Update models

README.md

@@ -105,11 +105,11 @@ to download this repo with submodules.
 
 If you find that this project helps your research, please consider citing the related paper:
 ```
-@article{dong2020autohas,
+@inproceedings{dong2021autohas,
   title={{AutoHAS}: Efficient Hyperparameter and Architecture Search},
   author={Dong, Xuanyi and Tan, Mingxing and Yu, Adams Wei and Peng, Daiyi and Gabrys, Bogdan and Le, Quoc V},
-  journal={arXiv preprint arXiv:2006.03656},
-  year={2020}
+  booktitle = {International Conference on Learning Representations (ICLR) Workshop on Neural Architecture Search},
+  year={2021}
 }
 @article{dong2021nats,
   title   = {{NATS-Bench}: Benchmarking NAS Algorithms for Architecture Topology and Size},

docs/README_CN.md

@@ -62,13 +62,13 @@
     <tr> <!-- (6-th row) -->
     <td align="center" valign="middle"> NATS-Bench </td>
     <td align="center" valign="middle"> <a href="https://xuanyidong.com/assets/projects/NATS-Bench"> NATS-Bench: Benchmarking NAS Algorithms for Architecture Topology and Size</a> </td>
-    <td align="center" valign="middle"> <a href="https://github.com/D-X-Y/AutoDL-Projects/tree/main/docs/NATS-Bench.md">NATS-Bench.md</a> </td>
+    <td align="center" valign="middle"> <a href="https://github.com/D-X-Y/NATS-Bench">NATS-Bench.md</a> </td>
     </tr>
     <tr> <!-- (7-th row) -->
     <td align="center" valign="middle"> ... </td>
     <td align="center" valign="middle"> ENAS / REA / REINFORCE / BOHB </td>
     <td align="center" valign="middle"> Please check the original papers. </td>
-    <td align="center" valign="middle"> <a href="https://github.com/D-X-Y/AutoDL-Projects/tree/main/docs/NAS-Bench-201.md">NAS-Bench-201.md</a> <a href="https://github.com/D-X-Y/AutoDL-Projects/tree/main/docs/NATS-Bench.md">NATS-Bench.md</a> </td>
+    <td align="center" valign="middle"> <a href="https://github.com/D-X-Y/AutoDL-Projects/tree/main/docs/NAS-Bench-201.md">NAS-Bench-201.md</a> <a href="https://github.com/D-X-Y/NATS-Bench">NATS-Bench.md</a> </td>
     </tr>
     <tr> <!-- (start second block) -->
     <td rowspan="1" align="center" valign="middle" halign="middle"> HPO </td>
@@ -98,6 +98,12 @@ Some methods use knowledge distillation (KD), which require pre-trained models.
 
 如果您发现该项目对您的科研或工程有帮助，请考虑引用下列的某些文献：
 ```
+@inproceedings{dong2021autohas,
+  title={{AutoHAS}: Efficient Hyperparameter and Architecture Search},
+  author={Dong, Xuanyi and Tan, Mingxing and Yu, Adams Wei and Peng, Daiyi and Gabrys, Bogdan and Le, Quoc V},
+  booktitle = {International Conference on Learning Representations (ICLR) Workshop on Neural Architecture Search},
+  year={2021}
+}
 @article{dong2021nats,
   title   = {{NATS-Bench}: Benchmarking NAS Algorithms for Architecture Topology and Size},
   author  = {Dong, Xuanyi and Liu, Lu and Musial, Katarzyna and Gabrys, Bogdan},

exps/trading/baselines.py

@@ -67,14 +67,26 @@ def extend_transformer_settings(alg2configs, name):
     return alg2configs
 
 
-def remove_PortAnaRecord(alg2configs):
+def refresh_record(alg2configs):
     alg2configs = copy.deepcopy(alg2configs)
     for key, config in alg2configs.items():
         xlist = config["task"]["record"]
         new_list = []
         for x in xlist:
-            if x["class"] != "PortAnaRecord":
+            # remove PortAnaRecord and SignalMseRecord
+            if x["class"] != "PortAnaRecord" and x["class"] != "SignalMseRecord":
                 new_list.append(x)
+        ## add MultiSegRecord
+        new_list.append(
+            {
+                "class": "MultiSegRecord",
+                "module_path": "qlib.contrib.workflow",
+                "generate_kwargs": {
+                    "segments": {"train": "train", "valid": "valid", "test": "test"},
+                    "save": True,
+                },
+            }
+        )
         config["task"]["record"] = new_list
     return alg2configs
 
@@ -117,7 +129,7 @@ def retrieve_configs():
             )
         )
     alg2configs = extend_transformer_settings(alg2configs, "TSF")
-    alg2configs = remove_PortAnaRecord(alg2configs)
+    alg2configs = refresh_record(alg2configs)
     print(
         "There are {:} algorithms : {:}".format(
             len(alg2configs), list(alg2configs.keys())

lib/procedures/q_exps.py

@@ -99,7 +99,12 @@ def run_exp(
 
         # Train model
         try:
-            model = R.load_object(model_obj_name)
+            if hasattr(model, "to"):  # Recoverable model
+                device = model.device
+                model = R.load_object(model_obj_name)
+                model.to(device)
+            else:
+                model = R.load_object(model_obj_name)
             logger.info("[Find existing object from {:}]".format(model_obj_name))
         except OSError:
             R.log_params(**flatten_dict(task_config))
@@ -112,16 +117,29 @@ def run_exp(
                     recorder_root_dir, "model-ckps"
                 )
             model.fit(**model_fit_kwargs)
-            R.save_objects(**{model_obj_name: model})
-        except:
-            raise ValueError("Something wrong.")
+            # remove model to CPU for saving
+            if hasattr(model, "to"):
+                model.to("cpu")
+                R.save_objects(**{model_obj_name: model})
+                model.to()
+            else:
+                R.save_objects(**{model_obj_name: model})
+        except Exception as e:
+            import pdb
+
+            pdb.set_trace()
+            raise ValueError("Something wrong: {:}".format(e))
         # Get the recorder
         recorder = R.get_recorder()
 
         # Generate records: prediction, backtest, and analysis
         for record in task_config["record"]:
             record = deepcopy(record)
-            if record["class"] == "SignalRecord":
+            if record["class"] == "MultiSegRecord":
+                record["kwargs"] = dict(model=model, dataset=dataset, recorder=recorder)
+                sr = init_instance_by_config(record)
+                sr.generate(**record["generate_kwargs"])
+            elif record["class"] == "SignalRecord":
                 srconf = {"model": model, "dataset": dataset, "recorder": recorder}
                 record["kwargs"].update(srconf)
                 sr = init_instance_by_config(record)

lib/trade_models/quant_transformer.py

@@ -112,6 +112,12 @@ class QuantTransformer(Model):
     def use_gpu(self):
         return self.device != torch.device("cpu")
 
+    def to(self, device):
+        if device is None:
+            self.model.to(self.device)
+        else:
+            self.model.to("cpu")
+
     def loss_fn(self, pred, label):
         mask = ~torch.isnan(label)
         if self.opt_config["loss"] == "mse":

scripts/trade/tsf-all.sh

@@ -0,0 +1,29 @@
+#!/bin/bash
+#
+# bash scripts/trade/tsf-all.sh 0 csi300 0
+# bash scripts/trade/tsf-all.sh 0 csi300 0.1
+# bash scripts/trade/tsf-all.sh 1 all    
+#
+set -e
+echo script name: $0
+echo $# arguments
+
+if [ "$#" -ne 3 ] ;then
+  echo "Input illegal number of parameters " $#
+  exit 1
+fi
+
+gpu=$1
+market=$2
+drop=$3
+
+channels="6 12 24 32 48 64"
+depths="1 2 3 4 5 6"
+
+for channel in ${channels}
+do
+  for depth in ${depths}
+  do
+    python exps/trading/baselines.py --alg TSF-${depth}x${channel}-d${drop} --gpu ${gpu} --market ${market}
+  done
+done