AmineTMIRI/backend
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

update

Browse Source
This commit is contained in:
MuslemRahimi 2024-06-05 08:29:44 +02:00
parent 4a8500e8ca
commit 7a2925a46a
4 changed files with 26 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
app/cron_congress_trading.py
View File

@ -88,9 +88,12 @@ def replace_representative(office):
'Vance, J.D. (Senator)': 'James Vance', 'Vance, J.D. (Senator)': 'James Vance',
'Neal Patrick MD, Facs Dunn': 'Neal Dunn', 'Neal Patrick MD, Facs Dunn': 'Neal Dunn',
'Neal Patrick MD, Facs Dunn (Senator)': 'Neal Dunn', 'Neal Patrick MD, Facs Dunn (Senator)': 'Neal Dunn',
'Neal Patrick Dunn, MD, FACS': 'Neal Dunn',
'Neal P. Dunn': 'Neal Dunn',
'Tillis, Thom (Senator)': 'Thom Tillis', 'Tillis, Thom (Senator)': 'Thom Tillis',
'W. Gregory Steube': 'Greg Steube', 'W. Gregory Steube': 'Greg Steube',
'W. Grego Steube': 'Greg Steube', 'W. Grego Steube': 'Greg Steube',
'W. Greg Steube': 'Greg Steube',
'David David Madison Cawthorn': 'David Madison Cawthorn', 'David David Madison Cawthorn': 'David Madison Cawthorn',
'Blunt, Roy (Senator)': 'Roy Blunt', 'Blunt, Roy (Senator)': 'Roy Blunt',
'Thune, John (Senator)': 'John Thune', 'Thune, John (Senator)': 'John Thune',
@ -303,7 +306,7 @@ async def run():
total_symbols = crypto_symbols +etf_symbols + stock_symbols total_symbols = crypto_symbols +etf_symbols + stock_symbols
total_raw_data = stock_raw_data + etf_raw_data + crypto_raw_data total_raw_data = stock_raw_data + etf_raw_data + crypto_raw_data
chunk_size = 250 chunk_size = 500
politician_list = [] politician_list = []
except Exception as e: except Exception as e:

19
app/ml_models/fundamental_predictor.py
View File

@ -45,8 +45,10 @@ async def download_data(ticker, con, start_date, end_date):
income = ujson.loads(query_df['income'].iloc[0]) income = ujson.loads(query_df['income'].iloc[0])
#Only consider company with at least 10 year worth of data #Only consider company with at least 10 year worth of data
'''
if len(income) < 40: if len(income) < 40:
raise ValueError("Income data length is too small.") raise ValueError("Income data length is too small.")
'''
income = [{k: v for k, v in item.items() if k not in ["symbol","reportedCurrency","calendarYear","fillingDate","acceptedDate","period","cik","link", "finalLink"]} for item in income if int(item["date"][:4]) >= 2000] income = [{k: v for k, v in item.items() if k not in ["symbol","reportedCurrency","calendarYear","fillingDate","acceptedDate","period","cik","link", "finalLink"]} for item in income if int(item["date"][:4]) >= 2000]
income_growth = ujson.loads(query_df['income_growth'].iloc[0]) income_growth = ujson.loads(query_df['income_growth'].iloc[0])
@ -109,11 +111,11 @@ async def download_data(ticker, con, start_date, end_date):
combined_data = sorted(combined_data, key=lambda x: x['date']) combined_data = sorted(combined_data, key=lambda x: x['date'])
df_income = pd.DataFrame(combined_data).dropna() df_combined = pd.DataFrame(combined_data).dropna()
df_income['Target'] = ((df_income['price'].shift(-1) - df_income['price']) / df_income['price'] > 0).astype(int) df_combined['Target'] = ((df_combined['price'].shift(-1) - df_combined['price']) / df_combined['price'] > 0).astype(int)
df_copy = df_income.copy() df_copy = df_combined.copy()
return df_copy return df_copy
@ -208,7 +210,7 @@ class FundamentalPredictor:
return {'accuracy': round(test_accuracy*100), 'precision': round(test_precision*100), 'sentiment': 'Bullish' if next_value_prediction == 1 else 'Bearish'}, test_predictions return {'accuracy': round(test_accuracy*100), 'precision': round(test_precision*100), 'sentiment': 'Bullish' if next_value_prediction == 1 else 'Bearish'}, test_predictions
def feature_selection(self, X_train, y_train,k=8): def feature_selection(self, X_train, y_train,k=8):
'''
selector = SelectKBest(score_func=f_classif, k=8) selector = SelectKBest(score_func=f_classif, k=8)
selector.fit(X_train, y_train) selector.fit(X_train, y_train)
@ -216,8 +218,9 @@ class FundamentalPredictor:
selected_features = [col for i, col in enumerate(X_train.columns) if selector.get_support()[i]] selected_features = [col for i, col in enumerate(X_train.columns) if selector.get_support()[i]]
return selected_features return selected_features
'''
# Calculate the variance of each feature with respect to the target # Calculate the variance of each feature with respect to the target
'''
variances = {} variances = {}
for col in X_train.columns: for col in X_train.columns:
grouped_variance = X_train.groupby(y_train)[col].var().mean() grouped_variance = X_train.groupby(y_train)[col].var().mean()
@ -226,6 +229,7 @@ class FundamentalPredictor:
# Sort features by variance and select top k features # Sort features by variance and select top k features
sorted_features = sorted(variances, key=variances.get, reverse=True)[:k] sorted_features = sorted(variances, key=variances.get, reverse=True)[:k]
return sorted_features return sorted_features
'''
#Train mode #Train mode
async def train_process(tickers, con): async def train_process(tickers, con):
@ -272,7 +276,7 @@ async def test_process(con):
start_date = datetime(2000, 1, 1).strftime("%Y-%m-%d") start_date = datetime(2000, 1, 1).strftime("%Y-%m-%d")
end_date = datetime.today().strftime("%Y-%m-%d") end_date = datetime.today().strftime("%Y-%m-%d")
predictor = FundamentalPredictor() predictor = FundamentalPredictor()
df = await download_data('GME', con, start_date, end_date) df = await download_data('RDDT', con, start_date, end_date)
split_size = int(len(df) * (1-test_size)) split_size = int(len(df) * (1-test_size))
test_data = df.iloc[split_size:] test_data = df.iloc[split_size:]
#selected_features = [col for col in test_data if col not in ['price','date','Target']] #selected_features = [col for col in test_data if col not in ['price','date','Target']]
@ -283,8 +287,9 @@ async def main():
con = sqlite3.connect('../stocks.db') con = sqlite3.connect('../stocks.db')
cursor = con.cursor() cursor = con.cursor()
cursor.execute("PRAGMA journal_mode = wal") cursor.execute("PRAGMA journal_mode = wal")
cursor.execute("SELECT DISTINCT symbol FROM stocks WHERE marketCap >= 500E9") cursor.execute("SELECT DISTINCT symbol FROM stocks WHERE marketCap >= 100E9")
stock_symbols = [row[0] for row in cursor.fetchall()] stock_symbols = [row[0] for row in cursor.fetchall()]
print('Number of Stocks')
print(len(stock_symbols)) print(len(stock_symbols))
await train_process(stock_symbols, con) await train_process(stock_symbols, con)
await test_process(con) await test_process(con)

4
app/restart_json.py
View File

@ -491,6 +491,7 @@ async def get_delisted_list():
def replace_representative(office): def replace_representative(office):
replacements = { replacements = {
'Carper, Thomas R. (Senator)': 'Tom Carper', 'Carper, Thomas R. (Senator)': 'Tom Carper',
@ -541,9 +542,12 @@ def replace_representative(office):
'Vance, J.D. (Senator)': 'James Vance', 'Vance, J.D. (Senator)': 'James Vance',
'Neal Patrick MD, Facs Dunn': 'Neal Dunn', 'Neal Patrick MD, Facs Dunn': 'Neal Dunn',
'Neal Patrick MD, Facs Dunn (Senator)': 'Neal Dunn', 'Neal Patrick MD, Facs Dunn (Senator)': 'Neal Dunn',
'Neal Patrick Dunn, MD, FACS': 'Neal Dunn',
'Neal P. Dunn': 'Neal Dunn',
'Tillis, Thom (Senator)': 'Thom Tillis', 'Tillis, Thom (Senator)': 'Thom Tillis',
'W. Gregory Steube': 'Greg Steube', 'W. Gregory Steube': 'Greg Steube',
'W. Grego Steube': 'Greg Steube', 'W. Grego Steube': 'Greg Steube',
'W. Greg Steube': 'Greg Steube',
'David David Madison Cawthorn': 'David Madison Cawthorn', 'David David Madison Cawthorn': 'David Madison Cawthorn',
'Blunt, Roy (Senator)': 'Roy Blunt', 'Blunt, Roy (Senator)': 'Roy Blunt',
'Thune, John (Senator)': 'John Thune', 'Thune, John (Senator)': 'John Thune',

12
fastify/get-post/server.js
View File

@ -78,8 +78,8 @@ module.exports = function (fastify, opts, done) {
filter += `&& created >= "${startDateStr}" && created <= "${endDateStr}" && pinned = false` filter += `&& created >= "${startDateStr}" && created <= "${endDateStr}" && pinned = false`
} }
posts = (await pb.collection('posts').getList(data?.startPage, 50, { posts = (await pb.collection('posts').getList(data?.startPage, 5, {
sort: sort, sort: '-created',
filter: filter, filter: filter,
expand: 'user,comments(post),alreadyVoted(post)', expand: 'user,comments(post),alreadyVoted(post)',
fields: "*,expand.user,expand.comments(post), expand.alreadyVoted(post).user,expand.alreadyVoted(post).type" fields: "*,expand.user,expand.comments(post), expand.alreadyVoted(post).user,expand.alreadyVoted(post).type"
@ -99,7 +99,7 @@ module.exports = function (fastify, opts, done) {
if (data?.userId) { if (data?.userId) {
posts = (await pb.collection('posts').getList(data?.startPage, 10, { posts = (await pb.collection('posts').getList(data?.startPage, 5, {
sort: sort, sort: sort,
filter: `user="${data?.userId}" && pinned=false`, filter: `user="${data?.userId}" && pinned=false`,
expand: `user,comments(post),alreadyVoted(post)`, expand: `user,comments(post),alreadyVoted(post)`,
@ -111,7 +111,7 @@ module.exports = function (fastify, opts, done) {
else if (data?.filterTicker) { else if (data?.filterTicker) {
posts = await pb.collection('posts').getList(data?.startPage, 10, { posts = await pb.collection('posts').getList(data?.startPage, 5, {
sort: sort, sort: sort,
filter: `tagline="${data?.filterTicker}" && pinned=false`, filter: `tagline="${data?.filterTicker}" && pinned=false`,
expand: `user,comments(post),alreadyVoted(post)`, expand: `user,comments(post),alreadyVoted(post)`,
@ -141,8 +141,8 @@ module.exports = function (fastify, opts, done) {
else { else {
filter = `pinned=false`; filter = `pinned=false`;
} }
posts = await pb.collection('posts').getList(data?.startPage, 50, { posts = await pb.collection('posts').getList(1, 5, {
sort: sort, sort: '-created',
filter: filter, filter: filter,
expand: 'user, comments(post), alreadyVoted(post)', expand: 'user, comments(post), alreadyVoted(post)',
fields: "*,expand.user,expand.comments(post), expand.alreadyVoted(post).user,expand.alreadyVoted(post).type" fields: "*,expand.user,expand.comments(post), expand.alreadyVoted(post).user,expand.alreadyVoted(post).type"