task 3 with wandb

plr_exercise.egg-info/PKG-INFO

@@ -0,0 +1,10 @@
+Metadata-Version: 2.1
+Name: plr_exercise
+Version: 1.0.0
+Summary: A small example package
+Author: Jonas Frey
+Author-email: [email protected]
+Requires-Python: >=3.7
+License-File: LICENSE
+Requires-Dist: numpy
+Requires-Dist: torch>=1.21

plr_exercise.egg-info/SOURCES.txt

@@ -0,0 +1,9 @@
+LICENSE
+README.md
+setup.py
+plr_exercise/__init__.py
+plr_exercise.egg-info/PKG-INFO
+plr_exercise.egg-info/SOURCES.txt
+plr_exercise.egg-info/dependency_links.txt
+plr_exercise.egg-info/requires.txt
+plr_exercise.egg-info/top_level.txt

plr_exercise.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+https://download.pytorch.org/whl/torch-2.1.0+cu121-cp38-cp38-linux_x86_64.whl

plr_exercise.egg-info/requires.txt

@@ -0,0 +1,2 @@
+numpy
+torch>=1.21

plr_exercise.egg-info/top_level.txt

@@ -0,0 +1 @@
+plr_exercise

train.py

@@ -6,15 +6,25 @@
 import torch.optim as optim
 from torchvision import datasets, transforms
 from torch.optim.lr_scheduler import StepLR
+import wandb
+
+
+run = wandb.init(
+    project="mon projet",
+    config={
+        "learning_rate": 0.01,
+        "epochs": 10,
+
+    },
+)
 
 
 class Net(nn.Module):
     def __init__(self):
 
-
         super(Net, self).__init__()
         self.conv1 = nn.Conv2d(1,   32, 3, 1)
-        self.conv2 = nn.Conv2d( 32, 64, 3, 1)
+        self.conv2 = nn.Conv2d(32, 64, 3, 1)
         self.dropout1 = nn.Dropout(0.25)
         self.dropout2 = nn.Dropout(0.5)
         self.fc1 = nn.Linear(9216, 128)
@@ -43,7 +53,7 @@ def train(args,    model, device, train_loader, optimizer, epoch):
         data, target = data.to(device),   target.to(device)
         optimizer.zero_grad()
         output = model(data)
-        loss = F.nll_loss(   output, target)
+        loss = F.nll_loss(output, target)
         loss.backward()
         optimizer.step()
         if batch_idx % args.log_interval == 0:
@@ -52,7 +62,7 @@ def train(args,    model, device, train_loader, optimizer, epoch):
                     epoch,
                     batch_idx * len(data),
                     len(train_loader.dataset),
-                    100.0 *    batch_idx / len(train_loader),
+                    100.0 * batch_idx / len(train_loader),
                     loss.item(),
                 )
             )
@@ -70,17 +80,21 @@ def test(model, device, test_loader, epoch):
 
             data, target = data.to(device), target.to(device)
             output = model(data)
-            test_loss += F.nll_loss(   output,   target,   reduction="sum").item()  # sum up batch loss
-            pred = output.argmax(dim=1, keepdim=True)  # get the index of the max log-probability
-            correct += pred.eq(target.view_as(pred)  ).sum().item()
+            # sum up batch loss
+            test_loss += F.nll_loss(output,   target,   reduction="sum").item()
+            # get the index of the max log-probability
+            pred = output.argmax(dim=1, keepdim=True)
+            correct += pred.eq(target.view_as(pred)).sum().item()
 
-    test_loss /= len(test_loader.dataset  )
+    test_loss /= len(test_loader.dataset)
 
     print(
         "\nTest set: Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)\n".format(
-            test_loss, correct, len(test_loader.dataset), 100.0 * correct / len(test_loader.dataset)
+            test_loss, correct, len(
+                test_loader.dataset), 100.0 * correct / len(test_loader.dataset)
         )
     )
+    wandb.log({"Loss": test_loss, "epoch": epoch})
 
 
 def main():
@@ -92,20 +106,27 @@ def main():
     parser.add_argument(
         "--test-batch-size", type=int, default=1000, metavar="N", help="input batch size for testing (default: 1000)"
     )
-    parser.add_argument("--epochs", type=int, default=2, metavar="N", help="number of epochs to train (default: 14)")
-    parser.add_argument("--lr", type=float, default=1.0, metavar="LR", help="learning rate (default: 1.0)")
-    parser.add_argument("--gamma", type=float, default=0.7, metavar="M", help="Learning rate step gamma (default: 0.7)")
-    parser.add_argument("--no-cuda", action="store_true", default=False, help="disables CUDA training")
-    parser.add_argument("--dry-run", action="store_true", default=False, help="quickly check a single pass")
-    parser.add_argument("--seed", type=int, default=1, metavar="S", help="random seed (default: 1)")
+    parser.add_argument("--epochs", type=int, default=2, metavar="N",
+                        help="number of epochs to train (default: 14)")
+    parser.add_argument("--lr", type=float, default=1.0,
+                        metavar="LR", help="learning rate (default: 1.0)")
+    parser.add_argument("--gamma", type=float, default=0.7,
+                        metavar="M", help="Learning rate step gamma (default: 0.7)")
+    parser.add_argument("--no-cuda", action="store_true",
+                        default=False, help="disables CUDA training")
+    parser.add_argument("--dry-run", action="store_true",
+                        default=False, help="quickly check a single pass")
+    parser.add_argument("--seed", type=int, default=1,
+                        metavar="S", help="random seed (default: 1)")
     parser.add_argument(
         "--log-interval",
         type=int,
         default=10,
         metavar="N",
         help="how many batches to wait before logging training status",
     )
-    parser.add_argument("--save-model", action="store_true", default=False, help="For Saving the current Model")
+    parser.add_argument("--save-model", action="store_true",
+                        default=False, help="For Saving the current Model")
     args = parser.parse_args()
     use_cuda = not args.no_cuda and torch.cuda.is_available()
 
@@ -123,8 +144,10 @@ def main():
         train_kwargs.update(cuda_kwargs)
         test_kwargs.update(cuda_kwargs)
 
-    transform = transforms.Compose([transforms.ToTensor(), transforms.Normalize((0.1307,), (0.3081,))])
-    dataset1 = datasets.MNIST("../data", train=True, download=True, transform=transform)
+    transform = transforms.Compose(
+        [transforms.ToTensor(), transforms.Normalize((0.1307,), (0.3081,))])
+    dataset1 = datasets.MNIST("../data", train=True,
+                              download=True, transform=transform)
     dataset2 = datasets.MNIST("../data", train=False, transform=transform)
     train_loader = torch.utils.data.DataLoader(dataset1, **train_kwargs)
     test_loader = torch.utils.data.DataLoader(dataset2, **test_kwargs)

wandb/debug-internal.log

@@ -0,0 +1 @@
+run-20240306_104616-pufbvux4/logs/debug-internal.log

wandb/debug.log

@@ -0,0 +1 @@
+run-20240306_104616-pufbvux4/logs/debug.log

wandb/latest-run

@@ -0,0 +1 @@
+run-20240306_104616-pufbvux4

wandb/run-20240306_104616-pufbvux4/files/config.yaml

@@ -0,0 +1,35 @@
+wandb_version: 1
+
+learning_rate:
+  desc: null
+  value: 0.01
+epochs:
+  desc: null
+  value: 10
+_wandb:
+  desc: null
+  value:
+    python_version: 3.11.5
+    cli_version: 0.16.4
+    framework: torch
+    is_jupyter_run: false
+    is_kaggle_kernel: false
+    start_time: 1709718376.0
+    t:
+      1:
+      - 1
+      - 41
+      - 55
+      2:
+      - 1
+      - 41
+      - 55
+      3:
+      - 16
+      - 23
+      4: 3.11.5
+      5: 0.16.4
+      8:
+      - 4
+      - 5
+      13: darwin-arm64

wandb/run-20240306_104616-pufbvux4/files/output.log

@@ -0,0 +1,190 @@
+Train Epoch: 0 [0/60000 (0%)]	Loss: 2.329474
+Train Epoch: 0 [640/60000 (1%)]	Loss: 3.163749
+Train Epoch: 0 [1280/60000 (2%)]	Loss: 2.810723
+Train Epoch: 0 [1920/60000 (3%)]	Loss: 2.462003
+Train Epoch: 0 [2560/60000 (4%)]	Loss: 2.393042
+Train Epoch: 0 [3200/60000 (5%)]	Loss: 727.100647
+Train Epoch: 0 [3840/60000 (6%)]	Loss: 2.365071
+Train Epoch: 0 [4480/60000 (7%)]	Loss: 2.351858
+Train Epoch: 0 [5120/60000 (9%)]	Loss: 2.322984
+Train Epoch: 0 [5760/60000 (10%)]	Loss: 2.334341
+Train Epoch: 0 [6400/60000 (11%)]	Loss: 2.350302
+Train Epoch: 0 [7040/60000 (12%)]	Loss: 2.328267
+Train Epoch: 0 [7680/60000 (13%)]	Loss: 2.351661
+Train Epoch: 0 [8320/60000 (14%)]	Loss: 2.312091
+Train Epoch: 0 [8960/60000 (15%)]	Loss: 2.401574
+Train Epoch: 0 [9600/60000 (16%)]	Loss: 2.336918
+Train Epoch: 0 [10240/60000 (17%)]	Loss: 2.294845
+Train Epoch: 0 [10880/60000 (18%)]	Loss: 2.343280
+Train Epoch: 0 [11520/60000 (19%)]	Loss: 2.362715
+Train Epoch: 0 [12160/60000 (20%)]	Loss: 2.298522
+Train Epoch: 0 [12800/60000 (21%)]	Loss: 2.286546
+Train Epoch: 0 [13440/60000 (22%)]	Loss: 2.339040
+Train Epoch: 0 [14080/60000 (23%)]	Loss: 2.336880
+Train Epoch: 0 [14720/60000 (25%)]	Loss: 2.348043
+Train Epoch: 0 [15360/60000 (26%)]	Loss: 2.330961
+Train Epoch: 0 [16000/60000 (27%)]	Loss: 2.320678
+Train Epoch: 0 [16640/60000 (28%)]	Loss: 2.373642
+Train Epoch: 0 [17280/60000 (29%)]	Loss: 2.362350
+Train Epoch: 0 [17920/60000 (30%)]	Loss: 2.285732
+Train Epoch: 0 [18560/60000 (31%)]	Loss: 2.387622
+Train Epoch: 0 [19200/60000 (32%)]	Loss: 2.334290
+Train Epoch: 0 [19840/60000 (33%)]	Loss: 2.351022
+Train Epoch: 0 [20480/60000 (34%)]	Loss: 2.451580
+Train Epoch: 0 [21120/60000 (35%)]	Loss: 2.432407
+Train Epoch: 0 [21760/60000 (36%)]	Loss: 2.357752
+Train Epoch: 0 [22400/60000 (37%)]	Loss: 2.384685
+Train Epoch: 0 [23040/60000 (38%)]	Loss: 2.421635
+Train Epoch: 0 [23680/60000 (39%)]	Loss: 2.375398
+Train Epoch: 0 [24320/60000 (41%)]	Loss: 2.312469
+Train Epoch: 0 [24960/60000 (42%)]	Loss: 2.293869
+Train Epoch: 0 [25600/60000 (43%)]	Loss: 2.306481
+Train Epoch: 0 [26240/60000 (44%)]	Loss: 2.339118
+Train Epoch: 0 [26880/60000 (45%)]	Loss: 2.329201
+Train Epoch: 0 [27520/60000 (46%)]	Loss: 2.402405
+Train Epoch: 0 [28160/60000 (47%)]	Loss: 2.362206
+Train Epoch: 0 [28800/60000 (48%)]	Loss: 2.346026
+Train Epoch: 0 [29440/60000 (49%)]	Loss: 2.359084
+Train Epoch: 0 [30080/60000 (50%)]	Loss: 2.382750
+Train Epoch: 0 [30720/60000 (51%)]	Loss: 2.302515
+Train Epoch: 0 [31360/60000 (52%)]	Loss: 2.340575
+Train Epoch: 0 [32000/60000 (53%)]	Loss: 2.352647
+Train Epoch: 0 [32640/60000 (54%)]	Loss: 2.329711
+Train Epoch: 0 [33280/60000 (55%)]	Loss: 2.331157
+Train Epoch: 0 [33920/60000 (57%)]	Loss: 2.365947
+Train Epoch: 0 [34560/60000 (58%)]	Loss: 2.310303
+Train Epoch: 0 [35200/60000 (59%)]	Loss: 2.403242
+Train Epoch: 0 [35840/60000 (60%)]	Loss: 2.366431
+Train Epoch: 0 [36480/60000 (61%)]	Loss: 2.329873
+Train Epoch: 0 [37120/60000 (62%)]	Loss: 2.329306
+Train Epoch: 0 [37760/60000 (63%)]	Loss: 2.313803
+Train Epoch: 0 [38400/60000 (64%)]	Loss: 2.392514
+Train Epoch: 0 [39040/60000 (65%)]	Loss: 2.323086
+Train Epoch: 0 [39680/60000 (66%)]	Loss: 2.471573
+Train Epoch: 0 [40320/60000 (67%)]	Loss: 3.164428
+Train Epoch: 0 [40960/60000 (68%)]	Loss: 2.648396
+Train Epoch: 0 [41600/60000 (69%)]	Loss: 2.387139
+Train Epoch: 0 [42240/60000 (70%)]	Loss: 2.344292
+Train Epoch: 0 [42880/60000 (71%)]	Loss: 2.378800
+Train Epoch: 0 [43520/60000 (72%)]	Loss: 2.317856
+Train Epoch: 0 [44160/60000 (74%)]	Loss: 2.365886
+Train Epoch: 0 [44800/60000 (75%)]	Loss: 2.275558
+Train Epoch: 0 [45440/60000 (76%)]	Loss: 2.385407
+Train Epoch: 0 [46080/60000 (77%)]	Loss: 2.352269
+Train Epoch: 0 [46720/60000 (78%)]	Loss: 2.331173
+Train Epoch: 0 [47360/60000 (79%)]	Loss: 2.333803
+Train Epoch: 0 [48000/60000 (80%)]	Loss: 2.368239
+Train Epoch: 0 [48640/60000 (81%)]	Loss: 2.323156
+Train Epoch: 0 [49280/60000 (82%)]	Loss: 2.348527
+Train Epoch: 0 [49920/60000 (83%)]	Loss: 2.382334
+Train Epoch: 0 [50560/60000 (84%)]	Loss: 2.492201
+Train Epoch: 0 [51200/60000 (85%)]	Loss: 2.390026
+Train Epoch: 0 [51840/60000 (86%)]	Loss: 2.466917
+Train Epoch: 0 [52480/60000 (87%)]	Loss: 2.296052
+Train Epoch: 0 [53120/60000 (88%)]	Loss: 2.334232
+Train Epoch: 0 [53760/60000 (90%)]	Loss: 2.384785
+Train Epoch: 0 [54400/60000 (91%)]	Loss: 2.349325
+Train Epoch: 0 [55040/60000 (92%)]	Loss: 2.294627
+Train Epoch: 0 [55680/60000 (93%)]	Loss: 2.314354
+Train Epoch: 0 [56320/60000 (94%)]	Loss: 2.313751
+Train Epoch: 0 [56960/60000 (95%)]	Loss: 2.326008
+Train Epoch: 0 [57600/60000 (96%)]	Loss: 2.256036
+Train Epoch: 0 [58240/60000 (97%)]	Loss: 2.343266
+Train Epoch: 0 [58880/60000 (98%)]	Loss: 2.317806
+Train Epoch: 0 [59520/60000 (99%)]	Loss: 2.336383
+Test set: Average loss: 2.3195, Accuracy: 1135/10000 (11%)
+Train Epoch: 1 [0/60000 (0%)]	Loss: 2.308809
+Train Epoch: 1 [640/60000 (1%)]	Loss: 2.314591
+Train Epoch: 1 [1280/60000 (2%)]	Loss: 2.356146
+Train Epoch: 1 [1920/60000 (3%)]	Loss: 2.341244
+Train Epoch: 1 [2560/60000 (4%)]	Loss: 2.321081
+Train Epoch: 1 [3200/60000 (5%)]	Loss: 2.308159
+Train Epoch: 1 [3840/60000 (6%)]	Loss: 2.312081
+Train Epoch: 1 [4480/60000 (7%)]	Loss: 2.357321
+Train Epoch: 1 [5120/60000 (9%)]	Loss: 2.317735
+Train Epoch: 1 [5760/60000 (10%)]	Loss: 2.331640
+Train Epoch: 1 [6400/60000 (11%)]	Loss: 2.342279
+Train Epoch: 1 [7040/60000 (12%)]	Loss: 2.320646
+Train Epoch: 1 [7680/60000 (13%)]	Loss: 2.380904
+Train Epoch: 1 [8320/60000 (14%)]	Loss: 2.305196
+Train Epoch: 1 [8960/60000 (15%)]	Loss: 2.394487
+Train Epoch: 1 [9600/60000 (16%)]	Loss: 2.340882
+Train Epoch: 1 [10240/60000 (17%)]	Loss: 2.319427
+Train Epoch: 1 [10880/60000 (18%)]	Loss: 2.332401
+Train Epoch: 1 [11520/60000 (19%)]	Loss: 2.360700
+Train Epoch: 1 [12160/60000 (20%)]	Loss: 2.303836
+Train Epoch: 1 [12800/60000 (21%)]	Loss: 2.277171
+Train Epoch: 1 [13440/60000 (22%)]	Loss: 2.339409
+Train Epoch: 1 [14080/60000 (23%)]	Loss: 2.283450
+Train Epoch: 1 [14720/60000 (25%)]	Loss: 2.319676
+Train Epoch: 1 [15360/60000 (26%)]	Loss: 2.313021
+Train Epoch: 1 [16000/60000 (27%)]	Loss: 2.321848
+Train Epoch: 1 [16640/60000 (28%)]	Loss: 2.348344
+Train Epoch: 1 [17280/60000 (29%)]	Loss: 2.369822
+Train Epoch: 1 [17920/60000 (30%)]	Loss: 2.285888
+Train Epoch: 1 [18560/60000 (31%)]	Loss: 2.375418
+Train Epoch: 1 [19200/60000 (32%)]	Loss: 2.317253
+Train Epoch: 1 [19840/60000 (33%)]	Loss: 2.306999
+Train Epoch: 1 [20480/60000 (34%)]	Loss: 2.406102
+Train Epoch: 1 [21120/60000 (35%)]	Loss: 2.388851
+Train Epoch: 1 [21760/60000 (36%)]	Loss: 2.360869
+Train Epoch: 1 [22400/60000 (37%)]	Loss: 2.338306
+Train Epoch: 1 [23040/60000 (38%)]	Loss: 2.380447
+Train Epoch: 1 [23680/60000 (39%)]	Loss: 2.347673
+Train Epoch: 1 [24320/60000 (41%)]	Loss: 2.299994
+Train Epoch: 1 [24960/60000 (42%)]	Loss: 2.299256
+Train Epoch: 1 [25600/60000 (43%)]	Loss: 2.298962
+Train Epoch: 1 [26240/60000 (44%)]	Loss: 2.312130
+Train Epoch: 1 [26880/60000 (45%)]	Loss: 2.297455
+Train Epoch: 1 [27520/60000 (46%)]	Loss: 2.376775
+Train Epoch: 1 [28160/60000 (47%)]	Loss: 2.319507
+Train Epoch: 1 [28800/60000 (48%)]	Loss: 2.295763
+Train Epoch: 1 [29440/60000 (49%)]	Loss: 2.335815
+Train Epoch: 1 [30080/60000 (50%)]	Loss: 2.386428
+Train Epoch: 1 [30720/60000 (51%)]	Loss: 2.337291
+Train Epoch: 1 [31360/60000 (52%)]	Loss: 2.318209
+Train Epoch: 1 [32000/60000 (53%)]	Loss: 2.345703
+Train Epoch: 1 [32640/60000 (54%)]	Loss: 2.323367
+Train Epoch: 1 [33280/60000 (55%)]	Loss: 2.324066
+Train Epoch: 1 [33920/60000 (57%)]	Loss: 2.332751
+Train Epoch: 1 [34560/60000 (58%)]	Loss: 2.294739
+Train Epoch: 1 [35200/60000 (59%)]	Loss: 2.370771
+Train Epoch: 1 [35840/60000 (60%)]	Loss: 2.352789
+Train Epoch: 1 [36480/60000 (61%)]	Loss: 2.306558
+Train Epoch: 1 [37120/60000 (62%)]	Loss: 2.318676
+Train Epoch: 1 [37760/60000 (63%)]	Loss: 2.304147
+Train Epoch: 1 [38400/60000 (64%)]	Loss: 2.341470
+Train Epoch: 1 [39040/60000 (65%)]	Loss: 2.317718
+Train Epoch: 1 [39680/60000 (66%)]	Loss: 2.329848
+Train Epoch: 1 [40320/60000 (67%)]	Loss: 2.358087
+Train Epoch: 1 [40960/60000 (68%)]	Loss: 2.280651
+Train Epoch: 1 [41600/60000 (69%)]	Loss: 2.342746
+Train Epoch: 1 [42240/60000 (70%)]	Loss: 2.318500
+Train Epoch: 1 [42880/60000 (71%)]	Loss: 2.359805
+Train Epoch: 1 [43520/60000 (72%)]	Loss: 2.327282
+Train Epoch: 1 [44160/60000 (74%)]	Loss: 2.355948
+Train Epoch: 1 [44800/60000 (75%)]	Loss: 2.277532
+Train Epoch: 1 [45440/60000 (76%)]	Loss: 2.362204
+Train Epoch: 1 [46080/60000 (77%)]	Loss: 2.353934
+Train Epoch: 1 [46720/60000 (78%)]	Loss: 2.374333
+Train Epoch: 1 [47360/60000 (79%)]	Loss: 2.362021
+Train Epoch: 1 [48000/60000 (80%)]	Loss: 2.352453
+Train Epoch: 1 [48640/60000 (81%)]	Loss: 2.329662
+Train Epoch: 1 [49280/60000 (82%)]	Loss: 2.343026
+Train Epoch: 1 [49920/60000 (83%)]	Loss: 2.345965
+Train Epoch: 1 [50560/60000 (84%)]	Loss: 2.438956
+Train Epoch: 1 [51200/60000 (85%)]	Loss: 2.365172
+Train Epoch: 1 [51840/60000 (86%)]	Loss: 2.411402
+Train Epoch: 1 [52480/60000 (87%)]	Loss: 2.321453
+Train Epoch: 1 [53120/60000 (88%)]	Loss: 2.325870
+Train Epoch: 1 [53760/60000 (90%)]	Loss: 2.372273
+Train Epoch: 1 [54400/60000 (91%)]	Loss: 2.340661
+Train Epoch: 1 [55040/60000 (92%)]	Loss: 2.308806
+Train Epoch: 1 [55680/60000 (93%)]	Loss: 2.310355
+Train Epoch: 1 [56320/60000 (94%)]	Loss: 2.310350
+Train Epoch: 1 [56960/60000 (95%)]	Loss: 2.312412
+Train Epoch: 1 [57600/60000 (96%)]	Loss: 2.227252
+Train Epoch: 1 [58240/60000 (97%)]	Loss: 2.308315
+Train Epoch: 1 [58880/60000 (98%)]	Loss: 2.317189
+Train Epoch: 1 [59520/60000 (99%)]	Loss: 2.317034
+Test set: Average loss: 2.3126, Accuracy: 1135/10000 (11%)