X-Git-Url: https://fleuret.org/cgi-bin/gitweb/gitweb.cgi?a=blobdiff_plain;f=main.py;h=1d52b6defc3cd22ddea610e1fbfc071358354a47;hb=cd3329fc206bacfd90a8e2cbe364244359568733;hp=496a6034b857d303baa1a32ab8bb48fd68bf84eb;hpb=503298855a80bde0bf856f1a34b532079d3c7ef6;p=picoclvr.git

diff --git a/main.py b/main.py
index 496a603..1d52b6d 100755
--- a/main.py
+++ b/main.py
@@ -33,7 +33,7 @@ parser.add_argument(
     "--task",
     type=str,
     default="twotargets",
-    help="byheart, learnop, guessop, mixing, twotargets, addition, picoclvr, mnist, maze, snake, stack, expr, rpl, grid, qmlp",
+    help="byheart, learnop, guessop, mixing, memory, twotargets, addition, picoclvr, mnist, maze, snake, stack, expr, rpl, grid, qmlp",
 )
 
 parser.add_argument("--log_filename", type=str, default="train.log", help=" ")
@@ -160,10 +160,12 @@ parser.add_argument("--expr_result_max", type=int, default=99)
 parser.add_argument("--expr_input_file", type=str, default=None)
 
 ##############################
-# Misc
+# Mixing
 
 parser.add_argument("--mixing_hard", action="store_true", default=False)
 
+parser.add_argument("--mixing_deterministic_start", action="store_true", default=False)
+
 ######################################################################
 
 args = parser.parse_args()
@@ -254,6 +256,12 @@ default_task_args = {
         "nb_train_samples": 50000,
         "nb_test_samples": 10000,
     },
+    "memory": {
+        "model": "4M",
+        "batch_size": 100,
+        "nb_train_samples": 5000,
+        "nb_test_samples": 1000,
+    },
     "mixing": {
         "model": "37M",
         "batch_size": 25,
@@ -283,6 +291,13 @@ default_model_args = {
         "nb_heads": 2,
         "nb_blocks": 2,
     },
+    "4M": {
+        "dim_model": 256,
+        "dim_keys": 32,
+        "dim_hidden": 1024,
+        "nb_heads": 4,
+        "nb_blocks": 6,
+    },
     "37M": {
         "dim_model": 512,
         "dim_keys": 64,
@@ -346,6 +361,8 @@ def log_string(s):
     sys.stdout.flush()
 
 
+log_string(f"argv {' '.join(sys.argv)}")
+
 for n in vars(args):
     log_string(f"args.{n} {getattr(args, n)}")
 
@@ -414,9 +431,21 @@ elif args.task == "twotargets":
         device=device,
     )
 
+elif args.task == "memory":
+    task = tasks.SandBox(
+        problem=problems.ProblemMemory(),
+        nb_train_samples=args.nb_train_samples,
+        nb_test_samples=args.nb_test_samples,
+        batch_size=args.batch_size,
+        logger=log_string,
+        device=device,
+    )
+
 elif args.task == "mixing":
     task = tasks.SandBox(
-        problem=problems.ProblemMixing(hard=args.mixing_hard),
+        problem=problems.ProblemMixing(
+            hard=args.mixing_hard, random_start=not args.mixing_deterministic_start
+        ),
         nb_train_samples=args.nb_train_samples,
         nb_test_samples=args.nb_test_samples,
         batch_size=args.batch_size,