setup for bloomberg

Author: lukas

examples/scikit/cross-validation-log.py

@@ -1,12 +1,12 @@
-
-
-
 import pandas as pd
 import numpy as np
-from wandblog import log
 import wandb
+
 run = wandb.init(job_type='eval')
 config = run.config
+config.lowercase=True
+config.ngram_min=1
+config.ngram_max=1
 
 df = pd.read_csv('tweets.csv')
 target = df['is_there_an_emotion_directed_at_a_brand_or_product']
@@ -18,15 +18,14 @@
 from sklearn.feature_extraction.text import CountVectorizer
 count_vect = CountVectorizer(lowercase=config.lowercase,
                              ngram_range=(config.ngram_min,
-                                          config.ngram_max),
-                             token_pattern=config.token_pattern
+                                          config.ngram_max)
                                         )
 count_vect.fit(fixed_text)
 
 counts = count_vect.transform(fixed_text)
 
 from sklearn.naive_bayes import MultinomialNB
-nb = MultinomialNB(alpha=config.alpha)
+nb = MultinomialNB()
 
 from sklearn.model_selection import cross_val_score, cross_val_predict
 
@@ -35,4 +34,4 @@
 print(scores.mean())
 
 predictions = cross_val_predict(nb, counts, fixed_target)
-log(run, fixed_text, fixed_target, predictions)
+wandb.log({"Accuracy": scores.mean()})

examples/scikit/wandb/settings

@@ -1,4 +1,5 @@
 [default]
-entity: qualcomm
-project: scikit-dec4
-base_url: https://api.wandb.ai
+entity = bloomberg-class
+project = text-sentiment
+base_url = https://api.wandb.ai
+