davidmezzetti commited on
Commit
4e23bb1
·
verified ·
1 Parent(s): 6ad7d2c

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +7 -11
app.py CHANGED
@@ -123,11 +123,7 @@ class Stats:
123
  vectors = {f'{row["yearID"]}{row["playerID"]}': self.transform(row) for _, row in self.stats.iterrows()}
124
  data = {f'{row["yearID"]}{row["playerID"]}': dict(row) for _, row in self.stats.iterrows()}
125
 
126
- embeddings = Embeddings(
127
- {
128
- "transform": Stats.transform
129
- }
130
- )
131
 
132
  embeddings.index((uid, vectors[uid], None) for uid in vectors)
133
 
@@ -254,9 +250,9 @@ class Batting(Stats):
254
 
255
  def load(self):
256
  # Retrieve raw data
257
- players = pd.read_csv("https://hf.co/datasets/neuml/baseballdatabank/resolve/main/People.csv")
258
- batting = pd.read_csv("https://hf.co/datasets/neuml/baseballdatabank/resolve/main/Batting.csv")
259
- fielding = pd.read_csv("https://hf.co/datasets/neuml/baseballdatabank/resolve/main/Fielding.csv")
260
 
261
  # Merge player data in
262
  batting = pd.merge(players, batting, how="inner", on=["playerID"])
@@ -388,8 +384,8 @@ class Pitching(Stats):
388
 
389
  def load(self):
390
  # Retrieve raw data
391
- players = pd.read_csv("https://hf.co/datasets/neuml/baseballdatabank/resolve/main/People.csv")
392
- pitching = pd.read_csv("https://hf.co/datasets/neuml/baseballdatabank/resolve/main/Pitching.csv")
393
 
394
  # Merge player data in
395
  pitching = pd.merge(players, pitching, how="inner", on=["playerID"])
@@ -439,7 +435,7 @@ class Application:
439
  st.markdown(
440
  """
441
  This application finds the best matching historical players using vector search with [txtai](https://github.com/neuml/txtai).
442
- Raw data is from the [Baseball Databank](https://github.com/chadwickbureau) Project. Read [this
443
  article](https://medium.com/neuml/explore-baseball-history-with-vector-search-5778d98d6846) for more details.
444
  """
445
  )
 
123
  vectors = {f'{row["yearID"]}{row["playerID"]}': self.transform(row) for _, row in self.stats.iterrows()}
124
  data = {f'{row["yearID"]}{row["playerID"]}': dict(row) for _, row in self.stats.iterrows()}
125
 
126
+ embeddings = Embeddings({"transform": Stats.transform})
 
 
 
 
127
 
128
  embeddings.index((uid, vectors[uid], None) for uid in vectors)
129
 
 
250
 
251
  def load(self):
252
  # Retrieve raw data
253
+ players = pd.read_csv("https://hf.co/datasets/neuml/baseballdata/resolve/main/People.csv")
254
+ batting = pd.read_csv("https://hf.co/datasets/neuml/baseballdata/resolve/main/Batting.csv")
255
+ fielding = pd.read_csv("https://hf.co/datasets/neuml/baseballdata/resolve/main/Fielding.csv")
256
 
257
  # Merge player data in
258
  batting = pd.merge(players, batting, how="inner", on=["playerID"])
 
384
 
385
  def load(self):
386
  # Retrieve raw data
387
+ players = pd.read_csv("https://hf.co/datasets/neuml/baseballdata/resolve/main/People.csv")
388
+ pitching = pd.read_csv("https://hf.co/datasets/neuml/baseballdata/resolve/main/Pitching.csv")
389
 
390
  # Merge player data in
391
  pitching = pd.merge(players, pitching, how="inner", on=["playerID"])
 
435
  st.markdown(
436
  """
437
  This application finds the best matching historical players using vector search with [txtai](https://github.com/neuml/txtai).
438
+ Raw data is from the [Lahman Baseball Database](https://sabr.org/lahman-database/). Read [this
439
  article](https://medium.com/neuml/explore-baseball-history-with-vector-search-5778d98d6846) for more details.
440
  """
441
  )