Module: LlmCostTracker::Dashboard::DataQuality

Defined in:
app/services/llm_cost_tracker/dashboard/data_quality.rb

Defined Under Namespace

Classes: QuarantinedInbox, StreamingHealthRow, Summary, UnknownPricingRow

Class Method Summary collapse

Class Method Details

.call(scope: LlmCostTracker::Call.all) ⇒ Object 



33
34
35
 
# File 'app/services/llm_cost_tracker/dashboard/data_quality.rb', line 33

def call(scope: LlmCostTracker::Call.all)
  scope.unscope(:order).select(aggregate_selects(scope)).take
end

.component_costs(scope) ⇒ Object 



155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
 
# File 'app/services/llm_cost_tracker/dashboard/data_quality.rb', line 155

def component_costs(scope)
  line_item_table = LlmCostTracker::CallLineItem.quoted_table_name
  rows = LlmCostTracker::CallLineItem
         .where(unit: "token")
         .joins(:call)
         .merge(scope.unscope(:select, :order, :group))
         .group("#{line_item_table}.kind",
                "#{line_item_table}.direction",
                "#{line_item_table}.cache_state")
         .pluck(Arel.sql("#{line_item_table}.kind"),
                Arel.sql("#{line_item_table}.direction"),
                Arel.sql("#{line_item_table}.cache_state"),
                Arel.sql("COALESCE(SUM(#{line_item_table}.cost), 0)"))
  index_costs_by_component(rows)
end

.hidden_output_summary(stats) ⇒ Object 



198
199
200
201
202
203
204
205
206
207
 
# File 'app/services/llm_cost_tracker/dashboard/data_quality.rb', line 198

def hidden_output_summary(stats)
  output_tokens = stats.output_tokens.to_i
  return unless output_tokens.positive?

  {
    hidden_output_tokens: stats.hidden_output_tokens.to_i,
    output_tokens: output_tokens,
    share_percent: stats.hidden_output_share.to_f
  }
end

.quarantined_inbox ⇒ Object 



37
38
39
40
41
42
43
44
45
46
 
# File 'app/services/llm_cost_tracker/dashboard/data_quality.rb', line 37

def quarantined_inbox
  return nil unless Ingestion.async?
  return nil unless Ingestion::InboxEntry.table_exists?

  row = Ingestion::InboxEntry
        .quarantined
        .select("COUNT(*) AS quarantined_count, COALESCE(SUM(total_cost), 0) AS quarantined_cost")
        .take
  QuarantinedInbox.new(count: row.quarantined_count.to_i, total_cost: row.quarantined_cost.to_d)
end

.service_charge_rows(scope) ⇒ Object 



103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
 
# File 'app/services/llm_cost_tracker/dashboard/data_quality.rb', line 103

def service_charge_rows(scope)
  call_table = LlmCostTracker::Call.quoted_table_name
  line_item_table = LlmCostTracker::CallLineItem.quoted_table_name
  relation = LlmCostTracker::CallLineItem
             .where.not(unit: "token")
             .joins(:call)
             .merge(scope.unscope(:select, :order))

  relation
    .group("#{call_table}.provider", "#{line_item_table}.kind", "#{line_item_table}.cost_status")
    .order(Arel.sql("COALESCE(SUM(#{line_item_table}.cost), 0) DESC"), Arel.sql("COUNT(*) DESC"))
    .select(
      "#{call_table}.provider AS provider",
      "#{line_item_table}.kind AS component",
      "#{line_item_table}.cost_status AS cost_status",
      "COUNT(*) AS charges_count",
      "COALESCE(SUM(#{line_item_table}.quantity), 0) AS quantity",
      "COALESCE(SUM(#{line_item_table}.cost), 0) AS total_cost"
    )
    .limit(10)
end

.streaming_health_rows(scope, total_streaming:) ⇒ Object 



171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
 
# File 'app/services/llm_cost_tracker/dashboard/data_quality.rb', line 171

def streaming_health_rows(scope, total_streaming:)
  return [] unless total_streaming.positive?

  unknown_predicate = unknown_usage_source_predicate(scope)
  rows = scope.unscope(:select, :order, :group)
              .where(stream: true)
              .group(:provider)
              .order(Arel.sql("COUNT(*) DESC"), :provider)
              .pluck(
                :provider,
                Arel.sql("COUNT(*)"),
                Arel.sql("SUM(CASE WHEN #{unknown_predicate} THEN 1 ELSE 0 END)")
              )

  rows.map do |provider, streams, unknown|
    streams_count = streams.to_i
    unknown_count = unknown.to_i
    StreamingHealthRow.new(
      provider: provider,
      streams: streams_count,
      with_usage: streams_count - unknown_count,
      unknown: unknown_count,
      unknown_share: percentage(unknown_count, streams_count)
    )
  end
end

.summary(stats) ⇒ Object 



48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
 
# File 'app/services/llm_cost_tracker/dashboard/data_quality.rb', line 48

def summary(stats)
  total = stats.total_calls.to_i
  unknown_pricing_count = stats.unknown_pricing_count.to_i
  untagged_calls_count = stats.untagged_calls_count.to_i
  missing_latency_count = stats.missing_latency_count.to_i
  streaming_count = stats.streaming_count.to_i
  streaming_missing_usage = stats.streaming_missing_usage_count.to_i
  missing_provider_response_id_count = stats.missing_provider_response_id_count.to_i
  calls_with_pricing = total - unknown_pricing_count
  tagged_calls = total - untagged_calls_count
  calls_with_latency = total - missing_latency_count
  streams_with_usage = streaming_count - streaming_missing_usage
  calls_with_provider_response_id = total - missing_provider_response_id_count

  Summary.new(
    total,
    unknown_pricing_count,
    untagged_calls_count,
    missing_latency_count,
    streaming_count,
    streaming_missing_usage,
    missing_provider_response_id_count,
    calls_with_pricing,
    tagged_calls,
    calls_with_latency,
    streams_with_usage,
    calls_with_provider_response_id,
    percentage(unknown_pricing_count, total),
    percentage(untagged_calls_count, total),
    percentage(missing_latency_count, total),
    percentage(streaming_count, total),
    percentage(streaming_missing_usage, streaming_count),
    percentage(calls_with_pricing, total),
    percentage(tagged_calls, total),
    percentage(calls_with_latency, total),
    percentage(streams_with_usage, streaming_count),
    percentage(calls_with_provider_response_id, total)
  )
end

.unknown_pricing_by_model(scope, total_calls:) ⇒ Object 



88
89
90
91
92
93
94
95
96
97
98
99
100
101
 
# File 'app/services/llm_cost_tracker/dashboard/data_quality.rb', line 88

def unknown_pricing_by_model(scope, total_calls:)
  scope.unknown_pricing
       .group(:provider, :model)
       .order(Arel.sql("COUNT(*) DESC"))
       .select("provider, model, COUNT(*) AS calls")
       .limit(10)
       .map do |row|
         calls = row.calls.to_i
         UnknownPricingRow.new(provider: row.provider,
                               model: row.model,
                               calls: calls,
                               share_percent: percentage(calls, total_calls))
       end
end

.usage_rows(stats, component_costs: {}) ⇒ Object 



125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
 
# File 'app/services/llm_cost_tracker/dashboard/data_quality.rb', line 125

def usage_rows(stats, component_costs: {})
  billable_tokens = stats.billable_tokens.to_f

  rows = Usage::Catalog.token_priced.map do |component|
    token_value = stats[component.token_key].to_i

    {
      price_key: component.key,
      token_key: component.token_key,
      cost_key: component.cost_key,
      token_value: token_value,
      cost_value: component_costs[component.key],
      share_percent: percentage(token_value, billable_tokens),
      share_basis: nil
    }
  end

  rows + [
    {
      price_key: nil,
      token_key: :hidden_output_tokens,
      cost_key: nil,
      token_value: stats.hidden_output_tokens.to_i,
      cost_value: nil,
      share_percent: stats.hidden_output_share.to_f,
      share_basis: :output
    }
  ]
end