misc

.gitignore

@@ -1,4 +1,5 @@
 /data/
+saved-outputs/
 
 */.ipynb_checkpoints/*
 */__pycache__/*

experiments/stocks/ForecastGenerator.py

@@ -14,19 +14,13 @@ warnings.simplefilter("ignore", NumericalWarning)
 
 def main(args):
     
-    
-    ticker_file = args.ticker_fname + ".txt"
+    data_path = "../../voltron/data/"
+    ticker_file = data_path + args.ticker_fname + ".txt"
     tckr_list = make_ticker_list(ticker_file)
-    
-    if args.end_date.lower() == "none":
-        end_date = datetime.date.today()
-    else:
-            end_date = datetime.datetime.strptime(args.end_date, "%Y-%m-%d")
-    
 
     for tckr in tckr_list:
 #         try:
-        data = GetStockHistory(tckr, history=args.ntrain + args.lookback)
+        data = GetStockHistory(tckr, end_date=args.end_date, history=args.ntrain + args.lookback)
         if args.kernel.lower() == 'volt':
             GenerateStockPredictions(tckr, data, forecast_horizon=args.forecast_horizon,
                                     train_iters=args.train_iters, 
@@ -92,7 +86,7 @@ if __name__ == "__main__":
     )
     parser.add_argument(
         "--end_date",
-        default="none",
+        default=None,
     )
     parser.add_argument(
         "--lookback",

experiments/stocks/LSTMGenerator.py

@@ -25,15 +25,10 @@ def main(args):
     ntrain = args.ntrain
     seq_len = args.seq_length
     
-    if args.end_date.lower() == "none":
-        end_date = datetime.date.today()
-    else:
-        end_date = datetime.datetime.strptime(args.end_date, "%Y-%m-%d")
-    
 
     for tckr in tckr_list:
         try:
-            data = GetStockHistory(tckr, history= ntrain + args.lookback)
+            data = GetStockHistory(tckr, end_date=args.end_date, history= ntrain + args.lookback)
             end_idxs = torch.arange(args.ntrain, data.shape[0],
                        int((data.shape[0]-args.ntrain)/args.ntimes))
 
@@ -116,7 +111,7 @@ if __name__ == "__main__":
     )
     parser.add_argument(
         "--end_date",
-        default="none",
+        default=None,
     )
     parser.add_argument(
         "--lookback",
@@ -130,4 +125,4 @@ if __name__ == "__main__":
     )
     args = parser.parse_args()
 
-    main(args)
+    main(args)

experiments/stocks/README.md

@@ -14,7 +14,11 @@ python GPGenerator.py
     --n_test_times=100 ## number of test time points
 ```
 
+```sh
+# note test_tickers0, resolves to `data/test_tickers0.txt`
+python ForecastGenerator.py \
+    --ticker_fname test_tickers0 --end_date 
+python LSTMGenerator.py \
+    --ticker_fname test_tickers0
 
-experiments/stocks/LSTMGenerator.py
-experiments/stocks/GenerateMultiMeanPreds.py
-experiments/stocks/ForecastGenerator.py
+```

experiments/stocks/example_tickers.txt

@@ -1,3 +1,8 @@
 AAPL
 MSFT
-GOOG
+GOOG
+TSLA
+XOM
+BTU
+BX
+PFE

notebooks/00.01_mjc_example.ipynb

@@ -60,7 +60,7 @@
   {
    "cell_type": "code",
    "execution_count": 14,
-   "id": "3bf4fe43",
+   "id": "2b7ed8c6",
    "metadata": {
     "ExecuteTime": {
      "end_time": "2022-07-16T07:23:41.316461Z",
@@ -446,6 +446,32 @@
     "    px_paths[vidx*npx:(vidx*npx + npx), :] = px_pred.detach().T"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 34,
+   "id": "4c9a0f25",
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2022-07-16T08:31:56.274593Z",
+     "start_time": "2022-07-16T08:31:56.252823Z"
+    }
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "torch.Size([8, 199])"
+      ]
+     },
+     "execution_count": 34,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "px_paths.shape"
+   ]
+  },
   {
    "cell_type": "markdown",
    "id": "ae84ab49-1e70-4922-a035-c607a5bbf551",
@@ -546,7 +572,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "eba9ee4f",
+   "id": "1d35cbfa",
    "metadata": {},
    "outputs": [],
    "source": []

voltron/data/MakeData.py

@@ -34,10 +34,13 @@ def DataGetter(history = 500, fpath="../data/", printing=False, end_date=None,
     tickers = make_ticker_list(fpath + ticker_file)
     make_price_files(tickers, start_date, end_date, fpath, printing)
     
-def GetStockHistory(ticker, end_date=str(datetime.date.today()), history=500):
-    end_date = datetime.datetime.strptime(end_date, "%Y-%m-%d").date()
+def GetStockHistory(ticker, end_date=None, history=500):
+    if end_date is None:
+        end_date = pd.Timestamp.now().floor('D') - pd.Timedelta('1d')
+    else:
+        end_date = pd.Timestamp(end_date)
     data = yf.download(tickers=ticker, period='10y', progress=False)
-    end_idx = np.where(data.index == pd.to_datetime(end_date))[0][0]
+    end_idx = np.where(data.index == end_date)[0][0]
     
     return data.iloc[end_idx-history:end_idx]