diiferent config

.idea/workspace.xml

@@ -2,7 +2,7 @@
 <project version="4">
   <component name="ChangeListManager">
     <list default="true" id="b4fb7f33-5387-4628-bcdb-b1b79dd926d0" name="Default Changelist" comment="">
-      <change beforePath="$PROJECT_DIR$/run_interaction.py" beforeDir="false" afterPath="$PROJECT_DIR$/run_interaction.py" afterDir="false" />
+      <change beforePath="$PROJECT_DIR$/.idea/workspace.xml" beforeDir="false" afterPath="$PROJECT_DIR$/.idea/workspace.xml" afterDir="false" />
     </list>
     <option name="SHOW_DIALOG" value="false" />
     <option name="HIGHLIGHT_CONFLICTS" value="true" />
@@ -22,6 +22,12 @@
     <property name="RunOnceActivity.ShowReadmeOnStart" value="true" />
     <property name="WebServerToolWindowFactoryState" value="false" />
   </component>
+  <component name="RecentsManager">
+    <key name="MoveFile.RECENT_KEYS">
+      <recent name="E:\CPI\project\CPI\log\lr-5e-6-batch-30-layer3" />
+      <recent name="E:\CPI\project\CPI" />
+    </key>
+  </component>
   <component name="SpellCheckerSettings" RuntimeDictionaries="0" Folders="0" CustomDictionaries="0" DefaultDictionary="application-level" UseSingleDictionary="true" transferred="true" />
   <component name="TaskManager">
     <task active="true" id="Default" summary="Default task">
@@ -30,7 +36,7 @@
       <option name="number" value="Default" />
       <option name="presentableId" value="Default" />
       <updated>1617112323231</updated>
-      <workItem from="1617112324809" duration="1883000" />
+      <workItem from="1617112324809" duration="4170000" />
     </task>
     <servers />
   </component>

config/config_layer_3.json

+{
+  "architectures": [
+    "BertForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 3,
+  "pad_token_id": 0,
+  "type_vocab_size": 2,
+  "vocab_size": 23615
+}

config/config_layer_6.json

+{
+  "architectures": [
+    "BertForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 6,
+  "pad_token_id": 0,
+  "type_vocab_size": 2,
+  "vocab_size": 23615
+}

config/config_layer_9.json

+{
+  "architectures": [
+    "BertForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 9,
+  "pad_token_id": 0,
+  "type_vocab_size": 2,
+  "vocab_size": 23615
+}

run_interaction.py

@@ -54,7 +54,7 @@ def main(args):
     if args.task == 'train':
         opt = torch.optim.Adam(model.parameters(), lr=args.lr)
         loss_fct = torch.nn.MSELoss()
-        writer = SummaryWriter('./log/' + args.log)
+        writer = SummaryWriter('./log/' + args.savedir)
         num_step = args.epochs * len(data_generator)
 
         # detect GPU
@@ -64,9 +64,9 @@ def main(args):
         for epoch in range(args.epochs):
             for i, (input, affinity) in enumerate(data_generator):
                 # use cuda
-                # if torch.cuda.is_available():
-                #     input.cuda()
-                #     affinity.cuda()
+#                 if torch.cuda.is_available():
+#                     input.cuda()
+#                     affinity.cuda()
                 # input model
                 pred_affinity = model(input.cuda().long())
                 loss = loss_fct(pred_affinity, affinity.cuda().unsqueeze(-1))
@@ -81,7 +81,7 @@ def main(args):
                         loss.cpu().detach().numpy()))
                 # save
                 if epoch > 1:
-                    save_path = './model/' + args.save + '/'
+                    save_path = './model/' + args.savedir + '/'
                     if not os.path.exists(save_path):
                         os.mkdir(save_path)
                     torch.save(model.state_dict(), save_path + 'epoch-{}-step-{}-loss-{}.pth'.format(epoch, i, loss))
@@ -110,8 +110,8 @@ if __name__ == '__main__':
     parser.add_argument('--lr', '--learning-rate', default=1e-4, type=float,
                         metavar='LR', help='initial learning rate', dest='lr')
     parser.add_argument('--config', default='./config/config.json', type=str, help='model config file path')
-    parser.add_argument('--log', default='training_log', type=str, help='training log')
-    parser.add_argument('--savedir', default='train', type=str, help='model save path')
+#     parser.add_argument('--log', default='training_log', type=str, help='training log')
+    parser.add_argument('--savedir', default='train', type=str, help='log and model save path')
     # parser.add_argument('--device', default='0', type=str, help='name of GPU')
     args = parser.parse_args()
 

train.sh

-CUDA_VISIBLE_DEVICES=0 python run_interaction.py --b=30 --task=train --epochs=10 --lr=5e-5 --log=lr-5e-5-batch-30 --savedir=lr-5e-5-batch-30
\ No newline at end of file
+CUDA_VISIBLE_DEVICES=3 python run_interaction.py --b=30 --task=train --epochs=10 --lr=5e-5 --savedir=lr-5e-5-batch-30-layer3 --config=./config/config_layer_3.json
\ No newline at end of file