[DOCS] Document cloudpickle dependency in tutorials

docs/install/from_source.rst

@@ -248,7 +248,7 @@ like ``virtualenv``.
 
    .. code:: bash
 
-       pip3 install --user tornado psutil xgboost
+       pip3 install --user tornado psutil xgboost cloudpickle
 
 
 Install Contrib Libraries

tutorials/autotvm/tune_conv2d_cuda.py

@@ -36,7 +36,7 @@
 #
 # .. code-block:: bash
 #
-#   pip3 install --user psutil xgboost tornado
+#   pip3 install --user psutil xgboost tornado cloudpickle
 #
 # To make TVM run faster in tuning, it is recommended to use cython
 # as FFI of tvm. In the root directory of tvm, execute

tutorials/autotvm/tune_relay_arm.py

@@ -49,7 +49,7 @@
 #
 # .. code-block:: bash
 #
-#   pip3 install --user psutil xgboost tornado
+#   pip3 install --user psutil xgboost tornado cloudpickle
 #
 # To make TVM run faster during tuning, it is recommended to use cython
 # as FFI of TVM. In the root directory of TVM, execute

tutorials/autotvm/tune_relay_cuda.py

@@ -47,7 +47,7 @@
 #
 # .. code-block:: bash
 #
-#   pip3 install --user psutil xgboost tornado
+#   pip3 install --user psutil xgboost tornado cloudpickle
 #
 # To make TVM run faster during tuning, it is recommended to use cython
 # as FFI of tvm. In the root directory of tvm, execute:

tutorials/autotvm/tune_relay_mobile_gpu.py

@@ -47,7 +47,7 @@
 #
 # .. code-block:: bash
 #
-#   pip3 install --user psutil xgboost tornado
+#   pip3 install --user psutil xgboost tornado cloudpickle
 #
 # To make TVM run faster during tuning, it is recommended to use cython
 # as FFI of tvm. In the root directory of tvm, execute

tutorials/autotvm/tune_simple_template.py

@@ -41,7 +41,7 @@
 #
 # .. code-block:: bash
 #
-#   pip3 install --user psutil xgboost
+#   pip3 install --user psutil xgboost cloudpickle
 #
 # To make TVM run faster in tuning, it is recommended to use cython
 # as FFI of TVM. In the root directory of TVM, execute

vta/tutorials/autotvm/tune_relay_vta.py

@@ -40,7 +40,7 @@
 #
 # .. code-block:: bash
 #
-#   pip3 install --user psutil xgboost tornado mxnet requests "Pillow<7"
+#   pip3 install --user psutil xgboost tornado mxnet requests "Pillow<7" cloudpickle
 #
 # To make TVM run faster during tuning, it is recommended to use cython
 # as FFI of TVM. In the root directory of TVM, execute