vimrc

.vimrc

@@ -0,0 +1,42 @@
+set encoding=utf8
+set expandtab
+set textwidth=0
+set tabstop=4
+set softtabstop=4
+set shiftwidth=4
+set backspace=indent,eol,start
+set incsearch
+set ruler
+set wildmenu
+set commentstring=\ #\ %s
+set foldlevel=0
+set clipboard+=unnamed
+set tags=/home/huanghuang/shire/tags
+syntax on
+set list listchars=tab:▷⋅,trail:⋅,nbsp:⋅
+set statusline=%F%m%r%h%w\ [TYPE=%Y\ %{&ff}]\
+\ [%l/%L\ (%p%%)
+filetype plugin indent on
+au FileType py set autoindent
+au FileType py set smartindent
+au FileType py set textwidth=79 " PEP-8 Friendly
+" NERD_tree config
+let NERDTreeChDirMode=2
+let NERDTreeIgnore=['\.vim$', '\~$', '\.pyc$', '\.swp$']
+let NERDTreeSortOrder=['^__\.py$', '\/$', '*', '\.swp$',  '\~$']
+let NERDTreeShowBookmarks=1
+map <F3> :NERDTreeToggle<CR>
+" Viewport Controls
+" ie moving between split panes
+map <silent>,h <C-w>h
+map <silent>,j <C-w>j
+map <silent>,k <C-w>k
+map <silent>,l <C-w>l
+
+let g:pydiction_location = '/home/huanghuang/.vim/pydiction-1.2/complete-dict'
+"autocmd FileType python compiler pylint
+"let g:pylint_onwrite = 0
+au BufRead,BufNewFile *.c,*.cpp,*.py 2match Underlined /.\%81v/
+if has("gui_running")
+    highlight SpellBad term=underline gui=undercurl guisp=Orange 
+endif

hw-4.py

@@ -0,0 +1,84 @@
+# --------------
+# USER INSTRUCTIONS
+#
+# Write a function called stochastic_value that 
+# takes no input and RETURNS two grids. The
+# first grid, value, should contain the computed
+# value of each cell as shown in the video. The
+# second grid, policy, should contain the optimum
+# policy for each cell.
+#
+# Stay tuned for a homework help video! This should
+# be available by Thursday and will be visible
+# in the course content tab.
+#
+# Good luck! Keep learning!
+#
+# --------------
+# GRADING NOTES
+#
+# We will be calling your stochastic_value function
+# with several different grids and different values
+# of success_prob, collision_cost, and cost_step.
+# In order to be marked correct, your function must
+# RETURN (it does not have to print) two grids,
+# value and policy.
+#
+# When grading your value grid, we will compare the
+# value of each cell with the true value according
+# to this model. If your answer for each cell
+# is sufficiently close to the correct answer
+# (within 0.001), you will be marked as correct.
+#
+# NOTE: Please do not modify the values of grid,
+# success_prob, collision_cost, or cost_step inside
+# your function. Doing so could result in your
+# submission being inappropriately marked as incorrect.
+
+# -------------
+# GLOBAL VARIABLES
+#
+# You may modify these variables for testing
+# purposes, but you should only modify them here.
+# Do NOT modify them inside your stochastic_value
+# function.
+
+grid = [[0, 0, 0, 0],
+        [0, 0, 0, 0],
+        [0, 0, 0, 0],
+        [0, 1, 1, 0]]
+
+goal = [0, len(grid[0])-1] # Goal is in top right corner
+
+
+delta = [[-1, 0 ], # go up
+         [ 0, -1], # go left
+         [ 1, 0 ], # go down
+         [ 0, 1 ]] # go right
+
+delta_name = ['^', '<', 'v', '>'] # Use these when creating your policy grid.
+
+success_prob = 0.5                      
+failure_prob = (1.0 - success_prob)/2.0 # Probability(stepping left) = prob(stepping right) = failure_prob
+collision_cost = 100                    
+cost_step = 1        
+
+
+############## INSERT/MODIFY YOUR CODE BELOW ##################
+#
+# You may modify the code below if you want, but remember that
+# your function must...
+#
+# 1) ...be called stochastic_value().
+# 2) ...NOT take any arguments.
+# 3) ...return two grids: FIRST value and THEN policy.
+
+def stochastic_value():
+    value = [[1000 for row in range(len(grid[0]))] for col in range(len(grid))]
+    policy = [[' ' for row in range(len(grid[0]))] for col in range(len(grid))]
+
+    return value, policy
+
+
+
+