X-Git-Url: https://fleuret.org/cgi-bin/gitweb/gitweb.cgi?a=blobdiff_plain;f=expr.py;h=8690504d81d63820d1f450f3b357302497759df6;hb=e38b98574f1966ea3a91ffb8fd9042f10a75ca88;hp=e073dcae327368c1edcda745f712ce75a5113d8e;hpb=6234d5163912a5b0483384ca7c44e03b0ea54333;p=picoclvr.git

diff --git a/expr.py b/expr.py
index e073dca..8690504 100755
--- a/expr.py
+++ b/expr.py
@@ -24,7 +24,7 @@ def random_expr(variables, budget):
         else:
             return str(torch.randint(10, (1,)).item())
     else:
-        op = torch.randint(4, (1,)).item()
+        op = torch.randint(3, (1,)).item()
         if op == 0:
             e = random_expr(variables, budget - 2)
             if ("+" in e or "-" in e or "*" in e) and (e[0] != "(" or e[-1] != ")"):
@@ -38,18 +38,22 @@ def random_expr(variables, budget):
             if op == 1:
                 return e1 + "+" + e2
             elif op == 2:
-                return e1 + "+" + e2
-            elif op == 3:
                 return e1 + "*" + e2
 
 
 def generate_program(nb_variables, length):
     s = ""
     variables = set()
+
+    # We take length itself half of the time, and uniform between 1
+    # and length otherwise. The actual length can be slightly greater
+
+    length = min(length, 1 + torch.randint(length * 2, (1,)).item())
     while len(s) < length:
         v = random_var(nb_variables=nb_variables)
         s += v + "=" + random_expr(variables, budget=20) + ";"
         variables.add(v)
+
     return s, variables
 
 
@@ -62,14 +66,14 @@ def extract_results(seq):
     return results
 
 
-def generate_sequences(nb, nb_variables=5, length=20, randomize_length=False):
+def generate_sequences(nb, nb_variables=5, length=20):
+    assert nb_variables <= 26
     sequences = []
+    result_max = 99
     for n in range(nb):
         result = None
-        while result == None or max(result.values()) > 100:
+        while result == None or max(result.values()) > result_max:
             l = length
-            if l > 5 and randomize_length:
-                l = 5 + torch.randint(l - 5, (1,)).item()
             p, v = generate_program(nb_variables, l)
             v = ", ".join(['"' + v + '": ' + v for v in v])
             ldict = {}
@@ -87,7 +91,7 @@ if __name__ == "__main__":
     import time
 
     start_time = time.perf_counter()
-    sequences = generate_sequences(1000)
+    sequences = generate_sequences(1000, length=40)
     end_time = time.perf_counter()
     for s in sequences[:10]:
         print(s)