Update.
authorFrançois Fleuret <francois@fleuret.org>
Sun, 24 Mar 2024 15:13:37 +0000 (16:13 +0100)
committerFrançois Fleuret <francois@fleuret.org>
Sun, 24 Mar 2024 15:13:37 +0000 (16:13 +0100)
tasks.py

index 2f3db6a..02c44bb 100755 (executable)
--- a/tasks.py
+++ b/tasks.py
@@ -1909,6 +1909,54 @@ class Escape(Task):
     def vocabulary_size(self):
         return self.nb_codes
 
+    def thinking_autoregression(
+        self, n_epoch, model, result_dir, logger, deterministic_synthesis, nmax=1000
+    ):
+        result = self.test_input[:100].clone()
+        t = torch.arange(result.size(1), device=result.device)
+        itl = self.height * self.width + 3
+
+        def ar():
+            masked_inplace_autoregression(
+                model,
+                self.batch_size,
+                result,
+                ar_mask,
+                deterministic_synthesis,
+                device=self.device,
+            )
+
+        for u in range(itl, result.size(1) - itl + 1, itl):
+            print(f"{itl=} {u=} {result.size(1)=}")
+            result[:, u - 1] = (-1) + 1 + escape.first_lookahead_rewards_code
+            ar_mask = (t >= u).long() * (t < u + self.height * self.width).long()
+            ar_mask = ar_mask[None, :]
+            ar_mask = ar_mask.expand_as(result)
+            result *= 1 - ar_mask
+            ar()
+            result[:, u - 1] = (1) + 1 + escape.first_lookahead_rewards_code
+            ar_mask = (t >= self.height * self.width).long() * (
+                t < self.height * self.width + 2
+            ).long()
+            ar_mask = ar_mask[None, :]
+            ar_mask = ar_mask.expand_as(result)
+            result *= 1 - ar_mask
+            ar()
+
+        # Saving the generated sequences
+
+        s, a, r, lr = escape.seq2episodes(
+            result, self.height, self.width, lookahead=True
+        )
+        str = escape.episodes2str(
+            s, a, r, lookahead_rewards=lr, unicode=True, ansi_colors=True
+        )
+
+        filename = os.path.join(result_dir, f"test_thinking_seq_{n_epoch:04d}.txt")
+        with open(filename, "w") as f:
+            f.write(str)
+            logger(f"wrote {filename}")
+
     def produce_results(
         self, n_epoch, model, result_dir, logger, deterministic_synthesis, nmax=1000
     ):
@@ -1932,7 +1980,7 @@ class Escape(Task):
 
         ar_mask = (
             torch.arange(result.size(1), device=result.device)
-            > self.height * self.width + 2
+            >= self.height * self.width + 3
         ).long()[None, :]
         ar_mask = ar_mask.expand_as(result)
         result *= 1 - ar_mask  # paraaaaanoiaaaaaaa
@@ -1960,5 +2008,9 @@ class Escape(Task):
             f.write(str)
             logger(f"wrote {filename}")
 
+        self.thinking_autoregression(
+            n_epoch, model, result_dir, logger, deterministic_synthesis, nmax
+        )
+
 
 ######################################################################