-forge/linux-64::pthread-stubs-0.4-h36c2ea0_1001\n"," ptyprocess conda-forge/noarch::ptyprocess-0.7.0-pyhd3deb0d_0\n"," py-xgboost rapidsai/linux-64::py-xgboost-1.4.2dev.rapidsai21.06-cuda11.0py37_0\n"," pyarrow conda-forge/linux-64::pyarrow-1.0.1-py37hb63ea2f_40_cuda\n"," pyasn1 conda-forge/noarch::pyasn1-0.4.8-py_0\n"," pyasn1-modules conda-forge/noarch::pyasn1-modules-0.2.7-py_0\n"," pyct conda-forge/noarch::pyct-0.4.6-py_0\n"," pyct-core conda-forge/noarch::pyct-core-0.4.6-py_0\n"," pydeck conda-forge/noarch::pydeck-0.5.0-pyh9f0ad1d_0\n"," pyee conda-forge/noarch::pyee-7.0.4-pyh9f0ad1d_0\n"," pygments conda-forge/noarch::pygments-2.9.0-pyhd8ed1ab_0\n"," pyhive conda-forge/noarch::pyhive-0.6.4-pyhd8ed1ab_0\n"," pyjwt conda-forge/noarch::pyjwt-2.1.0-pyhd8ed1ab_0\n"," pynvml conda-forge/noarch::pynvml-11.0.0-pyhd8ed1ab_0\n"," pyparsing conda-forge/noarch::pyparsing-2.4.7-pyh9f0ad1d_0\n"," pyppeteer conda-forge/noarch::pyppeteer-0.2.2-py_1\n"," pyproj conda-forge/linux-64::pyproj-3.0.1-py37h2bb2a07_1\n"," pyrsistent conda-forge/linux-64::pyrsistent-0.17.3-py37h5e8e339_2\n"," python-confluent-~ conda-forge/linux-64::python-confluent-kafka-1.5.0-py37h8f50634_0\n"," python-dateutil conda-forge/noarch::python-dateutil-2.8.1-py_0\n"," pytz conda-forge/noarch::pytz-2021.1-pyhd8ed1ab_0\n"," pyu2f conda-forge/noarch::pyu2f-0.1.5-pyhd8ed1ab_0\n"," pyviz_comms conda-forge/noarch::pyviz_comms-2.0.2-pyhd8ed1ab_0\n"," pyyaml conda-forge/linux-64::pyyaml-5.4.1-py37h5e8e339_0\n"," pyzmq conda-forge/linux-64::pyzmq-22.1.0-py37h336d617_0\n"," rapids rapidsai/linux-64::rapids-21.06.00-cuda11.0_py37_ge3c8282_427\n"," rapids-blazing rapidsai/linux-64::rapids-blazing-21.06.00-cuda11.0_py37_ge3c8282_427\n"," rapids-xgboost rapidsai/linux-64::rapids-xgboost-21.06.00-cuda11.0_py37_ge3c8282_427\n"," re2 conda-forge/linux-64::re2-2021.04.01-h9c3ff4c_0\n"," requests-oauthlib conda-forge/noarch::requests-oauthlib-1.3.0-pyh9f0ad1d_0\n"," rmm rapidsai/linux-64::rmm-21.06.00-cuda_11.0_py37_gee432a0_0\n"," rsa conda-forge/noarch::rsa-4.7.2-pyh44b312d_0\n"," rtree conda-forge/linux-64::rtree-0.9.7-py37h0b55af0_1\n"," s2n conda-forge/linux-64::s2n-1.0.10-h9b69904_0\n"," sasl conda-forge/linux-64::sasl-0.3a1-py37hcd2ae1e_0\n"," scikit-learn conda-forge/linux-64::scikit-learn-0.24.2-py37h18a542f_0\n"," scipy conda-forge/linux-64::scipy-1.6.3-py37h29e03ee_0\n"," send2trash conda-forge/noarch::send2trash-1.5.0-py_0\n"," shapely conda-forge/linux-64::shapely-1.7.1-py37h2d1e849_5\n"," simpervisor conda-forge/noarch::simpervisor-0.4-pyhd8ed1ab_0\n"," snappy conda-forge/linux-64::snappy-1.1.8-he1b5a44_3\n"," sniffio conda-forge/linux-64::sniffio-1.2.0-py37h89c1867_1\n"," sortedcontainers conda-forge/noarch::sortedcontainers-2.4.0-pyhd8ed1ab_0\n"," spdlog conda-forge/linux-64::spdlog-1.8.5-h4bd325d_0\n"," sqlalchemy conda-forge/linux-64::sqlalchemy-1.4.18-py37h5e8e339_0\n"," streamz conda-forge/noarch::streamz-0.6.2-pyh44b312d_0\n"," tblib conda-forge/noarch::tblib-1.7.0-pyhd8ed1ab_0\n"," terminado conda-forge/linux-64::terminado-0.10.1-py37h89c1867_0\n"," testpath conda-forge/noarch::testpath-0.5.0-pyhd8ed1ab_0\n"," threadpoolctl conda-forge/noarch::threadpoolctl-2.1.0-pyh5ca1d4c_0\n"," thrift conda-forge/linux-64::thrift-0.13.0-py37hcd2ae1e_2\n"," thrift_sasl conda-forge/linux-64::thrift_sasl-0.4.2-py37h8f50634_0\n"," tiledb conda-forge/linux-64::tiledb-2.2.9-h91fcb0e_0\n"," toolz conda-forge/noarch::toolz-0.11.1-py_0\n"," tornado conda-forge/linux-64::tornado-6.1-py37h5e8e339_1\n"," traitlets conda-forge/noarch::traitlets-5.0.5-py_0\n"," treelite conda-forge/linux-64::treelite-1.3.0-py37hfdac9b6_0\n"," typing-extensions conda-forge/noarch::typing-extensions-3.10.0.0-hd8ed1ab_0\n"," typing_extensions conda-forge/noarch::typing_extensions-3.10.0.0-pyha770c72_0\n"," tzcode conda-forge/linux-64::tzcode-2021a-h7f98852_1\n"," tzdata conda-forge/noarch::tzdata-2021a-he74cb21_0\n"," ucx rapidsai/linux-64::ucx-1.9.0+gcd9efd3-cuda11.0_0\n"," ucx-proc rapidsai/linux-64::ucx-proc-1.0.0-gpu\n"," ucx-py rapidsai/linux-64::ucx-py-0.20.0-py37_gcd9efd3_0\n"," wcwidth conda-forge/noarch::wcwidth-0.2.5-pyh9f0ad1d_2\n"," webencodings conda-forge/noarch::webencodings-0.5.1-py_1\n"," websocket-client conda-forge/linux-64::websocket-client-0.57.0-py37h89c1867_4\n"," websockets conda-forge/linux-64::websockets-8.1-py37h5e8e339_3\n"," widgetsnbextension conda-forge/linux-64::widgetsnbextension-3.5.1-py37h89c1867_4\n"," xarray conda-forge/noarch::xarray-0.18.2-pyhd8ed1ab_0\n"," xerces-c conda-forge/linux-64::xerces-c-3.2.3-h9d8b166_2\n"," xgboost rapidsai/linux-64::xgboost-1.4.2dev.rapidsai21.06-cuda11.0py37_0\n"," xorg-kbproto conda-forge/linux-64::xorg-kbproto-1.0.7-h7f98852_1002\n"," xorg-libice conda-forge/linux-64::xorg-libice-1.0.10-h7f98852_0\n"," xorg-libsm conda-forge/linux-64::xorg-libsm-1.2.3-hd9c2040_1000\n"," xorg-libx11 conda-forge/linux-64::xorg-libx11-1.7.2-h7f98852_0\n"," xorg-libxau conda-forge/linux-64::xorg-libxau-1.0.9-h7f98852_0\n"," xorg-libxdmcp conda-forge/linux-64::xorg-libxdmcp-1.1.3-h7f98852_0\n"," xorg-libxext conda-forge/linux-64::xorg-libxext-1.3.4-h7f98852_1\n"," xorg-libxrender conda-forge/linux-64::xorg-libxrender-0.9.10-h7f98852_1003\n"," xorg-renderproto conda-forge/linux-64::xorg-renderproto-0.11.1-h7f98852_1002\n"," xorg-xextproto conda-forge/linux-64::xorg-xextproto-7.3.0-h7f98852_1002\n"," xorg-xproto conda-forge/linux-64::xorg-xproto-7.0.31-h7f98852_1007\n"," yarl conda-forge/linux-64::yarl-1.6.3-py37h5e8e339_1\n"," zeromq conda-forge/linux-64::zeromq-4.3.4-h9c3ff4c_0\n"," zict conda-forge/noarch::zict-2.0.0-py_0\n"," zipp conda-forge/noarch::zipp-3.4.1-pyhd8ed1ab_0\n","\n","The following packages will be UPDATED:\n","\n"," ca-certificates 2020.12.5-ha878542_0 --> 2021.5.30-ha878542_0\n"," certifi 2020.12.5-py37h89c1867_1 --> 2021.5.30-py37h89c1867_0\n"," conda 4.9.2-py37h89c1867_0 --> 4.10.1-py37h89c1867_0\n"," krb5 1.17.2-h926e7f8_0 --> 1.19.1-hcc1bbae_0\n"," libcurl 7.75.0-hc4aaa36_0 --> 7.77.0-h2574ce0_0\n"," libxml2 2.9.10-h72842e0_3 --> 2.9.12-h72842e0_0\n"," openssl 1.1.1j-h7f98852_0 --> 1.1.1k-h7f98852_0\n"," readline 8.0-he28a2e2_2 --> 8.1-h46c0cb4_0\n","\n","\n","\n","Downloading and Extracting Packages\n","\n","thrift-0.13.0 | 120 KB | | 0%\n","thrift-0.13.0 | 120 KB | #3 | 13%\n","thrift-0.13.0 | 120 KB | ########## | 100%\n","\n","libcudf_kafka-21.06. | 125 KB | | 0%\n","libcudf_kafka-21.06. | 125 KB | #2 | 13%\n","libcudf_kafka-21.06. | 125 KB | ######3 | 64%\n","libcudf_kafka-21.06. | 125 KB | ########## | 100%\n","\n","nvtx-0.2.3 | 55 KB | | 0%\n","nvtx-0.2.3 | 55 KB | ########## | 100%\n","\n","parquet-cpp-1.5.1 | 3 KB | | 0%\n","parquet-cpp-1.5.1 | 3 KB | ########## | 100%\n","\n","boost-1.72.0 | 339 KB | | 0%\n","boost-1.72.0 | 339 KB | ########## | 100%\n","boost-1.72.0 | 339 KB | ########## | 100%\n","\n","fiona-1.8.20 | 1.1 MB | | 0%\n","fiona-1.8.20 | 1.1 MB | ########## | 100%\n","fiona-1.8.20 | 1.1 MB | ########## | 100%\n","\n","cugraph-21.06.00 | 65.0 MB | | 0%\n","cugraph-21.06.00 | 65.0 MB | | 0%\n","cugraph-21.06.00 | 65.0 MB | | 1%\n","cugraph-21.06.00 | 65.0 MB | 2 | 3%\n","cugraph-21.06.00 | 65.0 MB | 9 | 9%\n","cugraph-21.06.00 | 65.0 MB | #6 | 16%\n","cugraph-21.06.00 | 65.0 MB | ##3 | 23%\n","cugraph-21.06.00 | 65.0 MB | ### | 30%\n","cugraph-21.06.00 | 65.0 MB | ###6 | 37%\n","cugraph-21.06.00 | 65.0 MB | ####3 | 44%\n","cugraph-21.06.00 | 65.0 MB | ##### | 51%\n","cugraph-21.06.00 | 65.0 MB | #####8 | 58%\n","cugraph-21.06.00 | 65.0 MB | ######5 | 65%\n","cugraph-21.06.00 | 65.0 MB | #######3 | 73%\n","cugraph-21.06.00 | 65.0 MB | ######## | 80%\n","cugraph-21.06.00 | 65.0 MB | ########7 | 88%\n","cugraph-21.06.00 | 65.0 MB | #########4 | 94%\n","cugraph-21.06.00 | 65.0 MB | ########## | 100%\n","\n","libllvm10-10.0.1 | 26.4 MB | | 0%\n","libllvm10-10.0.1 | 26.4 MB | 3 | 3%\n","libllvm10-10.0.1 | 26.4 MB | 4 | 4%\n","libllvm10-10.0.1 | 26.4 MB | ###2 | 32%\n","libllvm10-10.0.1 | 26.4 MB | #######9 | 79%\n","libllvm10-10.0.1 | 26.4 MB | ########## | 100%\n","libllvm10-10.0.1 | 26.4 MB | ########## | 100%\n","\n","sasl-0.3a1 | 74 KB | | 0%\n","sasl-0.3a1 | 74 KB | ########## | 100%\n","\n","dask-2021.5.0 | 4 KB | | 0%\n","dask-2021.5.0 | 4 KB | ########## | 100%\n","\n","libxgboost-1.4.2dev. | 115.3 MB | | 0%\n","libxgboost-1.4.2dev. | 115.3 MB | | 0%\n","libxgboost-1.4.2dev. | 115.3 MB | | 0%\n","libxgboost-1.4.2dev. | 115.3 MB | | 0%\n","libxgboost-1.4.2dev. | 115.3 MB | 1 | 1%\n","libxgboost-1.4.2dev. | 115.3 MB | 4 | 5%\n","libxgboost-1.4.2dev. | 115.3 MB | 8 | 9%\n","libxgboost-1.4.2dev. | 115.3 MB | #2 | 13%\n","libxgboost-1.4.2dev. | 115.3 MB | #6 | 17%\n","libxgboost-1.4.2dev. | 115.3 MB | ## | 21%\n","libxgboost-1.4.2dev. | 115.3 MB | ##4 | 24%\n","libxgboost-1.4.2dev. | 115.3 MB | ##8 | 28%\n","libxgboost-1.4.2dev. | 115.3 MB | ###2 | 32%\n","libxgboost-1.4.2dev. | 115.3 MB | ###6 | 36%\n","libxgboost-1.4.2dev. | 115.3 MB | #### | 41%\n","libxgboost-1.4.2dev. | 115.3 MB | ####5 | 45%\n","libxgboost-1.4.2dev. | 115.3 MB | ####9 | 50%\n","libxgboost-1.4.2dev. | 115.3 MB | #####4 | 54%\n","libxgboost-1.4.2dev. | 115.3 MB | #####8 | 59%\n","libxgboost-1.4.2dev. | 115.3 MB | ######3 | 63%\n","libxgboost-1.4.2dev. | 115.3 MB | ######7 | 67%\n","libxgboost-1.4.2dev. | 115.3 MB | #######1 | 72%\n","libxgboost-1.4.2dev. | 115.3 MB | #######5 | 76%\n","libxgboost-1.4.2dev. | 115.3 MB | ######## | 80%\n","libxgboost-1.4.2dev. | 115.3 MB | ########4 | 84%\n","libxgboost-1.4.2dev. | 115.3 MB | ########8 | 89%\n","libxgboost-1.4.2dev. | 115.3 MB | #########3 | 93%\n","libxgboost-1.4.2dev. | 115.3 MB | #########7 | 98%\n","libxgboost-1.4.2dev. | 115.3 MB | ########## | 100%\n","\n","fontconfig-2.13.1 | 357 KB | | 0%\n","fontconfig-2.13.1 | 357 KB | ########## | 100%\n","fontconfig-2.13.1 | 357 KB | ########## | 100%\n","\n","geotiff-1.6.0 | 296 KB | | 0%\n","geotiff-1.6.0 | 296 KB | ########## | 100%\n","\n","custreamz-21.06.01 | 32 KB | | 0%\n","custreamz-21.06.01 | 32 KB | ##### | 50%\n","custreamz-21.06.01 | 32 KB | ########## | 100%\n","\n","aws-sdk-cpp-1.8.186 | 4.6 MB | | 0%\n","aws-sdk-cpp-1.8.186 | 4.6 MB | ########## | 100%\n","aws-sdk-cpp-1.8.186 | 4.6 MB | ########## | 100%\n","\n","arrow-cpp-proc-3.0.0 | 24 KB | | 0%\n","arrow-cpp-proc-3.0.0 | 24 KB | ########## | 100%\n","\n","numba-0.53.1 | 3.7 MB | | 0%\n","numba-0.53.1 | 3.7 MB | ########## | 100%\n","numba-0.53.1 | 3.7 MB | ########## | 100%\n","\n","cligj-0.7.2 | 10 KB | | 0%\n","cligj-0.7.2 | 10 KB | ########## | 100%\n","\n","poppler-21.03.0 | 15.9 MB | | 0%\n","poppler-21.03.0 | 15.9 MB | ####2 | 43%\n","poppler-21.03.0 | 15.9 MB | ########## | 100%\n","poppler-21.03.0 | 15.9 MB | ########## | 100%\n","\n","netifaces-0.10.9 | 17 KB | | 0%\n","netifaces-0.10.9 | 17 KB | ########## | 100%\n","\n","geopandas-0.9.0 | 5 KB | | 0%\n","geopandas-0.9.0 | 5 KB | ########## | 100%\n","\n","libxml2-2.9.12 | 772 KB | | 0%\n","libxml2-2.9.12 | 772 KB | ########## | 100%\n","libxml2-2.9.12 | 772 KB | ########## | 100%\n","\n","decorator-4.4.2 | 11 KB | | 0%\n","decorator-4.4.2 | 11 KB | ########## | 100%\n","\n","libgfortran-ng-9.3.0 | 22 KB | | 0%\n","libgfortran-ng-9.3.0 | 22 KB | ########## | 100%\n","\n","krb5-1.19.1 | 1.4 MB | | 0%\n","krb5-1.19.1 | 1.4 MB | ########## | 100%\n","krb5-1.19.1 | 1.4 MB | ########## | 100%\n","\n","sortedcontainers-2.4 | 26 KB | | 0%\n","sortedcontainers-2.4 | 26 KB | ########## | 100%\n","\n","pynvml-11.0.0 | 39 KB | | 0%\n","pynvml-11.0.0 | 39 KB | ########## | 100%\n","\n","libcudf-21.06.01 | 187.7 MB | | 0%\n","libcudf-21.06.01 | 187.7 MB | | 0%\n","libcudf-21.06.01 | 187.7 MB | | 0%\n","libcudf-21.06.01 | 187.7 MB | | 0%\n","libcudf-21.06.01 | 187.7 MB | 1 | 1%\n","libcudf-21.06.01 | 187.7 MB | 3 | 3%\n","libcudf-21.06.01 | 187.7 MB | 5 | 5%\n","libcudf-21.06.01 | 187.7 MB | 7 | 7%\n","libcudf-21.06.01 | 187.7 MB | 9 | 9%\n","libcudf-21.06.01 | 187.7 MB | #1 | 11%\n","libcudf-21.06.01 | 187.7 MB | #3 | 14%\n","libcudf-21.06.01 | 187.7 MB | #6 | 16%\n","libcudf-21.06.01 | 187.7 MB | #8 | 18%\n","libcudf-21.06.01 | 187.7 MB | ## | 21%\n","libcudf-21.06.01 | 187.7 MB | ##3 | 23%\n","libcudf-21.06.01 | 187.7 MB | ##5 | 26%\n","libcudf-21.06.01 | 187.7 MB | ##8 | 28%\n","libcudf-21.06.01 | 187.7 MB | ### | 30%\n","libcudf-21.06.01 | 187.7 MB | ###2 | 33%\n","libcudf-21.06.01 | 187.7 MB | ###5 | 36%\n","libcudf-21.06.01 | 187.7 MB | ###8 | 39%\n","libcudf-21.06.01 | 187.7 MB | #### | 41%\n","libcudf-21.06.01 | 187.7 MB | ####3 | 44%\n","libcudf-21.06.01 | 187.7 MB | ####6 | 47%\n","libcudf-21.06.01 | 187.7 MB | ####9 | 49%\n","libcudf-21.06.01 | 187.7 MB | #####1 | 52%\n","libcudf-21.06.01 | 187.7 MB | #####4 | 55%\n","libcudf-21.06.01 | 187.7 MB | #####7 | 57%\n","libcudf-21.06.01 | 187.7 MB | #####9 | 60%\n","libcudf-21.06.01 | 187.7 MB | ######2 | 63%\n","libcudf-21.06.01 | 187.7 MB | ######5 | 65%\n","libcudf-21.06.01 | 187.7 MB | ######7 | 68%\n","libcudf-21.06.01 | 187.7 MB | ####### | 71%\n","libcudf-21.06.01 | 187.7 MB | #######2 | 73%\n","libcudf-21.06.01 | 187.7 MB | #######5 | 75%\n","libcudf-21.06.01 | 187.7 MB | #######8 | 78%\n","libcudf-21.06.01 | 187.7 MB | ######## | 81%\n","libcudf-21.06.01 | 187.7 MB | ########4 | 84%\n","libcudf-21.06.01 | 187.7 MB | ########6 | 87%\n","libcudf-21.06.01 | 187.7 MB | ########9 | 89%\n","libcudf-21.06.01 | 187.7 MB | #########2 | 92%\n","libcudf-21.06.01 | 187.7 MB | #########5 | 95%\n","libcudf-21.06.01 | 187.7 MB | #########8 | 98%\n","libcudf-21.06.01 | 187.7 MB | ########## | 100%\n","\n","llvmlite-0.36.0 | 2.7 MB | | 0%\n","llvmlite-0.36.0 | 2.7 MB | ########## | 100%\n","llvmlite-0.36.0 | 2.7 MB | ########## | 100%\n","\n","hdf4-4.2.15 | 950 KB | | 0%\n","hdf4-4.2.15 | 950 KB | ########## | 100%\n","hdf4-4.2.15 | 950 KB | ########## | 100%\n","\n","olefile-0.46 | 32 KB | | 0%\n","olefile-0.46 | 32 KB | ########## | 100%\n","\n","conda-4.10.1 | 3.1 MB | | 0%\n","conda-4.10.1 | 3.1 MB | ########## | 100%\n","conda-4.10.1 | 3.1 MB | ########## | 100%\n","\n","greenlet-1.1.0 | 83 KB | | 0%\n","greenlet-1.1.0 | 83 KB | ########## | 100%\n","\n","zipp-3.4.1 | 11 KB | | 0%\n","zipp-3.4.1 | 11 KB | ########## | 100%\n","\n","pyppeteer-0.2.2 | 104 KB | | 0%\n","pyppeteer-0.2.2 | 104 KB | ########## | 100%\n","\n","xorg-xproto-7.0.31 | 73 KB | | 0%\n","xorg-xproto-7.0.31 | 73 KB | ########## | 100%\n","\n","numpy-1.20.3 | 5.7 MB | | 0%\n","numpy-1.20.3 | 5.7 MB | ########## | 100%\n","numpy-1.20.3 | 5.7 MB | ########## | 100%\n","\n","parso-0.8.2 | 68 KB | | 0%\n","parso-0.8.2 | 68 KB | ########## | 100%\n","\n","pyparsing-2.4.7 | 60 KB | | 0%\n","pyparsing-2.4.7 | 60 KB | ########## | 100%\n","\n","proj-8.0.0 | 3.1 MB | | 0%\n","proj-8.0.0 | 3.1 MB | ########## | 100%\n","proj-8.0.0 | 3.1 MB | ########## | 100%\n","\n","libthrift-0.14.1 | 4.5 MB | | 0%\n","libthrift-0.14.1 | 4.5 MB | ########## | 100%\n","libthrift-0.14.1 | 4.5 MB | ########## | 100%\n","\n","nodejs-14.15.4 | 15.7 MB | | 0%\n","nodejs-14.15.4 | 15.7 MB | ###8 | 38%\n","nodejs-14.15.4 | 15.7 MB | ########## | 100%\n","nodejs-14.15.4 | 15.7 MB | ########## | 100%\n","\n","libgsasl-1.8.0 | 125 KB | | 0%\n","libgsasl-1.8.0 | 125 KB | ########## | 100%\n","\n","tzdata-2021a | 121 KB | | 0%\n","tzdata-2021a | 121 KB | ########## | 100%\n","\n","pyrsistent-0.17.3 | 89 KB | | 0%\n","pyrsistent-0.17.3 | 89 KB | ########## | 100%\n","\n","yarl-1.6.3 | 141 KB | | 0%\n","yarl-1.6.3 | 141 KB | ########## | 100%\n","\n","pixman-0.40.0 | 627 KB | | 0%\n","pixman-0.40.0 | 627 KB | ########## | 100%\n","pixman-0.40.0 | 627 KB | ########## | 100%\n","\n","param-1.10.1 | 64 KB | | 0%\n","param-1.10.1 | 64 KB | ########## | 100%\n","\n","wcwidth-0.2.5 | 33 KB | | 0%\n","wcwidth-0.2.5 | 33 KB | ########## | 100%\n","\n","cuxfilter-21.06.00 | 136 KB | | 0%\n","cuxfilter-21.06.00 | 136 KB | #1 | 12%\n","cuxfilter-21.06.00 | 136 KB | #####8 | 59%\n","cuxfilter-21.06.00 | 136 KB | ########## | 100%\n","\n","re2-2021.04.01 | 218 KB | | 0%\n","re2-2021.04.01 | 218 KB | ########## | 100%\n","\n","tblib-1.7.0 | 15 KB | | 0%\n","tblib-1.7.0 | 15 KB | ########## | 100%\n","\n","websockets-8.1 | 90 KB | | 0%\n","websockets-8.1 | 90 KB | ########## | 100%\n","\n","cudatoolkit-11.0.221 | 953.0 MB | | 0%\n","cudatoolkit-11.0.221 | 953.0 MB | ########## | 100%\n","cudatoolkit-11.0.221 | 953.0 MB | ########## | 100%\n","\n","sniffio-1.2.0 | 15 KB | | 0%\n","sniffio-1.2.0 | 15 KB | ########## | 100%\n","\n","backports.functools_ | 9 KB | | 0%\n","backports.functools_ | 9 KB | ########## | 100%\n","\n","nbclient-0.5.3 | 67 KB | | 0%\n","nbclient-0.5.3 | 67 KB | ########## | 100%\n","\n","async-timeout-3.0.1 | 11 KB | | 0%\n","async-timeout-3.0.1 | 11 KB | ########## | 100%\n","\n","prompt-toolkit-3.0.1 | 244 KB | | 0%\n","prompt-toolkit-3.0.1 | 244 KB | ########## | 100%\n","\n","grpc-cpp-1.38.0 | 3.6 MB | | 0%\n","grpc-cpp-1.38.0 | 3.6 MB | ########## | 100%\n","grpc-cpp-1.38.0 | 3.6 MB | ########## | 100%\n","\n","fsspec-2021.6.0 | 79 KB | | 0%\n","fsspec-2021.6.0 | 79 KB | ########## | 100%\n","\n","pandocfilters-1.4.2 | 9 KB | | 0%\n","pandocfilters-1.4.2 | 9 KB | ########## | 100%\n","\n","msgpack-python-1.0.2 | 91 KB | | 0%\n","msgpack-python-1.0.2 | 91 KB | ########## | 100%\n","\n","markdown-3.3.4 | 67 KB | | 0%\n","markdown-3.3.4 | 67 KB | ########## | 100%\n","\n","cycler-0.10.0 | 9 KB | | 0%\n","cycler-0.10.0 | 9 KB | ########## | 100%\n","\n","notebook-6.4.0 | 6.1 MB | | 0%\n","notebook-6.4.0 | 6.1 MB | ########## | 100%\n","notebook-6.4.0 | 6.1 MB | ########## | 100%\n","\n","boost-cpp-1.72.0 | 16.3 MB | | 0%\n","boost-cpp-1.72.0 | 16.3 MB | ##3 | 23%\n","boost-cpp-1.72.0 | 16.3 MB | #######1 | 71%\n","boost-cpp-1.72.0 | 16.3 MB | ########## | 100%\n","boost-cpp-1.72.0 | 16.3 MB | ########## | 100%\n","\n","streamz-0.6.2 | 59 KB | | 0%\n","streamz-0.6.2 | 59 KB | ##6 | 27%\n","streamz-0.6.2 | 59 KB | ########## | 100%\n","\n","blazingsql-21.06.00 | 190.2 MB | | 0%\n","blazingsql-21.06.00 | 190.2 MB | | 0%\n","blazingsql-21.06.00 | 190.2 MB | | 0%\n","blazingsql-21.06.00 | 190.2 MB | | 0%\n","blazingsql-21.06.00 | 190.2 MB | | 1%\n","blazingsql-21.06.00 | 190.2 MB | 2 | 2%\n","blazingsql-21.06.00 | 190.2 MB | 4 | 5%\n","blazingsql-21.06.00 | 190.2 MB | 7 | 8%\n","blazingsql-21.06.00 | 190.2 MB | # | 10%\n","blazingsql-21.06.00 | 190.2 MB | #2 | 13%\n","blazingsql-21.06.00 | 190.2 MB | #5 | 16%\n","blazingsql-21.06.00 | 190.2 MB | #8 | 18%\n","blazingsql-21.06.00 | 190.2 MB | ##1 | 21%\n","blazingsql-21.06.00 | 190.2 MB | ##4 | 24%\n","blazingsql-21.06.00 | 190.2 MB | ##7 | 27%\n","blazingsql-21.06.00 | 190.2 MB | ### | 30%\n","blazingsql-21.06.00 | 190.2 MB | ###2 | 33%\n","blazingsql-21.06.00 | 190.2 MB | ###5 | 36%\n","blazingsql-21.06.00 | 190.2 MB | ###8 | 39%\n","blazingsql-21.06.00 | 190.2 MB | ####1 | 41%\n","blazingsql-21.06.00 | 190.2 MB | ####4 | 44%\n","blazingsql-21.06.00 | 190.2 MB | ####7 | 47%\n","blazingsql-21.06.00 | 190.2 MB | ##### | 50%\n","blazingsql-21.06.00 | 190.2 MB | #####2 | 53%\n","blazingsql-21.06.00 | 190.2 MB | #####5 | 55%\n","blazingsql-21.06.00 | 190.2 MB | #####7 | 58%\n","blazingsql-21.06.00 | 190.2 MB | ###### | 60%\n","blazingsql-21.06.00 | 190.2 MB | ######2 | 63%\n","blazingsql-21.06.00 | 190.2 MB | ######5 | 65%\n","blazingsql-21.06.00 | 190.2 MB | ######8 | 68%\n","blazingsql-21.06.00 | 190.2 MB | ####### | 71%\n","blazingsql-21.06.00 | 190.2 MB | #######3 | 73%\n","blazingsql-21.06.00 | 190.2 MB | #######5 | 76%\n","blazingsql-21.06.00 | 190.2 MB | #######8 | 78%\n","blazingsql-21.06.00 | 190.2 MB | ######## | 81%\n","blazingsql-21.06.00 | 190.2 MB | ########3 | 83%\n","blazingsql-21.06.00 | 190.2 MB | ########6 | 86%\n","blazingsql-21.06.00 | 190.2 MB | ########8 | 89%\n","blazingsql-21.06.00 | 190.2 MB | #########1 | 91%\n","blazingsql-21.06.00 | 190.2 MB | #########3 | 94%\n","blazingsql-21.06.00 | 190.2 MB | #########6 | 96%\n","blazingsql-21.06.00 | 190.2 MB | #########8 | 99%\n","blazingsql-21.06.00 | 190.2 MB | ########## | 100%\n","\n","argon2-cffi-20.1.0 | 47 KB | | 0%\n","argon2-cffi-20.1.0 | 47 KB | ########## | 100%\n","\n","backports-1.0 | 4 KB | | 0%\n","backports-1.0 | 4 KB | ########## | 100%\n","\n","cupy-9.0.0 | 50.3 MB | | 0%\n","cupy-9.0.0 | 50.3 MB | 7 | 7%\n","cupy-9.0.0 | 50.3 MB | ##7 | 28%\n","cupy-9.0.0 | 50.3 MB | ####7 | 48%\n","cupy-9.0.0 | 50.3 MB | ####### | 70%\n","cupy-9.0.0 | 50.3 MB | ######### | 90%\n","cupy-9.0.0 | 50.3 MB | ########## | 100%\n","\n","terminado-0.10.1 | 26 KB | | 0%\n","terminado-0.10.1 | 26 KB | ########## | 100%\n","\n","pyct-0.4.6 | 3 KB | | 0%\n","pyct-0.4.6 | 3 KB | ########## | 100%\n","\n","prometheus_client-0. | 46 KB | | 0%\n","prometheus_client-0. | 46 KB | ########## | 100%\n","\n","libuv-1.41.0 | 1.0 MB | | 0%\n","libuv-1.41.0 | 1.0 MB | ########## | 100%\n","libuv-1.41.0 | 1.0 MB | ########## | 100%\n","\n","pyproj-3.0.1 | 484 KB | | 0%\n","pyproj-3.0.1 | 484 KB | ########## | 100%\n","pyproj-3.0.1 | 484 KB | ########## | 100%\n","\n","freexl-1.0.6 | 48 KB | | 0%\n","freexl-1.0.6 | 48 KB | ########## | 100%\n","\n","backcall-0.2.0 | 13 KB | | 0%\n","backcall-0.2.0 | 13 KB | ########## | 100%\n","\n","lcms2-2.12 | 443 KB | | 0%\n","lcms2-2.12 | 443 KB | ########## | 100%\n","lcms2-2.12 | 443 KB | ########## | 100%\n","\n","cairo-1.16.0 | 1.5 MB | | 0%\n","cairo-1.16.0 | 1.5 MB | ########## | 100%\n","cairo-1.16.0 | 1.5 MB | ########## | 100%\n","\n","abseil-cpp-20210324. | 1015 KB | | 0%\n","abseil-cpp-20210324. | 1015 KB | ########## | 100%\n","abseil-cpp-20210324. | 1015 KB | ########## | 100%\n","\n","markupsafe-2.0.1 | 22 KB | | 0%\n","markupsafe-2.0.1 | 22 KB | ########## | 100%\n","\n","pandoc-2.14.0.2 | 12.0 MB | | 0%\n","pandoc-2.14.0.2 | 12.0 MB | ###8 | 39%\n","pandoc-2.14.0.2 | 12.0 MB | ########## | 100%\n","pandoc-2.14.0.2 | 12.0 MB | ########## | 100%\n","\n","libglib-2.68.3 | 3.1 MB | | 0%\n","libglib-2.68.3 | 3.1 MB | ########## | 100%\n","libglib-2.68.3 | 3.1 MB | ########## | 100%\n","\n","spdlog-1.8.5 | 353 KB | | 0%\n","spdlog-1.8.5 | 353 KB | ########## | 100%\n","\n","anyio-3.2.0 | 138 KB | | 0%\n","anyio-3.2.0 | 138 KB | ########## | 100%\n","\n","gcsfs-2021.6.0 | 23 KB | | 0%\n","gcsfs-2021.6.0 | 23 KB | ########## | 100%\n","\n","matplotlib-inline-0. | 11 KB | | 0%\n","matplotlib-inline-0. | 11 KB | ########## | 100%\n","\n","libspatialindex-1.9. | 4.6 MB | | 0%\n","libspatialindex-1.9. | 4.6 MB | ########## | 100%\n","libspatialindex-1.9. | 4.6 MB | ########## | 100%\n","\n","protobuf-3.16.0 | 342 KB | | 0%\n","protobuf-3.16.0 | 342 KB | ########## | 100%\n","protobuf-3.16.0 | 342 KB | ########## | 100%\n","\n","pyjwt-2.1.0 | 17 KB | | 0%\n","pyjwt-2.1.0 | 17 KB | ########## | 100%\n","\n","libntlm-1.4 | 32 KB | | 0%\n","libntlm-1.4 | 32 KB | ########## | 100%\n","\n","libxcb-1.13 | 395 KB | | 0%\n","libxcb-1.13 | 395 KB | ########## | 100%\n","libxcb-1.13 | 395 KB | ########## | 100%\n","\n","cuspatial-21.06.00 | 15.2 MB | | 0%\n","cuspatial-21.06.00 | 15.2 MB | | 0%\n","cuspatial-21.06.00 | 15.2 MB | | 1%\n","cuspatial-21.06.00 | 15.2 MB | 2 | 3%\n","cuspatial-21.06.00 | 15.2 MB | #1 | 11%\n","cuspatial-21.06.00 | 15.2 MB | ###6 | 37%\n","cuspatial-21.06.00 | 15.2 MB | #######2 | 73%\n","cuspatial-21.06.00 | 15.2 MB | ########## | 100%\n","cuspatial-21.06.00 | 15.2 MB | ########## | 100%\n","\n","xorg-renderproto-0.1 | 9 KB | | 0%\n","xorg-renderproto-0.1 | 9 KB | ########## | 100%\n","\n","heapdict-1.0.1 | 7 KB | | 0%\n","heapdict-1.0.1 | 7 KB | ########## | 100%\n","\n","future-0.18.2 | 714 KB | | 0%\n","future-0.18.2 | 714 KB | ########## | 100%\n","future-0.18.2 | 714 KB | ########## | 100%\n","\n","partd-1.2.0 | 18 KB | | 0%\n","partd-1.2.0 | 18 KB | ########## | 100%\n","\n","tzcode-2021a | 68 KB | | 0%\n","tzcode-2021a | 68 KB | ########## | 100%\n","\n","dask-cudf-21.06.01 | 103 KB | | 0%\n","dask-cudf-21.06.01 | 103 KB | #5 | 15%\n","dask-cudf-21.06.01 | 103 KB | ########## | 100%\n","\n","liblapack-3.9.0 | 11 KB | | 0%\n","liblapack-3.9.0 | 11 KB | ########## | 100%\n","\n","kiwisolver-1.3.1 | 78 KB | | 0%\n","kiwisolver-1.3.1 | 78 KB | ########## | 100%\n","\n","zeromq-4.3.4 | 352 KB | | 0%\n","zeromq-4.3.4 | 352 KB | ########## | 100%\n","\n","pyu2f-0.1.5 | 31 KB | | 0%\n","pyu2f-0.1.5 | 31 KB | ########## | 100%\n","\n","python-dateutil-2.8. | 220 KB | | 0%\n","python-dateutil-2.8. | 220 KB | ########## | 100%\n","\n","libcrc32c-1.1.1 | 20 KB | | 0%\n","libcrc32c-1.1.1 | 20 KB | ########## | 100%\n","\n","pygments-2.9.0 | 754 KB | | 0%\n","pygments-2.9.0 | 754 KB | ########## | 100%\n","pygments-2.9.0 | 754 KB | ########## | 100%\n","\n","datashader-0.11.1 | 14.0 MB | | 0%\n","datashader-0.11.1 | 14.0 MB | ###9 | 40%\n","datashader-0.11.1 | 14.0 MB | ########## | 100%\n","datashader-0.11.1 | 14.0 MB | ########## | 100%\n","\n","pyasn1-modules-0.2.7 | 60 KB | | 0%\n","pyasn1-modules-0.2.7 | 60 KB | ########## | 100%\n","\n","jsonschema-3.2.0 | 45 KB | | 0%\n","jsonschema-3.2.0 | 45 KB | ########## | 100%\n","\n","multipledispatch-0.6 | 12 KB | | 0%\n","multipledispatch-0.6 | 12 KB | ########## | 100%\n","\n","libprotobuf-3.16.0 | 2.5 MB | | 0%\n","libprotobuf-3.16.0 | 2.5 MB | ########## | 100%\n","libprotobuf-3.16.0 | 2.5 MB | ########## | 100%\n","\n","toolz-0.11.1 | 46 KB | | 0%\n","toolz-0.11.1 | 46 KB | ########## | 100%\n","\n","zict-2.0.0 | 10 KB | | 0%\n","zict-2.0.0 | 10 KB | ########## | 100%\n","\n","ipykernel-5.5.5 | 167 KB | | 0%\n","ipykernel-5.5.5 | 167 KB | ########## | 100%\n","\n","faiss-proc-1.0.0 | 24 KB | | 0%\n","faiss-proc-1.0.0 | 24 KB | ######7 | 68%\n","faiss-proc-1.0.0 | 24 KB | ########## | 100%\n","\n","libgpg-error-1.42 | 278 KB | | 0%\n","libgpg-error-1.42 | 278 KB | ########## | 100%\n","\n","kealib-1.4.14 | 186 KB | | 0%\n","kealib-1.4.14 | 186 KB | ########## | 100%\n","\n","libtiff-4.2.0 | 639 KB | | 0%\n","libtiff-4.2.0 | 639 KB | ########## | 100%\n","libtiff-4.2.0 | 639 KB | ########## | 100%\n","\n","distributed-2021.5.0 | 1.1 MB | | 0%\n","distributed-2021.5.0 | 1.1 MB | ########## | 100%\n","distributed-2021.5.0 | 1.1 MB | ########## | 100%\n","\n","aws-checksums-0.1.11 | 50 KB | | 0%\n","aws-checksums-0.1.11 | 50 KB | ########## | 100%\n","\n","librttopo-1.1.0 | 235 KB | | 0%\n","librttopo-1.1.0 | 235 KB | ########## | 100%\n","\n","pcre-8.45 | 253 KB | | 0%\n","pcre-8.45 | 253 KB | ########## | 100%\n","\n","fastavro-1.4.1 | 496 KB | | 0%\n","fastavro-1.4.1 | 496 KB | ########## | 100%\n","fastavro-1.4.1 | 496 KB | ########## | 100%\n","\n","pyzmq-22.1.0 | 500 KB | | 0%\n","pyzmq-22.1.0 | 500 KB | ########## | 100%\n","pyzmq-22.1.0 | 500 KB | ########## | 100%\n","\n","aiohttp-3.7.4.post0 | 625 KB | | 0%\n","aiohttp-3.7.4.post0 | 625 KB | ########## | 100%\n","aiohttp-3.7.4.post0 | 625 KB | ########## | 100%\n","\n","brotli-1.0.9 | 389 KB | | 0%\n","brotli-1.0.9 | 389 KB | ########## | 100%\n","\n","dask-cuda-21.06.00 | 110 KB | | 0%\n","dask-cuda-21.06.00 | 110 KB | #4 | 15%\n","dask-cuda-21.06.00 | 110 KB | ########## | 100%\n","\n","rapids-blazing-21.06 | 5 KB | | 0%\n","rapids-blazing-21.06 | 5 KB | ########## | 100%\n","rapids-blazing-21.06 | 5 KB | ########## | 100%\n","\n","arrow-cpp-1.0.1 | 21.1 MB | | 0%\n","arrow-cpp-1.0.1 | 21.1 MB | ##3 | 23%\n","arrow-cpp-1.0.1 | 21.1 MB | #######3 | 74%\n","arrow-cpp-1.0.1 | 21.1 MB | ########## | 100%\n","arrow-cpp-1.0.1 | 21.1 MB | ########## | 100%\n","\n","widgetsnbextension-3 | 1.8 MB | | 0%\n","widgetsnbextension-3 | 1.8 MB | ########## | 100%\n","widgetsnbextension-3 | 1.8 MB | ########## | 100%\n","\n","hdf5-1.10.6 | 3.1 MB | | 0%\n","hdf5-1.10.6 | 3.1 MB | ######9 | 70%\n","hdf5-1.10.6 | 3.1 MB | ########## | 100%\n","hdf5-1.10.6 | 3.1 MB | ########## | 100%\n","\n","rmm-21.06.00 | 7.0 MB | | 0%\n","rmm-21.06.00 | 7.0 MB | | 0%\n","rmm-21.06.00 | 7.0 MB | #######4 | 74%\n","rmm-21.06.00 | 7.0 MB | ########## | 100%\n","rmm-21.06.00 | 7.0 MB | ########## | 100%\n","\n","rapids-21.06.00 | 5 KB | | 0%\n","rapids-21.06.00 | 5 KB | ########## | 100%\n","rapids-21.06.00 | 5 KB | ########## | 100%\n","\n","mapclassify-2.4.2 | 36 KB | | 0%\n","mapclassify-2.4.2 | 36 KB | ########## | 100%\n","\n","cudf-21.06.01 | 108.4 MB | | 0%\n","cudf-21.06.01 | 108.4 MB | | 0%\n","cudf-21.06.01 | 108.4 MB | 4 | 5%\n","cudf-21.06.01 | 108.4 MB | 9 | 9%\n","cudf-21.06.01 | 108.4 MB | #4 | 14%\n","cudf-21.06.01 | 108.4 MB | #9 | 19%\n","cudf-21.06.01 | 108.4 MB | ##4 | 25%\n","cudf-21.06.01 | 108.4 MB | ### | 30%\n","cudf-21.06.01 | 108.4 MB | ###5 | 35%\n","cudf-21.06.01 | 108.4 MB | #### | 41%\n","cudf-21.06.01 | 108.4 MB | ####5 | 46%\n","cudf-21.06.01 | 108.4 MB | #####1 | 51%\n","cudf-21.06.01 | 108.4 MB | #####5 | 55%\n","cudf-21.06.01 | 108.4 MB | #####9 | 59%\n","cudf-21.06.01 | 108.4 MB | ######3 | 63%\n","cudf-21.06.01 | 108.4 MB | ######8 | 68%\n","cudf-21.06.01 | 108.4 MB | #######3 | 73%\n","cudf-21.06.01 | 108.4 MB | #######8 | 78%\n","cudf-21.06.01 | 108.4 MB | ########3 | 84%\n","cudf-21.06.01 | 108.4 MB | ########8 | 89%\n","cudf-21.06.01 | 108.4 MB | #########3 | 94%\n","cudf-21.06.01 | 108.4 MB | #########9 | 99%\n","cudf-21.06.01 | 108.4 MB | ########## | 100%\n","\n","xgboost-1.4.2dev.rap | 17 KB | | 0%\n","xgboost-1.4.2dev.rap | 17 KB | #########5 | 96%\n","xgboost-1.4.2dev.rap | 17 KB | ########## | 100%\n","\n","pexpect-4.8.0 | 47 KB | | 0%\n","pexpect-4.8.0 | 47 KB | ########## | 100%\n","\n","gdal-3.2.2 | 1.5 MB | | 0%\n","gdal-3.2.2 | 1.5 MB | ########## | 100%\n","gdal-3.2.2 | 1.5 MB | ########## | 100%\n","\n","jedi-0.18.0 | 923 KB | | 0%\n","jedi-0.18.0 | 923 KB | ########## | 100%\n","jedi-0.18.0 | 923 KB | ########## | 100%\n","\n","glog-0.5.0 | 104 KB | | 0%\n","glog-0.5.0 | 104 KB | ########## | 100%\n","\n","webencodings-0.5.1 | 12 KB | | 0%\n","webencodings-0.5.1 | 12 KB | ########## | 100%\n","\n","libnetcdf-4.7.4 | 1.3 MB | | 0%\n","libnetcdf-4.7.4 | 1.3 MB | ########## | 100%\n","libnetcdf-4.7.4 | 1.3 MB | ########## | 100%\n","\n","panel-0.10.3 | 6.1 MB | | 0%\n","panel-0.10.3 | 6.1 MB | #######9 | 80%\n","panel-0.10.3 | 6.1 MB | ########## | 100%\n","\n","google-cloud-cpp-1.2 | 9.3 MB | | 0%\n","google-cloud-cpp-1.2 | 9.3 MB | #######1 | 71%\n","google-cloud-cpp-1.2 | 9.3 MB | ########## | 100%\n","google-cloud-cpp-1.2 | 9.3 MB | ########## | 100%\n","\n","networkx-2.5.1 | 1.2 MB | | 0%\n","networkx-2.5.1 | 1.2 MB | ########## | 100%\n","networkx-2.5.1 | 1.2 MB | ########## | 100%\n","\n","psutil-5.8.0 | 342 KB | | 0%\n","psutil-5.8.0 | 342 KB | ########## | 100%\n","psutil-5.8.0 | 342 KB | ########## | 100%\n","\n","ucx-1.9.0+gcd9efd3 | 8.2 MB | | 0%\n","ucx-1.9.0+gcd9efd3 | 8.2 MB | | 0%\n","ucx-1.9.0+gcd9efd3 | 8.2 MB | | 1%\n","ucx-1.9.0+gcd9efd3 | 8.2 MB | 3 | 4%\n","ucx-1.9.0+gcd9efd3 | 8.2 MB | #6 | 16%\n","ucx-1.9.0+gcd9efd3 | 8.2 MB | #####2 | 52%\n","ucx-1.9.0+gcd9efd3 | 8.2 MB | #########9 | 100%\n","ucx-1.9.0+gcd9efd3 | 8.2 MB | ########## | 100%\n","\n","libcurl-7.77.0 | 334 KB | | 0%\n","libcurl-7.77.0 | 334 KB | ########## | 100%\n","\n","libhwloc-2.3.0 | 2.7 MB | | 0%\n","libhwloc-2.3.0 | 2.7 MB | ########## | 100%\n","libhwloc-2.3.0 | 2.7 MB | ########## | 100%\n","\n","xarray-0.18.2 | 599 KB | | 0%\n","xarray-0.18.2 | 599 KB | ########## | 100%\n","xarray-0.18.2 | 599 KB | ########## | 100%\n","\n","google-auth-1.30.2 | 77 KB | | 0%\n","google-auth-1.30.2 | 77 KB | ########## | 100%\n","\n","libcugraph-21.06.00 | 213.6 MB | | 0%\n","libcugraph-21.06.00 | 213.6 MB | | 0%\n","libcugraph-21.06.00 | 213.6 MB | 2 | 2%\n","libcugraph-21.06.00 | 213.6 MB | 4 | 5%\n","libcugraph-21.06.00 | 213.6 MB | 7 | 7%\n","libcugraph-21.06.00 | 213.6 MB | 9 | 10%\n","libcugraph-21.06.00 | 213.6 MB | #2 | 13%\n","libcugraph-21.06.00 | 213.6 MB | #5 | 15%\n","libcugraph-21.06.00 | 213.6 MB | #7 | 18%\n","libcugraph-21.06.00 | 213.6 MB | ## | 21%\n","libcugraph-21.06.00 | 213.6 MB | ##3 | 23%\n","libcugraph-21.06.00 | 213.6 MB | ##5 | 26%\n","libcugraph-21.06.00 | 213.6 MB | ##8 | 29%\n","libcugraph-21.06.00 | 213.6 MB | ###1 | 31%\n","libcugraph-21.06.00 | 213.6 MB | ###3 | 34%\n","libcugraph-21.06.00 | 213.6 MB | ###6 | 37%\n","libcugraph-21.06.00 | 213.6 MB | ###9 | 39%\n","libcugraph-21.06.00 | 213.6 MB | ####2 | 42%\n","libcugraph-21.06.00 | 213.6 MB | ####4 | 45%\n","libcugraph-21.06.00 | 213.6 MB | ####6 | 47%\n","libcugraph-21.06.00 | 213.6 MB | ####9 | 50%\n","libcugraph-21.06.00 | 213.6 MB | #####2 | 52%\n","libcugraph-21.06.00 | 213.6 MB | #####5 | 55%\n","libcugraph-21.06.00 | 213.6 MB | #####7 | 58%\n","libcugraph-21.06.00 | 213.6 MB | ###### | 60%\n","libcugraph-21.06.00 | 213.6 MB | ######2 | 63%\n","libcugraph-21.06.00 | 213.6 MB | ######5 | 66%\n","libcugraph-21.06.00 | 213.6 MB | ######8 | 68%\n","libcugraph-21.06.00 | 213.6 MB | #######1 | 71%\n","libcugraph-21.06.00 | 213.6 MB | #######3 | 74%\n","libcugraph-21.06.00 | 213.6 MB | #######6 | 76%\n","libcugraph-21.06.00 | 213.6 MB | #######9 | 79%\n","libcugraph-21.06.00 | 213.6 MB | ########1 | 82%\n","libcugraph-21.06.00 | 213.6 MB | ########4 | 84%\n","libcugraph-21.06.00 | 213.6 MB | ########6 | 87%\n","libcugraph-21.06.00 | 213.6 MB | ########8 | 89%\n","libcugraph-21.06.00 | 213.6 MB | ######### | 91%\n","libcugraph-21.06.00 | 213.6 MB | #########3 | 93%\n","libcugraph-21.06.00 | 213.6 MB | #########5 | 95%\n","libcugraph-21.06.00 | 213.6 MB | #########7 | 97%\n","libcugraph-21.06.00 | 213.6 MB | #########9 | 100%\n","libcugraph-21.06.00 | 213.6 MB | ########## | 100%\n","\n","google-auth-oauthlib | 19 KB | | 0%\n","google-auth-oauthlib | 19 KB | ########## | 100%\n","\n","pyct-core-0.4.6 | 13 KB | | 0%\n","pyct-core-0.4.6 | 13 KB | ########## | 100%\n","\n","bleach-3.3.0 | 111 KB | | 0%\n","bleach-3.3.0 | 111 KB | ########## | 100%\n","\n","cfitsio-3.470 | 1.3 MB | | 0%\n","cfitsio-3.470 | 1.3 MB | ########## | 100%\n","cfitsio-3.470 | 1.3 MB | ########## | 100%\n","\n","dlpack-0.5 | 12 KB | | 0%\n","dlpack-0.5 | 12 KB | ########## | 100%\n","\n","expat-2.4.1 | 182 KB | | 0%\n","expat-2.4.1 | 182 KB | ########## | 100%\n","\n","aws-c-io-0.10.4 | 121 KB | | 0%\n","aws-c-io-0.10.4 | 121 KB | ########## | 100%\n","\n","munch-2.5.0 | 12 KB | | 0%\n","munch-2.5.0 | 12 KB | ########## | 100%\n","\n","nbformat-5.1.3 | 47 KB | | 0%\n","nbformat-5.1.3 | 47 KB | ########## | 100%\n","\n","postgresql-13.3 | 5.3 MB | | 0%\n","postgresql-13.3 | 5.3 MB | #######5 | 75%\n","postgresql-13.3 | 5.3 MB | ########## | 100%\n","postgresql-13.3 | 5.3 MB | ########## | 100%\n","\n","scikit-learn-0.24.2 | 7.5 MB | | 0%\n","scikit-learn-0.24.2 | 7.5 MB | ####7 | 47%\n","scikit-learn-0.24.2 | 7.5 MB | ########## | 100%\n","scikit-learn-0.24.2 | 7.5 MB | ########## | 100%\n","\n","certifi-2021.5.30 | 141 KB | | 0%\n","certifi-2021.5.30 | 141 KB | ########## | 100%\n","\n","colorcet-2.0.6 | 1.5 MB | | 0%\n","colorcet-2.0.6 | 1.5 MB | ########## | 100%\n","colorcet-2.0.6 | 1.5 MB | ########## | 100%\n","\n","libcblas-3.9.0 | 11 KB | | 0%\n","libcblas-3.9.0 | 11 KB | ########## | 100%\n","\n","ucx-proc-1.0.0 | 9 KB | | 0%\n","ucx-proc-1.0.0 | 9 KB | ########## | 100%\n","ucx-proc-1.0.0 | 9 KB | ########## | 100%\n","\n","ipython-7.24.1 | 1.1 MB | | 0%\n","ipython-7.24.1 | 1.1 MB | ########## | 100%\n","ipython-7.24.1 | 1.1 MB | ########## | 100%\n","\n","libgfortran5-9.3.0 | 2.0 MB | | 0%\n","libgfortran5-9.3.0 | 2.0 MB | ########## | 100%\n","libgfortran5-9.3.0 | 2.0 MB | ########## | 100%\n","\n","cudf_kafka-21.06.01 | 1.7 MB | | 0%\n","cudf_kafka-21.06.01 | 1.7 MB | | 1%\n","cudf_kafka-21.06.01 | 1.7 MB | 4 | 5%\n","cudf_kafka-21.06.01 | 1.7 MB | #5 | 15%\n","cudf_kafka-21.06.01 | 1.7 MB | #####7 | 58%\n","cudf_kafka-21.06.01 | 1.7 MB | ########## | 100%\n","cudf_kafka-21.06.01 | 1.7 MB | ########## | 100%\n","\n","gflags-2.2.2 | 114 KB | | 0%\n","gflags-2.2.2 | 114 KB | ########## | 100%\n","\n","importlib-metadata-4 | 31 KB | | 0%\n","importlib-metadata-4 | 31 KB | ########## | 100%\n","\n","shapely-1.7.1 | 438 KB | | 0%\n","shapely-1.7.1 | 438 KB | 3 | 4%\n","shapely-1.7.1 | 438 KB | ########## | 100%\n","shapely-1.7.1 | 438 KB | ########## | 100%\n","\n","xorg-kbproto-1.0.7 | 27 KB | | 0%\n","xorg-kbproto-1.0.7 | 27 KB | ########## | 100%\n","\n","nlohmann_json-3.9.1 | 122 KB | | 0%\n","nlohmann_json-3.9.1 | 122 KB | ########## | 100%\n","\n","cloudpickle-1.6.0 | 22 KB | | 0%\n","cloudpickle-1.6.0 | 22 KB | ########## | 100%\n","\n","libuuid-2.32.1 | 28 KB | | 0%\n","libuuid-2.32.1 | 28 KB | ########## | 100%\n","\n","libkml-1.3.0 | 640 KB | | 0%\n","libkml-1.3.0 | 640 KB | ########## | 100%\n","libkml-1.3.0 | 640 KB | ########## | 100%\n","\n","click-plugins-1.1.1 | 9 KB | | 0%\n","click-plugins-1.1.1 | 9 KB | ########## | 100%\n","\n","ipywidgets-7.6.3 | 101 KB | | 0%\n","ipywidgets-7.6.3 | 101 KB | ########## | 100%\n","\n","cusignal-21.06.00 | 1.0 MB | | 0%\n","cusignal-21.06.00 | 1.0 MB | 1 | 2%\n","cusignal-21.06.00 | 1.0 MB | ########## | 100%\n","cusignal-21.06.00 | 1.0 MB | ########## | 100%\n","\n","aws-c-common-0.6.2 | 168 KB | | 0%\n","aws-c-common-0.6.2 | 168 KB | ########## | 100%\n","\n","openssl-1.1.1k | 2.1 MB | | 0%\n","openssl-1.1.1k | 2.1 MB | ########## | 100%\n","openssl-1.1.1k | 2.1 MB | ########## | 100%\n","\n","blinker-1.4 | 13 KB | | 0%\n","blinker-1.4 | 13 KB | ########## | 100%\n","\n","libpq-13.3 | 2.7 MB | | 0%\n","libpq-13.3 | 2.7 MB | ########## | 100%\n","libpq-13.3 | 2.7 MB | ########## | 100%\n","\n","click-7.1.2 | 64 KB | | 0%\n","click-7.1.2 | 64 KB | ########## | 100%\n","\n","thrift_sasl-0.4.2 | 14 KB | | 0%\n","thrift_sasl-0.4.2 | 14 KB | ########## | 100%\n","\n","requests-oauthlib-1. | 21 KB | | 0%\n","requests-oauthlib-1. | 21 KB | ########## | 100%\n","\n","libblas-3.9.0 | 11 KB | | 0%\n","libblas-3.9.0 | 11 KB | ########## | 100%\n","\n","librdkafka-1.5.3 | 11.2 MB | | 0%\n","librdkafka-1.5.3 | 11.2 MB | ###8 | 39%\n","librdkafka-1.5.3 | 11.2 MB | ########## | 100%\n","librdkafka-1.5.3 | 11.2 MB | ########## | 100%\n","\n","geopandas-base-0.9.0 | 950 KB | | 0%\n","geopandas-base-0.9.0 | 950 KB | ########## | 100%\n","geopandas-base-0.9.0 | 950 KB | ########## | 100%\n","\n","libgcrypt-1.9.3 | 677 KB | | 0%\n","libgcrypt-1.9.3 | 677 KB | ########## | 100%\n","libgcrypt-1.9.3 | 677 KB | ########## | 100%\n","\n","geos-3.9.1 | 1.1 MB | | 0%\n","geos-3.9.1 | 1.1 MB | ########## | 100%\n","geos-3.9.1 | 1.1 MB | ########## | 100%\n","\n","xorg-libx11-1.7.2 | 941 KB | | 0%\n","xorg-libx11-1.7.2 | 941 KB | ########## | 100%\n","xorg-libx11-1.7.2 | 941 KB | ########## | 100%\n","\n","cytoolz-0.11.0 | 403 KB | | 0%\n","cytoolz-0.11.0 | 403 KB | ########## | 100%\n","cytoolz-0.11.0 | 403 KB | ########## | 100%\n","\n","libdap4-3.20.6 | 11.3 MB | | 0%\n","libdap4-3.20.6 | 11.3 MB | ####8 | 49%\n","libdap4-3.20.6 | 11.3 MB | ########## | 100%\n","libdap4-3.20.6 | 11.3 MB | ########## | 100%\n","\n","libspatialite-5.0.1 | 4.4 MB | | 0%\n","libspatialite-5.0.1 | 4.4 MB | ########## | 100%\n","libspatialite-5.0.1 | 4.4 MB | ########## | 100%\n","\n","py-xgboost-1.4.2dev. | 151 KB | | 0%\n","py-xgboost-1.4.2dev. | 151 KB | # | 11%\n","py-xgboost-1.4.2dev. | 151 KB | ######3 | 64%\n","py-xgboost-1.4.2dev. | 151 KB | ########## | 100%\n","\n","jupyter_server-1.8.0 | 255 KB | | 0%\n","jupyter_server-1.8.0 | 255 KB | ########## | 100%\n","jupyter_server-1.8.0 | 255 KB | ########## | 100%\n","\n","poppler-data-0.4.10 | 3.8 MB | | 0%\n","poppler-data-0.4.10 | 3.8 MB | ########## | 100%\n","poppler-data-0.4.10 | 3.8 MB | ########## | 100%\n","\n","libopenblas-0.3.15 | 9.2 MB | | 0%\n","libopenblas-0.3.15 | 9.2 MB | #####7 | 57%\n","libopenblas-0.3.15 | 9.2 MB | ########## | 100%\n","libopenblas-0.3.15 | 9.2 MB | ########## | 100%\n","\n","pyarrow-1.0.1 | 2.4 MB | | 0%\n","pyarrow-1.0.1 | 2.4 MB | ########## | 100%\n","pyarrow-1.0.1 | 2.4 MB | ########## | 100%\n","\n","threadpoolctl-2.1.0 | 15 KB | | 0%\n","threadpoolctl-2.1.0 | 15 KB | ########## | 100%\n","\n","fastrlock-0.6 | 31 KB | | 0%\n","fastrlock-0.6 | 31 KB | ########## | 100%\n","\n","xorg-libxau-1.0.9 | 13 KB | | 0%\n","xorg-libxau-1.0.9 | 13 KB | ########## | 100%\n","\n","attrs-21.2.0 | 44 KB | | 0%\n","attrs-21.2.0 | 44 KB | ########## | 100%\n","\n","libfaiss-1.7.0 | 67.0 MB | | 0%\n","libfaiss-1.7.0 | 67.0 MB | 4 | 5%\n","libfaiss-1.7.0 | 67.0 MB | #9 | 19%\n","libfaiss-1.7.0 | 67.0 MB | ###7 | 37%\n","libfaiss-1.7.0 | 67.0 MB | #####4 | 55%\n","libfaiss-1.7.0 | 67.0 MB | #######2 | 73%\n","libfaiss-1.7.0 | 67.0 MB | ########6 | 86%\n","libfaiss-1.7.0 | 67.0 MB | ########## | 100%\n","libfaiss-1.7.0 | 67.0 MB | ########## | 100%\n","\n","librmm-21.06.00 | 57 KB | | 0%\n","librmm-21.06.00 | 57 KB | ##8 | 28%\n","librmm-21.06.00 | 57 KB | ########## | 100%\n","\n","send2trash-1.5.0 | 12 KB | | 0%\n","send2trash-1.5.0 | 12 KB | ########## | 100%\n","\n","pickle5-0.0.11 | 173 KB | | 0%\n","pickle5-0.0.11 | 173 KB | ########## | 100%\n","\n","pyasn1-0.4.8 | 53 KB | | 0%\n","pyasn1-0.4.8 | 53 KB | ########## | 100%\n","\n","pandas-1.2.4 | 11.8 MB | | 0%\n","pandas-1.2.4 | 11.8 MB | ###### | 61%\n","pandas-1.2.4 | 11.8 MB | ########## | 100%\n","pandas-1.2.4 | 11.8 MB | ########## | 100%\n","\n","aws-c-cal-0.5.11 | 37 KB | | 0%\n","aws-c-cal-0.5.11 | 37 KB | ########## | 100%\n","\n","libcuml-21.06.02 | 95.2 MB | | 0%\n","libcuml-21.06.02 | 95.2 MB | | 0%\n","libcuml-21.06.02 | 95.2 MB | | 0%\n","libcuml-21.06.02 | 95.2 MB | 1 | 1%\n","libcuml-21.06.02 | 95.2 MB | 3 | 4%\n","libcuml-21.06.02 | 95.2 MB | 7 | 8%\n","libcuml-21.06.02 | 95.2 MB | #2 | 12%\n","libcuml-21.06.02 | 95.2 MB | #7 | 17%\n","libcuml-21.06.02 | 95.2 MB | ##1 | 22%\n","libcuml-21.06.02 | 95.2 MB | ##7 | 27%\n","libcuml-21.06.02 | 95.2 MB | ###2 | 32%\n","libcuml-21.06.02 | 95.2 MB | ###7 | 38%\n","libcuml-21.06.02 | 95.2 MB | ####3 | 43%\n","libcuml-21.06.02 | 95.2 MB | ####8 | 49%\n","libcuml-21.06.02 | 95.2 MB | #####4 | 54%\n","libcuml-21.06.02 | 95.2 MB | ###### | 60%\n","libcuml-21.06.02 | 95.2 MB | ######5 | 65%\n","libcuml-21.06.02 | 95.2 MB | #######1 | 71%\n","libcuml-21.06.02 | 95.2 MB | #######6 | 77%\n","libcuml-21.06.02 | 95.2 MB | ########2 | 82%\n","libcuml-21.06.02 | 95.2 MB | ########8 | 88%\n","libcuml-21.06.02 | 95.2 MB | #########4 | 94%\n","libcuml-21.06.02 | 95.2 MB | ########## | 100%\n","libcuml-21.06.02 | 95.2 MB | ########## | 100%\n","\n","libcumlprims-21.06.0 | 1.1 MB | | 0%\n","libcumlprims-21.06.0 | 1.1 MB | ########## | 100%\n","libcumlprims-21.06.0 | 1.1 MB | ########## | 100%\n","\n","ucx-py-0.20.0 | 294 KB | | 0%\n","ucx-py-0.20.0 | 294 KB | 5 | 5%\n","ucx-py-0.20.0 | 294 KB | ###2 | 33%\n","ucx-py-0.20.0 | 294 KB | ########## | 100%\n","ucx-py-0.20.0 | 294 KB | ########## | 100%\n","\n","matplotlib-base-3.4. | 7.2 MB | | 0%\n","matplotlib-base-3.4. | 7.2 MB | ######6 | 66%\n","matplotlib-base-3.4. | 7.2 MB | ########## | 100%\n","matplotlib-base-3.4. | 7.2 MB | ########## | 100%\n","\n","nest-asyncio-1.5.1 | 9 KB | | 0%\n","nest-asyncio-1.5.1 | 9 KB | ########## | 100%\n","\n","jupyterlab_widgets-1 | 130 KB | | 0%\n","jupyterlab_widgets-1 | 130 KB | ########## | 100%\n","\n","rapids-xgboost-21.06 | 4 KB | | 0%\n","rapids-xgboost-21.06 | 4 KB | ########## | 100%\n","rapids-xgboost-21.06 | 4 KB | ########## | 100%\n","\n","gettext-0.19.8.1 | 3.6 MB | | 0%\n","gettext-0.19.8.1 | 3.6 MB | #####1 | 51%\n","gettext-0.19.8.1 | 3.6 MB | ########## | 100%\n","gettext-0.19.8.1 | 3.6 MB | ########## | 100%\n","\n","xorg-libice-1.0.10 | 58 KB | | 0%\n","xorg-libice-1.0.10 | 58 KB | ########## | 100%\n","\n","ipython_genutils-0.2 | 21 KB | | 0%\n","ipython_genutils-0.2 | 21 KB | ########## | 100%\n","\n","libpng-1.6.37 | 306 KB | | 0%\n","libpng-1.6.37 | 306 KB | ########## | 100%\n","\n","xorg-libsm-1.2.3 | 26 KB | | 0%\n","xorg-libsm-1.2.3 | 26 KB | ########## | 100%\n","\n","orc-1.6.7 | 751 KB | | 0%\n","orc-1.6.7 | 751 KB | ########## | 100%\n","orc-1.6.7 | 751 KB | ########## | 100%\n","\n","cyrus-sasl-2.1.27 | 224 KB | | 0%\n","cyrus-sasl-2.1.27 | 224 KB | ########## | 100%\n","\n","mistune-0.8.4 | 54 KB | | 0%\n","mistune-0.8.4 | 54 KB | ########## | 100%\n","\n","multidict-5.1.0 | 67 KB | | 0%\n","multidict-5.1.0 | 67 KB | ########## | 100%\n","\n","sqlalchemy-1.4.18 | 2.2 MB | | 0%\n","sqlalchemy-1.4.18 | 2.2 MB | ########## | 100%\n","sqlalchemy-1.4.18 | 2.2 MB | ########## | 100%\n","\n","ca-certificates-2021 | 136 KB | | 0%\n","ca-certificates-2021 | 136 KB | ########## | 100%\n","\n","openjpeg-2.4.0 | 444 KB | | 0%\n","openjpeg-2.4.0 | 444 KB | ########## | 100%\n","openjpeg-2.4.0 | 444 KB | ########## | 100%\n","\n","ptyprocess-0.7.0 | 16 KB | | 0%\n","ptyprocess-0.7.0 | 16 KB | ########## | 100%\n","\n","packaging-20.9 | 35 KB | | 0%\n","packaging-20.9 | 35 KB | ########## | 100%\n","\n","cachetools-4.2.2 | 12 KB | | 0%\n","cachetools-4.2.2 | 12 KB | ########## | 100%\n","\n","pyviz_comms-2.0.2 | 25 KB | | 0%\n","pyviz_comms-2.0.2 | 25 KB | ########## | 100%\n","\n","pydeck-0.5.0 | 3.6 MB | | 0%\n","pydeck-0.5.0 | 3.6 MB | ########## | 100%\n","pydeck-0.5.0 | 3.6 MB | ########## | 100%\n","\n","locket-0.2.0 | 6 KB | | 0%\n","locket-0.2.0 | 6 KB | ########## | 100%\n","\n","libgdal-3.2.2 | 13.2 MB | | 0%\n","libgdal-3.2.2 | 13.2 MB | ##3 | 23%\n","libgdal-3.2.2 | 13.2 MB | ###### | 61%\n","libgdal-3.2.2 | 13.2 MB | #########6 | 97%\n","libgdal-3.2.2 | 13.2 MB | ########## | 100%\n","\n","dask-core-2021.5.0 | 735 KB | | 0%\n","dask-core-2021.5.0 | 735 KB | ########## | 100%\n","dask-core-2021.5.0 | 735 KB | ########## | 100%\n","\n","typing_extensions-3. | 28 KB | | 0%\n","typing_extensions-3. | 28 KB | ########## | 100%\n","\n","tornado-6.1 | 646 KB | | 0%\n","tornado-6.1 | 646 KB | ########## | 100%\n","tornado-6.1 | 646 KB | ########## | 100%\n","\n","xorg-libxrender-0.9. | 32 KB | | 0%\n","xorg-libxrender-0.9. | 32 KB | ########## | 100%\n","\n","pthread-stubs-0.4 | 5 KB | | 0%\n","pthread-stubs-0.4 | 5 KB | ########## | 100%\n","\n","snappy-1.1.8 | 32 KB | | 0%\n","snappy-1.1.8 | 32 KB | ########## | 100%\n","\n","entrypoints-0.3 | 8 KB | | 0%\n","entrypoints-0.3 | 8 KB | ########## | 100%\n","\n","traitlets-5.0.5 | 81 KB | | 0%\n","traitlets-5.0.5 | 81 KB | ########## | 100%\n","\n","treelite-1.3.0 | 2.7 MB | | 0%\n","treelite-1.3.0 | 2.7 MB | ########## | 100%\n","treelite-1.3.0 | 2.7 MB | ########## | 100%\n","\n","aws-c-event-stream-0 | 47 KB | | 0%\n","aws-c-event-stream-0 | 47 KB | ########## | 100%\n","\n","jpeg-9d | 264 KB | | 0%\n","jpeg-9d | 264 KB | ########## | 100%\n","\n","async_generator-1.10 | 18 KB | | 0%\n","async_generator-1.10 | 18 KB | ########## | 100%\n","\n","bokeh-2.2.3 | 7.0 MB | | 0%\n","bokeh-2.2.3 | 7.0 MB | ##### | 50%\n","bokeh-2.2.3 | 7.0 MB | ########## | 100%\n","bokeh-2.2.3 | 7.0 MB | ########## | 100%\n","\n","jupyter_client-6.1.1 | 79 KB | | 0%\n","jupyter_client-6.1.1 | 79 KB | ########## | 100%\n","\n","xorg-xextproto-7.3.0 | 28 KB | | 0%\n","xorg-xextproto-7.3.0 | 28 KB | ########## | 100%\n","\n","pyyaml-5.4.1 | 189 KB | | 0%\n","pyyaml-5.4.1 | 189 KB | ########## | 100%\n","\n","libcuspatial-21.06.0 | 7.6 MB | | 0%\n","libcuspatial-21.06.0 | 7.6 MB | | 0%\n","libcuspatial-21.06.0 | 7.6 MB | 1 | 1%\n","libcuspatial-21.06.0 | 7.6 MB | 5 | 5%\n","libcuspatial-21.06.0 | 7.6 MB | ## | 21%\n","libcuspatial-21.06.0 | 7.6 MB | ######4 | 64%\n","libcuspatial-21.06.0 | 7.6 MB | ########## | 100%\n","libcuspatial-21.06.0 | 7.6 MB | ########## | 100%\n","\n","jupyter-server-proxy | 27 KB | | 0%\n","jupyter-server-proxy | 27 KB | ########## | 100%\n","\n","pickleshare-0.7.5 | 9 KB | | 0%\n","pickleshare-0.7.5 | 9 KB | ########## | 100%\n","\n","pytz-2021.1 | 239 KB | | 0%\n","pytz-2021.1 | 239 KB | ########## | 100%\n","pytz-2021.1 | 239 KB | ########## | 100%\n","\n","libsodium-1.0.18 | 366 KB | | 0%\n","libsodium-1.0.18 | 366 KB | ########## | 100%\n","\n","xorg-libxext-1.3.4 | 54 KB | | 0%\n","xorg-libxext-1.3.4 | 54 KB | ########## | 100%\n","\n","s2n-1.0.10 | 442 KB | | 0%\n","s2n-1.0.10 | 442 KB | ########## | 100%\n","s2n-1.0.10 | 442 KB | ########## | 100%\n","\n","oauthlib-3.1.1 | 87 KB | | 0%\n","oauthlib-3.1.1 | 87 KB | ########## | 100%\n","\n","typing-extensions-3. | 8 KB | | 0%\n","typing-extensions-3. | 8 KB | ########## | 100%\n","\n","defusedxml-0.7.1 | 23 KB | | 0%\n","defusedxml-0.7.1 | 23 KB | ########## | 100%\n","\n","cuml-21.06.02 | 78.9 MB | | 0%\n","cuml-21.06.02 | 78.9 MB | | 0%\n","cuml-21.06.02 | 78.9 MB | 6 | 7%\n","cuml-21.06.02 | 78.9 MB | #3 | 13%\n","cuml-21.06.02 | 78.9 MB | #9 | 19%\n","cuml-21.06.02 | 78.9 MB | ##6 | 26%\n","cuml-21.06.02 | 78.9 MB | ###1 | 31%\n","cuml-21.06.02 | 78.9 MB | ###7 | 38%\n","cuml-21.06.02 | 78.9 MB | ####4 | 44%\n","cuml-21.06.02 | 78.9 MB | ####9 | 50%\n","cuml-21.06.02 | 78.9 MB | #####6 | 57%\n","cuml-21.06.02 | 78.9 MB | ######2 | 62%\n","cuml-21.06.02 | 78.9 MB | ######8 | 68%\n","cuml-21.06.02 | 78.9 MB | #######4 | 75%\n","cuml-21.06.02 | 78.9 MB | ######## | 81%\n","cuml-21.06.02 | 78.9 MB | ########6 | 86%\n","cuml-21.06.02 | 78.9 MB | #########2 | 92%\n","cuml-21.06.02 | 78.9 MB | #########8 | 98%\n","cuml-21.06.02 | 78.9 MB | ########## | 100%\n","\n","jupyter_core-4.7.1 | 72 KB | | 0%\n","jupyter_core-4.7.1 | 72 KB | ########## | 100%\n","\n","xerces-c-3.2.3 | 1.8 MB | | 0%\n","xerces-c-3.2.3 | 1.8 MB | ########## | 100%\n","xerces-c-3.2.3 | 1.8 MB | ########## | 100%\n","\n","appdirs-1.4.4 | 13 KB | | 0%\n","appdirs-1.4.4 | 13 KB | ########## | 100%\n","\n","jinja2-3.0.1 | 99 KB | | 0%\n","jinja2-3.0.1 | 99 KB | ########## | 100%\n","\n","curl-7.77.0 | 149 KB | | 0%\n","curl-7.77.0 | 149 KB | ########## | 100%\n","\n","jupyterlab_pygments- | 8 KB | | 0%\n","jupyterlab_pygments- | 8 KB | ########## | 100%\n","\n","simpervisor-0.4 | 9 KB | | 0%\n","simpervisor-0.4 | 9 KB | ########## | 100%\n","\n","tiledb-2.2.9 | 4.0 MB | | 0%\n","tiledb-2.2.9 | 4.0 MB | #######7 | 78%\n","tiledb-2.2.9 | 4.0 MB | ########## | 100%\n","\n","libevent-2.1.10 | 1.1 MB | | 0%\n","libevent-2.1.10 | 1.1 MB | ########## | 100%\n","libevent-2.1.10 | 1.1 MB | ########## | 100%\n","\n","giflib-5.2.1 | 77 KB | | 0%\n","giflib-5.2.1 | 77 KB | ########## | 100%\n","\n","nbconvert-6.0.7 | 535 KB | | 0%\n","nbconvert-6.0.7 | 535 KB | ########## | 100%\n","nbconvert-6.0.7 | 535 KB | ########## | 100%\n","\n","pyee-7.0.4 | 14 KB | | 0%\n","pyee-7.0.4 | 14 KB | ########## | 100%\n","\n","testpath-0.5.0 | 86 KB | | 0%\n","testpath-0.5.0 | 86 KB | ########## | 100%\n","\n","xorg-libxdmcp-1.1.3 | 19 KB | | 0%\n","xorg-libxdmcp-1.1.3 | 19 KB | ########## | 100%\n","\n","pyhive-0.6.4 | 39 KB | | 0%\n","pyhive-0.6.4 | 39 KB | ########## | 100%\n","\n","nccl-2.9.9.1 | 82.3 MB | | 0%\n","nccl-2.9.9.1 | 82.3 MB | 3 | 4%\n","nccl-2.9.9.1 | 82.3 MB | 8 | 9%\n","nccl-2.9.9.1 | 82.3 MB | #8 | 18%\n","nccl-2.9.9.1 | 82.3 MB | ##9 | 29%\n","nccl-2.9.9.1 | 82.3 MB | ####2 | 42%\n","nccl-2.9.9.1 | 82.3 MB | #####3 | 54%\n","nccl-2.9.9.1 | 82.3 MB | ######3 | 63%\n","nccl-2.9.9.1 | 82.3 MB | #######4 | 74%\n","nccl-2.9.9.1 | 82.3 MB | ########8 | 88%\n","nccl-2.9.9.1 | 82.3 MB | #########9 | 100%\n","nccl-2.9.9.1 | 82.3 MB | ########## | 100%\n","\n","websocket-client-0.5 | 59 KB | | 0%\n","websocket-client-0.5 | 59 KB | ########## | 100%\n","\n","openjdk-8.0.282 | 99.3 MB | | 0%\n","openjdk-8.0.282 | 99.3 MB | 7 | 7%\n","openjdk-8.0.282 | 99.3 MB | #8 | 18%\n","openjdk-8.0.282 | 99.3 MB | ##9 | 30%\n","openjdk-8.0.282 | 99.3 MB | ####2 | 42%\n","openjdk-8.0.282 | 99.3 MB | #####2 | 52%\n","openjdk-8.0.282 | 99.3 MB | ######1 | 61%\n","openjdk-8.0.282 | 99.3 MB | ####### | 70%\n","openjdk-8.0.282 | 99.3 MB | ######## | 81%\n","openjdk-8.0.282 | 99.3 MB | #########2 | 92%\n","openjdk-8.0.282 | 99.3 MB | ########## | 100%\n","\n","libutf8proc-2.6.1 | 95 KB | | 0%\n","libutf8proc-2.6.1 | 95 KB | ########## | 100%\n","\n","rsa-4.7.2 | 28 KB | | 0%\n","rsa-4.7.2 | 28 KB | ########## | 100%\n","\n","datashape-0.5.4 | 49 KB | | 0%\n","datashape-0.5.4 | 49 KB | ########## | 100%\n","\n","joblib-1.0.1 | 206 KB | | 0%\n","joblib-1.0.1 | 206 KB | ########## | 100%\n","\n","libwebp-base-1.2.0 | 815 KB | | 0%\n","libwebp-base-1.2.0 | 815 KB | ########## | 100%\n","libwebp-base-1.2.0 | 815 KB | ########## | 100%\n","\n","scipy-1.6.3 | 20.5 MB | | 0%\n","scipy-1.6.3 | 20.5 MB | ##7 | 27%\n","scipy-1.6.3 | 20.5 MB | #######6 | 77%\n","scipy-1.6.3 | 20.5 MB | ########## | 100%\n","scipy-1.6.3 | 20.5 MB | ########## | 100%\n","\n","freetype-2.10.4 | 890 KB | | 0%\n","freetype-2.10.4 | 890 KB | ########## | 100%\n","freetype-2.10.4 | 890 KB | ########## | 100%\n","\n","pillow-8.2.0 | 684 KB | | 0%\n","pillow-8.2.0 | 684 KB | ########## | 100%\n","pillow-8.2.0 | 684 KB | ########## | 100%\n","\n","json-c-0.15 | 274 KB | | 0%\n","json-c-0.15 | 274 KB | ########## | 100%\n","\n","python-confluent-kaf | 122 KB | | 0%\n","python-confluent-kaf | 122 KB | ########## | 100%\n","\n","readline-8.1 | 295 KB | | 0%\n","readline-8.1 | 295 KB | ########## | 100%\n","\n","jpype1-1.3.0 | 482 KB | | 0%\n","jpype1-1.3.0 | 482 KB | ########## | 100%\n","jpype1-1.3.0 | 482 KB | ########## | 100%\n","\n","rtree-0.9.7 | 45 KB | | 0%\n","rtree-0.9.7 | 45 KB | ########## | 100%\n","\n","libwebp-1.2.0 | 85 KB | | 0%\n","libwebp-1.2.0 | 85 KB | ########## | 100%\n","Preparing transaction: ...working... done\n","Verifying transaction: ...working... done\n","Executing transaction: ...working... By downloading and using the CUDA Toolkit conda packages, you accept the terms and conditions of the CUDA End User License Agreement (EULA): https://docs.nvidia.com/cuda/eula/index.html\n","\n","Enabling notebook extension jupyter-js-widgets/extension...\n","Paths used for configuration of notebook:\n"," \t/usr/local/etc/jupyter/nbconfig/notebook.d/plotlywidget.json\n"," \t/usr/local/etc/jupyter/nbconfig/notebook.d/pydeck.json\n"," \t/usr/local/etc/jupyter/nbconfig/notebook.d/widgetsnbextension.json\n"," \t/usr/local/etc/jupyter/nbconfig/notebook.json\n","Paths used for configuration of notebook:\n"," \t/usr/local/etc/jupyter/nbconfig/notebook.d/plotlywidget.json\n"," \t/usr/local/etc/jupyter/nbconfig/notebook.d/pydeck.json\n"," \t/usr/local/etc/jupyter/nbconfig/notebook.d/widgetsnbextension.json\n"," - Validating: \u001b[32mOK\u001b[0m\n","Paths used for configuration of notebook:\n"," \t/usr/local/etc/jupyter/nbconfig/notebook.d/plotlywidget.json\n"," \t/usr/local/etc/jupyter/nbconfig/notebook.d/pydeck.json\n"," \t/usr/local/etc/jupyter/nbconfig/notebook.d/widgetsnbextension.json\n"," \t/usr/local/etc/jupyter/nbconfig/notebook.json\n","\n","done\n","RAPIDS conda installation complete. Updating Colab's libraries...\n","Copying /usr/local/lib/libcudf.so to /usr/lib/libcudf.so\n","Copying /usr/local/lib/libnccl.so to /usr/lib/libnccl.so\n","Copying /usr/local/lib/libcuml.so to /usr/lib/libcuml.so\n","Copying /usr/local/lib/libcugraph.so to /usr/lib/libcugraph.so\n","Copying /usr/local/lib/libxgboost.so to /usr/lib/libxgboost.so\n","Copying /usr/local/lib/libcuspatial.so to /usr/lib/libcuspatial.so\n","Copying /usr/local/lib/libgeos.so to /usr/lib/libgeos.so\n"],"name":"stdout"}]},{"cell_type":"markdown","metadata":{"id":"Q8E5moaL3pb6"},"source":["### Extra: Adding cat columns to datasets.\n","It is possible with Cudf. Giving memory error with Pandas."]},{"cell_type":"code","metadata":{"id":"pN-Foz-x2-uO"},"source":["# import IPython\n","\n","# import pandas as pd\n","# import cudf\n","# import numpy as np\n","# import cupy\n","# import matplotlib.pyplot as plt\n","\n","# !cp /content/drive/MyDrive/Recommendation/data_silver_l2.zip /content\n","# !unzip /content/data_silver_l2.zip\n","\n","# df_train = cudf.read_parquet('/content/train.parquet')\n","# df_valid = cudf.read_parquet('/content/valid.parquet')\n","# df_test = cudf.read_parquet('/content/test.parquet')\n","\n","# df_train.isna().sum()\n","\n","# _temp = df_train['category_code'].str.split(\".\", n=3, expand=True).fillna('NA')\n","# _temp.columns = ['cat_{}'.format(x) for x in _temp.columns]\n","# df_train.drop('category_code', axis=1, inplace=True)\n","# df_train = df_train.join(_temp)\n","\n","# _temp = df_valid['category_code'].str.split(\".\", n=3, expand=True).fillna('NA')\n","# _temp.columns = ['cat_{}'.format(x) for x in _temp.columns]\n","# df_valid.drop('category_code', axis=1, inplace=True)\n","# df_valid = df_valid.join(_temp)\n","\n","# _temp = df_test['category_code'].str.split(\".\", n=3, expand=True).fillna('NA')\n","# _temp.columns = ['cat_{}'.format(x) for x in _temp.columns]\n","# df_test.drop('category_code', axis=1, inplace=True)\n","# df_test = df_test.join(_temp)\n","\n","# !mkdir -p /content/data/silver_l3\n","# df_train.to_parquet('/content/data/silver_l3/train.parquet', index=False)\n","# df_valid.to_parquet('/content/data/silver_l3/valid.parquet', index=False)\n","# df_test.to_parquet('/content/data/silver_l3/test.parquet', index=False)\n","\n","# !cd /content/data/silver_l3 && zip /content/data_silver_l3.zip ./*.parquet\n","# !cp /content/data_silver_l3.zip /content/drive/MyDrive/Recommendation"],"execution_count":null,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"ADjBlJtF55_Z"},"source":["### Intro to NVTabular"]},{"cell_type":"markdown","metadata":{"id":"b7ibvy1P59rN"},"source":["With the rapid growth in scale of industry datasets, deep learning (DL) recommender models have started to gain advantages over traditional methods by capitalizing on large amounts of training data.\n","\n","The current challenges for training large-scale recommenders include:\n","- Huge datasets: Commercial recommenders are trained on huge datasets, often several terabytes in scale.\n","- Complex data preprocessing and feature engineering pipelines: Datasets need to be preprocessed and transformed into a form relevant to be used with DL models and frameworks. In addition, feature engineering creates an extensive set of new features from existing ones, requiring multiple iterations to arrive at an optimal solution.\n","- Input bottleneck: Data loading, if not well optimized, can be the slowest part of the training process, leading to under-utilization of high-throughput computing devices such as GPUs.\n","- Extensive repeated experimentation: The whole data engineering, training, and evaluation process is generally repeated many times, requiring significant time and computational resources.\n","\n","NVTabular is a library for fast tabular data tranformation and loading, manipulating terabyte-scale datasets quickly. It provides best practices for feature engineering and preprocessing and a high-level abstraction to simplify code accelerating computation on the GPU using the RAPIDS cuDF library."]},{"cell_type":"markdown","metadata":{"id":"_BASJ76x6EbX"},"source":["![image.png](data:image/png;base64,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)"]},{"cell_type":"markdown","metadata":{"id":"yN03XPiv6Tte"},"source":["NVTabular has 4 main components:\n","\n","1. Dataset: A dataset contains a list of files and iterates over the files. If necessary, it will read a file in chunks.\n","2. Op: An Op defines the calculation, which should be exectued. For example, an op could be to collect the mean/std for a column, fill in missing values or combine two categories.\n","3. Workflow: A workflow orchastrates the pipeline\n"," - It defines the contex, which columns are categorical, numerical or the label\n"," - It registers the operations (calculation) for the different column types\n"," - It optimizes the tasks by reordering the operations\n"," - It collects the required statistics for operations (e.g. the mean/std for normalization)\n"," - It applies the final operations to the dataset\n","4. Dataloader: NVTabular provides optimized dataloader for tabular data in PyTorch and Tensorflow"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"SSTQdoaX6qwq","executionInfo":{"status":"ok","timestamp":1624100229329,"user_tz":-330,"elapsed":6612,"user":{"displayName":"Sparsh Agarwal","photoUrl":"","userId":"13037694610922482904"}},"outputId":"2ba6c423-062f-4e5a-b855-2210ea3245a2"},"source":["!pip install nvtabular"],"execution_count":10,"outputs":[{"output_type":"stream","text":["Collecting nvtabular\n"," Downloading nvtabular-0.5.3.tar.gz (191 kB)\n","\u001b[?25l\r\u001b[K |█▊ | 10 kB 32.4 MB/s eta 0:00:01\r\u001b[K |███▍ | 20 kB 22.9 MB/s eta 0:00:01\r\u001b[K |█████▏ | 30 kB 17.4 MB/s eta 0:00:01\r\u001b[K |██████▉ | 40 kB 15.1 MB/s eta 0:00:01\r\u001b[K |████████▌ | 51 kB 8.0 MB/s eta 0:00:01\r\u001b[K |██████████▎ | 61 kB 7.5 MB/s eta 0:00:01\r\u001b[K |████████████ | 71 kB 8.6 MB/s eta 0:00:01\r\u001b[K |█████████████▋ | 81 kB 8.9 MB/s eta 0:00:01\r\u001b[K |███████████████▍ | 92 kB 9.3 MB/s eta 0:00:01\r\u001b[K |█████████████████ | 102 kB 7.5 MB/s eta 0:00:01\r\u001b[K |██████████████████▉ | 112 kB 7.5 MB/s eta 0:00:01\r\u001b[K |████████████████████▌ | 122 kB 7.5 MB/s eta 0:00:01\r\u001b[K |██████████████████████▏ | 133 kB 7.5 MB/s eta 0:00:01\r\u001b[K |████████████████████████ | 143 kB 7.5 MB/s eta 0:00:01\r\u001b[K |█████████████████████████▋ | 153 kB 7.5 MB/s eta 0:00:01\r\u001b[K |███████████████████████████▎ | 163 kB 7.5 MB/s eta 0:00:01\r\u001b[K |█████████████████████████████ | 174 kB 7.5 MB/s eta 0:00:01\r\u001b[K |██████████████████████████████▊ | 184 kB 7.5 MB/s eta 0:00:01\r\u001b[K |████████████████████████████████| 191 kB 7.5 MB/s \n","\u001b[?25h Installing build dependencies ... \u001b[?25l\u001b[?25hdone\n"," Getting requirements to build wheel ... \u001b[?25l\u001b[?25hdone\n"," Preparing wheel metadata ... \u001b[?25l\u001b[?25hdone\n","Building wheels for collected packages: nvtabular\n"," Building wheel for nvtabular (PEP 517) ... \u001b[?25l\u001b[?25hdone\n"," Created wheel for nvtabular: filename=nvtabular-0.5.3-py3-none-any.whl size=231724 sha256=787778b5f104a27a6a51a7573ef7c03c40b7127b7a93df87eebe59495d2283c9\n"," Stored in directory: /root/.cache/pip/wheels/d0/d5/d3/177c1a93a7f274135cf9b1bce626c7a73b7c6f131d5bd9dbe3\n","Successfully built nvtabular\n","Installing collected packages: nvtabular\n","Successfully installed nvtabular-0.5.3\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"_hufhUdO5260","executionInfo":{"status":"ok","timestamp":1624100512836,"user_tz":-330,"elapsed":717,"user":{"displayName":"Sparsh Agarwal","photoUrl":"","userId":"13037694610922482904"}}},"source":["import glob\n","\n","import nvtabular as nvt\n","from nvtabular import ops"],"execution_count":15,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"dQsWVpfm7YmL"},"source":["### Load data"]},{"cell_type":"code","metadata":{"id":"orTKgniX7YFf"},"source":["!cp /content/drive/MyDrive/Recommendation/data_silver_l3.zip /content\n","!unzip /content/data_silver_l3.zip"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"zmvxvyZS6jvn","executionInfo":{"status":"ok","timestamp":1624100515125,"user_tz":-330,"elapsed":684,"user":{"displayName":"Sparsh Agarwal","photoUrl":"","userId":"13037694610922482904"}}},"source":["train_paths = glob.glob('/content/train.parquet')\n","valid_paths = glob.glob('/content/valid.parquet')\n","\n","train_dataset = nvt.Dataset(train_paths, engine='parquet', part_mem_fraction=0.15)\n","valid_dataset = nvt.Dataset(valid_paths, engine='parquet', part_mem_fraction=0.15)"],"execution_count":16,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":479},"id":"RZt4yaUd7w5O","executionInfo":{"status":"ok","timestamp":1624100523071,"user_tz":-330,"elapsed":624,"user":{"displayName":"Sparsh Agarwal","photoUrl":"","userId":"13037694610922482904"}},"outputId":"ad165de8-452f-4099-dfb9-0dab9c108e67"},"source":["train_dataset.head()"],"execution_count":17,"outputs":[{"output_type":"execute_result","data":{"text/html":["
\n","\n","\n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n"," \n","
event_timeevent_typeproduct_idbrandpriceuser_iduser_sessiontargettimestampts_hourts_minutets_weekdayts_dayts_monthts_yearcat_0cat_1cat_2cat_3
02020-01-23 10:22:27 UTCcart1005101xiaomi419.31602604607ec725848-03db-4a0d-94b2-d0121f79dbb502020-01-23 10:22:27102232312020constructiontoolslightNA
12020-01-23 10:22:27 UTCcart17300751versace81.57543745697860e30ab-c461-40ef-aa17-1f74acc94f2102020-01-23 10:22:27102232312020apparelshoessandalsNA
22020-01-23 10:22:37 UTCcart4802273samsung6.675127560880b8d5157-50ee-46ae-9772-e599004d1cbc02020-01-23 10:22:37102232312020sportbicycleNANA
32020-01-23 10:22:38 UTCcart2900561midea48.865713873272366dffb-a199-4aa1-b6a3-12fe807ea80d02020-01-23 10:22:38102232312020furniturebedroomblanketNA
42020-01-23 10:22:40 UTCcart17501003enough6.6959090228038efa242-cba4-45ac-8d28-a96d964d0e5702020-01-23 10:22:40102232312020apparelcostumeNANA
\n","
"],"text/plain":[" event_time event_type product_id ... cat_1 cat_2 cat_3\n","0 2020-01-23 10:22:27 UTC cart 1005101 ... tools light NA\n","1 2020-01-23 10:22:27 UTC cart 17300751 ... shoes sandals NA\n","2 2020-01-23 10:22:37 UTC cart 4802273 ... bicycle NA NA\n","3 2020-01-23 10:22:38 UTC cart 2900561 ... bedroom blanket NA\n","4 2020-01-23 10:22:40 UTC cart 17501003 ... costume NA NA\n","\n","[5 rows x 19 columns]"]},"metadata":{"tags":[]},"execution_count":17}]},{"cell_type":"markdown","metadata":{"id":"3-1ZEG7g75M8"},"source":["### Define the data schema"]},{"cell_type":"code","metadata":{"id":"77n7VYRw8Q0q"},"source":["nvt.Workflow()"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":214},"id":"QIMmMSAS70XS","executionInfo":{"status":"error","timestamp":1624100695691,"user_tz":-330,"elapsed":722,"user":{"displayName":"Sparsh Agarwal","photoUrl":"","userId":"13037694610922482904"}},"outputId":"0d06b37e-62de-4a3f-8fb3-b36cc26c8647"},"source":["proc = nvt.Workflow(\n"," cat_features=['product_id', 'brand', 'user_id',\n"," 'user_session', 'cat_0', 'cat_1', 'cat_2', 'cat_3',\n"," 'ts_hour', 'ts_minute', 'ts_weekday', 'ts_day', 'ts_month', 'ts_year'],\n"," cont_features=['price', 'timestamp'],\n"," label_name=['target']\n",")"],"execution_count":19,"outputs":[{"output_type":"error","ename":"TypeError","evalue":"ignored","traceback":["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m","\u001b[0;31mTypeError\u001b[0m Traceback (most recent call last)","\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m()\u001b[0m\n\u001b[1;32m 4\u001b[0m 'ts_hour', 'ts_minute', 'ts_weekday', 'ts_day', 'ts_month', 'ts_year'],\n\u001b[1;32m 5\u001b[0m \u001b[0mcont_features\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'price'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'timestamp'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 6\u001b[0;31m \u001b[0mlabel_name\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'target'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 7\u001b[0m )\n","\u001b[0;31mTypeError\u001b[0m: __init__() got an unexpected keyword argument 'cat_features'"]}]},{"cell_type":"code","metadata":{"id":"VOxTTtwG77GH"},"source":["proc.add_feature([\n"," ops.LambdaOp(\n"," op_name = 'user_id',\n"," f = lambda col, gdf: col.astype(str) + '_' + gdf['user_id'].astype(str),\n"," columns = ['product_id', 'brand', 'ts_hour', 'ts_minute'],\n"," replace=False\n"," ),\n"," ops.LambdaOp(\n"," op_name = 'user_id_brand',\n"," f = lambda col, gdf: col.astype(str) + '_' + gdf['user_id'].astype(str) + '_' + gdf['brand'].astype(str),\n"," columns = ['ts_hour', 'ts_weekday', 'cat_0', 'cat_1', 'cat_2'],\n"," replace=False\n"," ),\n"," ops.Categorify(\n"," freq_threshold=15,\n"," columns = [x + '_user_id' for x in ['product_id', 'brand', 'ts_hour', 'ts_minute']] + [x + '_user_id_brand' for x in ['ts_hour', 'ts_weekday', 'cat_0', 'cat_1', 'cat_2']] + ['product_id', 'brand', 'user_id', 'user_session', 'cat_0', 'cat_1', 'cat_2', 'cat_3', 'ts_hour', 'ts_minute', 'ts_weekday', 'ts_day', 'ts_month', 'ts_year']\n"," ),\n"," ops.LambdaOp(\n"," op_name = 'product_id',\n"," f = lambda col, gdf: col.astype(str) + '_' + gdf['product_id'].astype(str),\n"," columns = ['brand', 'user_id', 'cat_0'],\n"," replace=False\n"," ),\n"," ops.JoinGroupby(\n"," cont_names=[]\n"," ),\n"," ops.TargetEncoding(\n"," cat_groups = ['brand', 'user_id', 'product_id', 'cat_2', ['ts_weekday','ts_day']],\n"," cont_target= 'target',\n"," kfold=5,\n"," fold_seed=42,\n"," p_smooth=20,\n"," )\n","])"],"execution_count":null,"outputs":[]}]}