Spaces:
Sleeping
Sleeping
File size: 7,535 Bytes
44db2f9 b8d0d32 44db2f9 b8d0d32 44db2f9 b8d0d32 44db2f9 b8d0d32 44db2f9 b8d0d32 44db2f9 b8d0d32 44db2f9 b8d0d32 44db2f9 b8d0d32 44db2f9 b8d0d32 44db2f9 b8d0d32 44db2f9 b8d0d32 44db2f9 b8d0d32 44db2f9 b8d0d32 44db2f9 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 |
import pytest
from . import wordle
from . import state
TESTWORDS = [
"APPAA",
"APPAB",
"APPAC",
"APPAD",
"BPPAB",
"BPPAC",
"BPPAD",
"CPPAB",
"CPPAC",
"CPPAD",
]
@pytest.fixture
def wordleEnv():
env = wordle.WordleEnvBase(
words=TESTWORDS,
max_turns=6,
)
return env
def test_reset(wordleEnv):
wordleEnv.reset(seed=13)
def test_guess_win(wordleEnv):
wordleEnv.reset(seed=13)
goal = wordleEnv.goal_word
new_state, reward, done, _ = wordleEnv.step(goal)
assert done
assert wordleEnv.done
assert reward == 0
try:
wordleEnv.step(goal)
raise ValueError("Shouldn't reach here!")
except ValueError:
pass
def test_win_reward(wordleEnv):
wordleEnv.reset(seed=13)
goal = wordleEnv.goal_word
word_index = (goal + 1) % len(wordleEnv.words)
new_state, reward, done, _ = wordleEnv.step(word_index)
assert state.remaining_steps(new_state) == wordleEnv.max_turns-1
assert not done
assert not wordleEnv.done
assert reward < 1
new_state, reward, done, _ = wordleEnv.step(goal)
assert state.remaining_steps(new_state) == wordleEnv.max_turns-2
assert done
assert wordleEnv.done
assert reward == wordle.REWARD
try:
wordleEnv.step(goal)
raise ValueError("Shouldn't reach here!")
except ValueError:
pass
def test_win_reward_6(wordleEnv):
wordleEnv.reset(seed=13)
goal = wordleEnv.goal_word
random_word = (goal + 1) % len(wordleEnv.words)
for i in range(5):
new_state, reward, done, _ = wordleEnv.step(random_word)
new_state, reward, done, _ = wordleEnv.step(goal)
assert wordleEnv.max_turns - state.remaining_steps(new_state) == 6
assert done
assert wordleEnv.done
assert reward == wordle.REWARD
def test_lose_reward(wordleEnv):
wordleEnv.reset(seed=13)
goal = wordleEnv.goal_word
random_word = (goal + 1) % len(wordleEnv.words)
for i in range(1, wordleEnv.max_turns):
new_state, reward, done, _ = wordleEnv.step(random_word)
assert state.remaining_steps(new_state) == wordleEnv.max_turns-i
assert not done
assert not wordleEnv.done
assert reward < 1
word_index = (goal + wordleEnv.max_turns) % len(wordleEnv.words)
random_word = word_index
new_state, reward, done, _ = wordleEnv.step(random_word)
assert state.remaining_steps(new_state) == 0
assert done
assert wordleEnv.done
assert reward < -wordle.REWARD + 1
try:
wordleEnv.step(goal)
raise ValueError("Shouldn't reach here!")
except ValueError:
pass
def test_step(wordleEnv):
wordleEnv.reset(seed=13)
wordleEnv.set_goal_encoded(0)
cur_state = wordleEnv.state
new_state, reward, done, _ = wordleEnv.step(1)
assert state.remaining_steps(cur_state) == wordleEnv.max_turns
assert state.remaining_steps(new_state) == wordleEnv.max_turns-1
# Expect B to be all 1,0,0
offset = 1+3*5*(ord('B')-ord('A'))
assert tuple(new_state[offset:offset+15]) == tuple([1, 0, 0]*5)
# Expect A to be right in position 0 4 and maybe otherwise
offset = 1
assert tuple(new_state[offset:offset+15]) == (0, 0, 1,
1, 0, 0,
1, 0, 0,
0, 0, 1,
0, 0, 0)
# Expect P to be right in position 2 3 and maybe otherwise
offset = 1 + 3*5*(ord('P') - ord('A'))
assert tuple(new_state[offset:offset+15]) == (1, 0, 0,
0, 0, 1,
0, 0, 1,
1, 0, 0,
0, 0, 0)
# Expect C to be maybes
offset = 1 + 3*5*(ord('C') - ord('A'))
assert tuple(new_state[offset:offset+15]) == (1, 0, 0,
1, 0, 0,
1, 0, 0,
1, 0, 0,
0, 0, 0)
cur_state = wordleEnv.state
new_state, reward, done, _ = wordleEnv.step(1)
assert state.remaining_steps(cur_state) == wordleEnv.max_turns-1
assert state.remaining_steps(new_state) == wordleEnv.max_turns-2
# Expect B to be all 1,0,0
offset = 1+3*5*(ord('B')-ord('A'))
assert tuple(new_state[offset:offset+15]) == tuple([1, 0, 0]*5)
# Expect A to be right in position 0 4 and maybe otherwise
offset = 1
assert tuple(new_state[offset:offset+15]) == (0, 0, 1,
1, 0, 0,
1, 0, 0,
0, 0, 1,
0, 0, 0)
# Expect P to be right in position 2 3 and maybe otherwise
offset = 1 + 3*5*(ord('P') - ord('A'))
assert tuple(new_state[offset:offset+15]) == (1, 0, 0,
0, 0, 1,
0, 0, 1,
1, 0, 0,
0, 0, 0)
new_state, reward, done, _ = wordleEnv.step(2)
assert state.remaining_steps(new_state) == wordleEnv.max_turns-3
# Expect B to be all 1,0,0
offset = 1 + 3*5*(ord('B')-ord('A'))
assert tuple(new_state[offset:offset+15]) == tuple([1, 0, 0]*5)
# Expect C to be all 1,0,0
offset = 1+3*5*(ord('C')-ord('A'))
assert tuple(new_state[offset:offset+15]) == tuple([1, 0, 0]*5)
# Expect A to be right in position 0 4 and maybe otherwise
offset = 1
assert tuple(new_state[offset:offset+15]) == (0, 0, 1,
1, 0, 0,
1, 0, 0,
0, 0, 1,
0, 0, 0)
# Expect P to be right in position 2 3 and maybe otherwise
offset = 1 + 3*5*(ord('P') - ord('A'))
assert tuple(new_state[offset:offset+15]) == (1, 0, 0,
0, 0, 1,
0, 0, 1,
1, 0, 0,
0, 0, 0)
new_state, reward, done, _ = wordleEnv.step(0)
# Expect A to be right in position 0 4 and 5
offset = 1
assert tuple(new_state[offset:offset+15]) == (0, 0, 1,
1, 0, 0,
1, 0, 0,
0, 0, 1,
0, 0, 1)
# Expect P to be right in position 2 3 and not otherwise
offset = 1 + 3*5*(ord('P') - ord('A'))
assert tuple(new_state[offset:offset+15]) == (1, 0, 0,
0, 0, 1,
0, 0, 1,
1, 0, 0,
1, 0, 0)
assert state.remaining_steps(new_state) == wordleEnv.max_turns-4
assert done
assert wordleEnv.done
assert reward == wordle.REWARD
|